autoencoder_custom_binary_dense_lib_.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "400e8acc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-22 10:51:19.398301: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "# import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "# import tensorflow\n",
    "from tensorflow.keras.layers import BatchNormalization, Dense, Input, BatchNormalization\n",
    "from tensorflow.keras import layers, Model\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "from tensorflow.keras.layers import GaussianNoise\n",
    "\n",
    "\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, BatchNormalization\n",
    "from tensorflow.keras.layers import GaussianNoise\n",
    "\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.layers import Dense, BatchNormalization, GaussianNoise, InputLayer\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e95bf246",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.layers import Dense, Convolution2D, Activation, Flatten, MaxPooling2D,Input,Dropout,GlobalAveragePooling2D\n",
    "from tensorflow.keras.layers import BatchNormalization\n",
    "from tensorflow.python.framework import ops\n",
    "from binarization_utils import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a3d48aad",
   "metadata": {},
   "outputs": [],
   "source": [
    "resid_levels = True\n",
    "BINARY = False\n",
    "LOGIC_SHRINKAGE = False\n",
    "custom_rand_seed = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "544f2644",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "# os.environ['TF_XLA_FLAGS'] = '--tf_xla_enable_xla_devices'\n",
    "\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"-1\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "415acd68",
   "metadata": {},
   "outputs": [],
   "source": [
    "def autoenco(k,n,SNRs):\n",
    "    BER = []\n",
    "    R = float(k/n)\n",
    "    Nt = 10000\n",
    "    M = 2**k\n",
    "\n",
    "\n",
    "    label = np.random.randint(M,size=Nt)\n",
    "    data = [] # one-hot encodedno\n",
    "\n",
    "\n",
    "    for i in label:\n",
    "        temp = np.zeros(M)\n",
    "        temp[i] = 1\n",
    "        data.append(temp) \n",
    "\n",
    "\n",
    "    data = np.array(data)\n",
    "\n",
    "\n",
    "#     autoencoder = Sequential()\n",
    "#     autoencoder.add(Dense(M, activation='relu', input_shape=(M,), name='input_layer'))\n",
    "#     autoencoder.add(Dense(n, activation='linear', name='encoder_layer_1'))\n",
    "#     autoencoder.add(BatchNormalization(name='encoder_batchnorm_layer'))\n",
    "    \n",
    "    \n",
    "    autoencoder = Sequential()\n",
    "    autoencoder.add(binary_dense(levels=resid_levels, n_in=M, n_out=M, input_shape=(M,), BINARY=BINARY, LOGIC_SHRINKAGE=LOGIC_SHRINKAGE, custom_rand_seed=custom_rand_seed, name=\"input_layer\"))\n",
    "    autoencoder.add(Activation('relu'))\n",
    "    \n",
    "    autoencoder.add(binary_dense(levels=resid_levels, n_in=M, n_out=n, BINARY=BINARY, LOGIC_SHRINKAGE=LOGIC_SHRINKAGE, custom_rand_seed=custom_rand_seed, name=\"encoder_layer_1\"))\n",
    "    autoencoder.add(Activation('linear'))\n",
    "    \n",
    "    autoencoder.add(BatchNormalization(name='encoder_batchnorm_layer'))\n",
    "\n",
    "    \n",
    "    \n",
    "    \n",
    "    EbNo_train = 10 ** (0.7)\n",
    "    noise_value = np.sqrt(1/(2*R*EbNo_train))\n",
    "    \n",
    "    \n",
    "    autoencoder.add(GaussianNoise(noise_value))\n",
    "    \n",
    "    \n",
    "#     autoencoder.add(Dense(M, activation='relu', name='decoder_layer_1'))\n",
    "#     autoencoder.add(Dense(M, activation='softmax', name='decoder_layer_2'))\n",
    "#     #sgd = SGD(lr=0.001)\n",
    "#     autoencoder.compile(optimizer='adam', loss='categorical_crossentropy')\n",
    "    \n",
    "    \n",
    "    \n",
    "    autoencoder.add(binary_dense(levels=resid_levels, n_in=n, n_out=M, BINARY=BINARY, LOGIC_SHRINKAGE=LOGIC_SHRINKAGE, custom_rand_seed=custom_rand_seed, name=\"decoder_layer_1\"))\n",
    "    autoencoder.add(Activation('relu'))\n",
    "    \n",
    "    autoencoder.add(binary_dense(levels=resid_levels, n_in=M, n_out=M, BINARY=BINARY, LOGIC_SHRINKAGE=LOGIC_SHRINKAGE, custom_rand_seed=custom_rand_seed, name=\"decoder_layer_2\"))\n",
    "    autoencoder.add(Activation('softmax'))\n",
    "    autoencoder.compile(optimizer='adam', loss='categorical_crossentropy')\n",
    "    \n",
    "    \n",
    "    print(autoencoder.summary())\n",
    "    \n",
    "    \n",
    "    \n",
    "    N_val = 1500\n",
    "    val_label = np.random.randint(M,size=N_val)\n",
    "    val_data = []\n",
    "    \n",
    "    for i in val_label:\n",
    "        temp = np.zeros(M)\n",
    "        temp[i] = 1\n",
    "        val_data.append(temp)\n",
    "    \n",
    "    val_data = np.array(val_data)\n",
    "    \n",
    "    \n",
    "\n",
    "\n",
    "    autoencoder.fit(data, data,epochs=30,\n",
    "                  batch_size=300,\n",
    "                  validation_data=(val_data, val_data),\n",
    "                  callbacks=[ModelCheckpoint(save_best_only=True, \n",
    "                                             filepath=\"ok\")])\n",
    " \n",
    "    \n",
    "    \n",
    "\n",
    "\n",
    "\n",
    "#     encoder = Sequential()\n",
    "#     encoder.add(BatchNormalization(input_shape=(M,), name='encoder_2_input_layer_with_batchnorm'))\n",
    "#     encoded_input = Input(shape=(n,))\n",
    "    \n",
    "\n",
    "\n",
    "    return BER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "28900bb2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /workspace/Logic-Shrinkage/MNIST-SVHN-CIFAR10/training-software/binarization_utils.py:917: The name tf.keras.initializers.RandomNormal is deprecated. Please use tf.compat.v1.keras.initializers.RandomNormal instead.\n",
      "\n",
      "WARNING:tensorflow:TensorFlow will not use sklearn by default. This improves performance in some cases. To enable sklearn export the environment variable  TF_ALLOW_IOLIBS=1.\n",
      "WARNING:tensorflow:TensorFlow will not use Dask by default. This improves performance in some cases. To enable Dask export the environment variable  TF_ALLOW_IOLIBS=1.\n",
      "WARNING:tensorflow:TensorFlow will not use Pandas by default. This improves performance in some cases. To enable Pandas export the environment variable  TF_ALLOW_IOLIBS=1.\n",
      "WARNING:tensorflow:Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc611f0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc611f0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING: Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc611f0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc611f0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING:tensorflow:Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc61c40>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc61c40>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING: Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc61c40>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f454bc61c40>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING:tensorflow:Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480db9a0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480db9a0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING: Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480db9a0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480db9a0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING:tensorflow:Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480adbb0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480adbb0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "WARNING: Entity <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480adbb0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Failed to parse source code of <bound method binary_dense.call of <binarization_utils.binary_dense object at 0x7f45480adbb0>>, which Python reported as:\n",
      "\tdef call(self, x,mask=None):\n",
      "\n",
      "\t\t#np.random.seed(self.custom_rand_seed)\n",
      "\t\t#tf.set_random_seed(self.rand_seed)\n",
      "\t\t#random.seed(self.rand_seed)\n",
      "\n",
      "\t\tconstraint_gamma=K.abs(self.gamma)#K.clip(self.gamma,0.01,10)\n",
      "\t\tif self.levels==1 or self.first_layer==True:\n",
      "\t\t\tif self.BINARY==False:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\telse:\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\tself.out=K.dot(x,self.clamped_w)\n",
      "\t\telif self.levels==2:\n",
      "\t\t\tif self.LUT==True:\n",
      "\n",
      "\t\t\t\tif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\t#SHRINKAGE\n",
      "                #Stacking c coef\tficients to create c_mat1 and c_mat2 tensors\n",
      "\t\t\t\t\tc_mat1=self.c_param[0 : 2**self.k_lut, :, :]\n",
      "\t\t\t\t\tc_mat2=self.c_param[2**self.k_lut : (2**self.k_lut)*2, :, :]\n",
      "\n",
      "\t\t\t\t\tc_mat1=tf.reshape(c_mat1,(1,2**self.k_lut,self.n_in*self.n_out))                \n",
      "\t\t\t\t\tc_mat2=tf.reshape(c_mat2,(1,2**self.k_lut,self.n_in*self.n_out))\n",
      "       \n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "            \n",
      "\t\t\t\t\tc_mat1=tf.matmul(c_mat1,self.shrinkage_map)\n",
      "\t\t\t\t\tc_mat2=tf.matmul(c_mat2,self.shrinkage_map)\n",
      "\t\t\t\t\t\n",
      "\t\t\t\t\tc_mat1=tf.transpose(c_mat1,(2,0,1))\n",
      "\t\t\t\t\tc_mat2=tf.transpose(c_mat2,(2,0,1))\n",
      "\n",
      "\t\t\t\t\tc_mat1 = tf.reshape(c_mat1, (2**self.k_lut, self.n_in, self.n_out))\n",
      "\t\t\t\t\tc_mat2 = tf.reshape(c_mat2, (2**self.k_lut, self.n_in, self.n_out))\n",
      "                \n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*K.clip(self.c_param, -1,1)\n",
      "\n",
      "\t\t\t\telif self.LOGIC_SHRINKAGE==True:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(tf.concat([c_mat1, c_mat2], axis=0))\n",
      "\n",
      "\t\t\t\telse:\n",
      "\n",
      "\t\t\t\t\tself.clamped_c_param= constraint_gamma*binarize(self.c_param)\n",
      "\t\t\t\t\n",
      "\t\t\t\t# Special hack for randomising the subsequent input connections: tensorflow does not support advanced matrix indexing\n",
      "\t\t\t\tshuf_x=tf.transpose(x, perm=[2, 0, 1])\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\tshuf_x_0 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_0, tf.int32))\n",
      "\t\t\t\t\tshuf_x_0=tf.transpose(shuf_x_0, perm=[1, 2, 0], name=\"extra_input_1_tapped\")\n",
      "\t\t\t\t\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\tshuf_x_1 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_1, tf.int32))\n",
      "\t\t\t\t\tshuf_x_1=tf.transpose(shuf_x_1, perm=[1, 2, 0], name=\"extra_input_2_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\tshuf_x_2 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_2, tf.int32))\n",
      "\t\t\t\t\tshuf_x_2=tf.transpose(shuf_x_2, perm=[1, 2, 0], name=\"extra_input_3_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\tshuf_x_3 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_3, tf.int32))\n",
      "\t\t\t\t\tshuf_x_3=tf.transpose(shuf_x_3, perm=[1, 2, 0], name=\"extra_input_4_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\tshuf_x_4 = tf.gather_nd(shuf_x, tf.cast(self.rand_map_4, tf.int32))\n",
      "\t\t\t\t\tshuf_x_4=tf.transpose(shuf_x_4, perm=[1, 2, 0], name=\"extra_input_5_tapped\")\n",
      "\t\t\t\n",
      "\t\t\t\tx_pos=(1+binarize(x))/2*abs(x)\n",
      "\t\t\t\tx_neg=(1-binarize(x))/2*abs(x)\n",
      "\t\t\t\tif self.k_lut > 1:\n",
      "\t\t\t\t\txs0_pos=(1+binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\t\txs0_neg=(1-binarize(shuf_x_0))/2#*abs(shuf_x_0)\n",
      "\t\t\t\tif self.k_lut > 2:\n",
      "\t\t\t\t\txs1_pos=(1+binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\t\txs1_neg=(1-binarize(shuf_x_1))/2#*abs(shuf_x_1)\n",
      "\t\t\t\tif self.k_lut > 3:\n",
      "\t\t\t\t\txs2_pos=(1+binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs2_neg=(1-binarize(shuf_x_2))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 4:\n",
      "\t\t\t\t\txs3_pos=(1+binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs3_neg=(1-binarize(shuf_x_3))/2#*abs(shuf_x_2)\n",
      "\t\t\t\tif self.k_lut > 5:\n",
      "\t\t\t\t\txs4_pos=(1+binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\t\t\t\t\txs4_neg=(1-binarize(shuf_x_4))/2#*abs(shuf_x_2)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 1:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 2:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 3:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 4:\n",
      "\t\t\t\t\tself.out=         tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask,name=\"inter_dot_output_0_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask,name=\"inter_dot_output_1_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask,name=\"inter_dot_output_2_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask,name=\"inter_dot_output_3_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask,name=\"inter_dot_output_4_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask,name=\"inter_dot_output_5_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask,name=\"inter_dot_output_6_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask,name=\"inter_dot_output_7_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask,name=\"inter_dot_output_8_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask,name=\"inter_dot_output_9_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask,name=\"inter_dot_output_10_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask,name=\"inter_dot_output_11_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask,name=\"inter_dot_output_12_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask,name=\"inter_dot_output_13_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask,name=\"inter_dot_output_14_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask,name=\"inter_dot_output_15_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[16]*self.pruning_mask,name=\"inter_dot_output_16_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[17]*self.pruning_mask,name=\"inter_dot_output_17_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[18]*self.pruning_mask,name=\"inter_dot_output_18_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[19]*self.pruning_mask,name=\"inter_dot_output_19_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[20]*self.pruning_mask,name=\"inter_dot_output_20_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[21]*self.pruning_mask,name=\"inter_dot_output_21_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[22]*self.pruning_mask,name=\"inter_dot_output_22_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[23]*self.pruning_mask,name=\"inter_dot_output_23_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[24]*self.pruning_mask,name=\"inter_dot_output_24_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[25]*self.pruning_mask,name=\"inter_dot_output_25_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[26]*self.pruning_mask,name=\"inter_dot_output_26_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[27]*self.pruning_mask,name=\"inter_dot_output_27_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[28]*self.pruning_mask,name=\"inter_dot_output_28_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[29]*self.pruning_mask,name=\"inter_dot_output_29_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:],self.clamped_c_param[30]*self.pruning_mask,name=\"inter_dot_output_30_tapped\")\n",
      "\t\t\t\t\tself.out=self.out+tf.matmul(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:],self.clamped_c_param[31]*self.pruning_mask,name=\"inter_dot_output_31_tapped\")\n",
      "\n",
      "\t\t\t\tif self.k_lut == 5:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\n",
      "\t\t\t\tif self.k_lut == 6:\n",
      "\t\t\t\t\tself.out=         K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[0]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[1]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[2]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[3]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[4]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[5]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[6]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[7]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[8]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[9]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[10]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[11]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[12]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[13]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[14]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[15]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[16]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[17]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[18]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[19]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[20]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[21]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[22]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[23]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[24]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[25]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[26]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[27]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[28]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[29]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[30]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[31]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[32]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[33]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[34]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[35]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[36]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[37]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[38]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[39]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[40]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[41]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[42]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[43]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[44]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[45]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[46]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_pos[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[47]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[48]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[49]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[50]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[51]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[52]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[53]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[54]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_pos[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[55]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[56]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[57]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[58]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_pos[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[59]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[60]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_pos[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[61]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_pos[0,:,:],self.clamped_c_param[62]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*xs0_neg[0,:,:]*xs1_neg[0,:,:]*xs2_neg[0,:,:]*xs3_neg[0,:,:]*xs4_neg[0,:,:],self.clamped_c_param[63]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[64]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[65]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[66]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[67]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[68]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[69]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[70]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[71]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[72]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[73]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[74]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[75]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[76]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[77]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[78]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[79]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[80]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[81]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[82]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[83]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[84]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[85]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[86]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[87]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[88]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[89]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[90]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[91]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[92]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[93]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[94]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_pos[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[95]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[96]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[97]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[98]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[99]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[100]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[101]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[102]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[103]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[104]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[105]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[106]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[107]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[108]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[109]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[110]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_pos[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[111]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[112]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[113]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[114]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[115]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[116]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[117]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[118]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_pos[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[119]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[120]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[121]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[122]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_pos[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[123]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[124]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_pos[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[125]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_pos[1,:,:],self.clamped_c_param[126]*self.pruning_mask)\n",
      "\t\t\t\t\tself.out=self.out+K.dot(x_neg[1,:,:]*xs0_neg[1,:,:]*xs1_neg[1,:,:]*xs2_neg[1,:,:]*xs3_neg[1,:,:]*xs4_neg[1,:,:],self.clamped_c_param[127]*self.pruning_mask)\n",
      "\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tif self.BINARY==False:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*K.clip(self.w,-1,1)\n",
      "\t\t\t\telse:\n",
      "\t\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w*self.pruning_mask)\n",
      "\t\telif self.levels==3:\n",
      "\t\t\tif self.LUT==True:\n",
      "\t\t\t\tself.clamped_w1=constraint_gamma*binarize(self.w1)\n",
      "\t\t\t\tself.clamped_w2=constraint_gamma*binarize(self.w2)\n",
      "\t\t\t\tself.clamped_w3=constraint_gamma*binarize(self.w3)\n",
      "\t\t\t\tself.clamped_w4=constraint_gamma*binarize(self.w4)\n",
      "\t\t\t\tself.clamped_w5=constraint_gamma*binarize(self.w5)\n",
      "\t\t\t\tself.clamped_w6=constraint_gamma*binarize(self.w6)\n",
      "\t\t\t\tself.clamped_w7=constraint_gamma*binarize(self.w7)\n",
      "\t\t\t\tself.clamped_w8=constraint_gamma*binarize(self.w8)\n",
      "\t\t\t\tx_pos=(1+x)/2\n",
      "\t\t\t\tx_neg=(1-x)/2\n",
      "\t\t\t\tself.out=K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w1)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w2)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w3)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_pos[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w4)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_pos[2,:,:],self.clamped_w5)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_pos[1,:,:]*x_neg[2,:,:],self.clamped_w6)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_pos[2,:,:],self.clamped_w7)\n",
      "\t\t\t\tself.out=self.out+K.dot(x_neg[0,:,:]*x_neg[1,:,:]*x_neg[2,:,:],self.clamped_w8)\n",
      "\t\t\telse:\n",
      "\t\t\t\tx_expanded=0\n",
      "\t\t\t\tself.clamped_w=constraint_gamma*binarize(self.w)\n",
      "\t\t\t\tfor l in range(self.levels):\n",
      "\t\t\t\t\tx_expanded=x_expanded+x[l,:,:]\n",
      "\t\t\t\tself.out=K.dot(x_expanded,self.clamped_w)\n",
      "\t\treturn self.out\n",
      "\n",
      "This may be caused by multiline strings or comments not indented at the same level as the code.\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_layer (binary_dense)   (None, 8)                 16537     \n",
      "_________________________________________________________________\n",
      "activation (Activation)      (None, 8)                 0         \n",
      "_________________________________________________________________\n",
      "encoder_layer_1 (binary_dens (None, 7)                 14473     \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 7)                 0         \n",
      "_________________________________________________________________\n",
      "encoder_batchnorm_layer (Bat (None, 7)                 28        \n",
      "_________________________________________________________________\n",
      "gaussian_noise (GaussianNois (None, 7)                 0         \n",
      "_________________________________________________________________\n",
      "decoder_layer_1 (binary_dens (None, 8)                 14470     \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 8)                 0         \n",
      "_________________________________________________________________\n",
      "decoder_layer_2 (binary_dens (None, 8)                 16537     \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 8)                 0         \n",
      "=================================================================\n",
      "Total params: 62,045\n",
      "Trainable params: 258\n",
      "Non-trainable params: 61,787\n",
      "_________________________________________________________________\n",
      "None\n",
      "Train on 10000 samples, validate on 1500 samples\n",
      "Epoch 1/30\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-22 10:51:22.662209: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1\n",
      "2023-10-22 10:51:23.147519: E tensorflow/stream_executor/cuda/cuda_driver.cc:282] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
      "2023-10-22 10:51:23.147600: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: 5310f4639566\n",
      "2023-10-22 10:51:23.147608: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: 5310f4639566\n",
      "2023-10-22 10:51:23.147731: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 470.82.1\n",
      "2023-10-22 10:51:23.147764: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 470.82.1\n",
      "2023-10-22 10:51:23.147771: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 470.82.1\n",
      "2023-10-22 10:51:23.162276: I tensorflow/core/platform/profile_utils/cpu_utils.cc:109] CPU Frequency: 2700000000 Hz\n",
      "2023-10-22 10:51:23.166439: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x139f1f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
      "2023-10-22 10:51:23.166479: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 1s 115us/sample - loss: 1.9585 - val_loss: 2.0243\n",
      "Epoch 2/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 1.6894 - val_loss: 1.9379\n",
      "Epoch 3/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 1.4293 - val_loss: 1.8007\n",
      "Epoch 4/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 1.1458 - val_loss: 1.5976\n",
      "Epoch 5/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.8774 - val_loss: 1.3309\n",
      "Epoch 6/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.6662 - val_loss: 1.0352\n",
      "Epoch 7/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.5010 - val_loss: 0.7546\n",
      "Epoch 8/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.3583 - val_loss: 0.5014\n",
      "Epoch 9/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.2186 - val_loss: 0.2916\n",
      "Epoch 10/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.1271 - val_loss: 0.1538\n",
      "Epoch 11/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0813 - val_loss: 0.0824\n",
      "Epoch 12/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0556 - val_loss: 0.0463\n",
      "Epoch 13/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0404 - val_loss: 0.0277\n",
      "Epoch 14/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0313 - val_loss: 0.0179\n",
      "Epoch 15/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0247 - val_loss: 0.0122\n",
      "Epoch 16/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0204 - val_loss: 0.0090\n",
      "Epoch 17/30\n",
      "10000/10000 [==============================] - 0s 9us/sample - loss: 0.0169 - val_loss: 0.0068\n",
      "Epoch 18/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0145 - val_loss: 0.0053\n",
      "Epoch 19/30\n",
      "10000/10000 [==============================] - 0s 10us/sample - loss: 0.0127 - val_loss: 0.0043\n",
      "Epoch 20/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0102 - val_loss: 0.0035\n",
      "Epoch 21/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0090 - val_loss: 0.0030\n",
      "Epoch 22/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0085 - val_loss: 0.0026\n",
      "Epoch 23/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0074 - val_loss: 0.0022\n",
      "Epoch 24/30\n",
      "10000/10000 [==============================] - 0s 7us/sample - loss: 0.0066 - val_loss: 0.0019\n",
      "Epoch 25/30\n",
      "10000/10000 [==============================] - 0s 7us/sample - loss: 0.0059 - val_loss: 0.0017\n",
      "Epoch 26/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0055 - val_loss: 0.0015\n",
      "Epoch 27/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0051 - val_loss: 0.0013\n",
      "Epoch 28/30\n",
      "10000/10000 [==============================] - 0s 7us/sample - loss: 0.0047 - val_loss: 0.0012\n",
      "Epoch 29/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0045 - val_loss: 0.0011\n",
      "Epoch 30/30\n",
      "10000/10000 [==============================] - 0s 8us/sample - loss: 0.0041 - val_loss: 9.8249e-04\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use k =3 and n = 7\n",
    "SNRs = np.arange(1.0,18.0,0.5)\n",
    "\n",
    "autoenco(3, 7, SNRs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2a46cc31",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}