refactor: PEP8 FLAKE8 (hunkim#127)

1. Format to PEP8
2. Remove unused var/package

Chainer/chlab-02-1-linear_regression.py

@@ -3,9 +3,8 @@
 
 import numpy as np
 import chainer
-from chainer import training, Variable
-from chainer import datasets, iterators, optimizers
-from chainer import Chain
+from chainer import training
+from chainer import datasets
 from chainer.training import extensions
 
 import chainer.functions as F

Keras/klab-10-1-mnist_softmax.py

@@ -49,7 +49,7 @@
 model.summary()
 
 model.compile(loss='categorical_crossentropy',
-              optimizer='adam', learning_rate = learning_rate,
+              optimizer='adam', learning_rate=learning_rate,
               metrics=['accuracy'])
 
 model.fit(X_train, Y_train, batch_size=batch_size, epochs=training_epochs)

Keras/klab-12-1-softmax_hello_char.py

@@ -1,6 +1,7 @@
 import numpy as np
 from keras.models import Sequential
-from keras.layers import Dense, TimeDistributed, Activation, LSTM
+from keras.layers import Activation
+from keras.layers import Dense
 from keras.utils import np_utils
 
 import os

Keras/klab-12-3-rnn_stock_prediction.py

@@ -1,7 +1,8 @@
 # http://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras/
 import numpy as np
 from keras.models import Sequential
-from keras.layers import Dense, LSTM, Activation
+from keras.layers import Activation
+from keras.layers import LSTM
 from sklearn.preprocessing import MinMaxScaler
 import os
 

PyTorch/lab-10-3-mnist_nn_xavier.py

@@ -4,7 +4,7 @@
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
 import random
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 

PyTorch/lab-10-4-mnist_nn_deep.py

@@ -4,7 +4,7 @@
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
 import random
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 

PyTorch/lab-10-5-mnist_nn_dropout.py

@@ -4,7 +4,7 @@
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
 import random
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 

PyTorch/lab-11-1-mnist_cnn.py

@@ -3,7 +3,7 @@
 from torch.autograd import Variable
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 
@@ -32,6 +32,7 @@
 
 
 class CNN(torch.nn.Module):
+
     def __init__(self):
         super(CNN, self).__init__()
         # L1 ImgIn shape=(?, 28, 28, 1)

PyTorch/lab-11-2-mnist_deep_cnn.py

@@ -3,7 +3,7 @@
 from torch.autograd import Variable
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 
@@ -33,6 +33,7 @@
 
 
 class CNN(torch.nn.Module):
+
     def __init__(self):
         super(CNN, self).__init__()
         # L1 ImgIn shape=(?, 28, 28, 1)

PyTorch/lab-11-3-mnist_cnn_class.py

@@ -3,7 +3,7 @@
 from torch.autograd import Variable
 import torchvision.datasets as dsets
 import torchvision.transforms as transforms
-from torch.nn import init
+import torch.nn.init
 
 torch.manual_seed(777)  # reproducibility
 
@@ -32,6 +32,7 @@
 
 
 class CNN(torch.nn.Module):
+
     def __init__(self):
         super(CNN, self).__init__()
         self._build_net()

PyTorch/lab-12-1-hello-rnn.py

@@ -2,7 +2,6 @@
 import torch
 import torch.nn as nn
 from torch.autograd import Variable
-import numpy as np
 
 torch.manual_seed(777)  # reproducibility
 
@@ -39,6 +38,7 @@
 
 
 class RNN(nn.Module):
+
     def __init__(self, num_classes, input_size, hidden_size, num_layers):
         super(RNN, self).__init__()
         self.num_classes = num_classes

PyTorch/lab-12-2-char-seq-rnn.py

@@ -2,7 +2,6 @@
 import torch
 import torch.nn as nn
 from torch.autograd import Variable
-import numpy as np
 
 torch.manual_seed(777)  # reproducibility
 
@@ -46,6 +45,7 @@ def one_hot(x, num_classes):
 
 
 class LSTM(nn.Module):
+
     def __init__(self, num_classes, input_size, hidden_size, num_layers):
         super(LSTM, self).__init__()
         self.num_classes = num_classes

PyTorch/lab-12-4-rnn-long_char.py

@@ -2,7 +2,6 @@
 import torch
 import torch.nn as nn
 from torch.autograd import Variable
-import numpy as np
 
 torch.manual_seed(777)  # reproducibility
 
@@ -59,6 +58,7 @@ def one_hot(x, num_classes):
 
 
 class LSTM(nn.Module):
+
     def __init__(self, num_classes, input_size, hidden_size, num_layers):
         super(LSTM, self).__init__()
         self.num_classes = num_classes

PyTorch/lab-12-5-stock_prediction.py

@@ -83,6 +83,7 @@ def MinMaxScaler(data):
 
 
 class LSTM(nn.Module):
+
     def __init__(self, num_classes, input_size, hidden_size, num_layers):
         super(LSTM, self).__init__()
         self.num_classes = num_classes

lab-07-1-learning_rate_and_evaluation.py

@@ -1,6 +1,5 @@
 # Lab 7 Learning rate and Evaluation
 import tensorflow as tf
-import numpy as np
 tf.set_random_seed(777)  # for reproducibility
 
 x_data = [[1, 2, 1], [1, 3, 2], [1, 3, 4], [1, 5, 5],

lab-09-4-xor_tensorboard.py

@@ -49,7 +49,7 @@
     # tensorboard --logdir=./logs/xor_logs
     merged_summary = tf.summary.merge_all()
     writer = tf.summary.FileWriter("./logs/xor_logs_r0_01")
-    writer.add_graph(sess.graph) # Show the graph
+    writer.add_graph(sess.graph)  # Show the graph
 
     # Initialize TensorFlow variables
     sess.run(tf.global_variables_initializer())

lab-10-7-mnist_nn_higher_level_API.py

@@ -77,7 +77,7 @@
         c = sess.run(cost, feed_dict=feed_dict_cost)
         avg_cost += c / total_batch
 
-    print("[Epoch: {:>4}] cost = {:>.9}".format(epoch+1, avg_cost))
+    print("[Epoch: {:>4}] cost = {:>.9}".format(epoch + 1, avg_cost))
     #print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
 
 print('Learning Finished!')

lab-11-3-mnist_cnn_class.py

@@ -1,6 +1,5 @@
 # Lab 11 MNIST and Deep learning CNN
 import tensorflow as tf
-import random
 # import matplotlib.pyplot as plt
 
 from tensorflow.examples.tutorials.mnist import input_data

lab-11-4-mnist_cnn_layers.py

@@ -1,6 +1,5 @@
 # Lab 11 MNIST and Deep learning CNN
 import tensorflow as tf
-import random
 # import matplotlib.pyplot as plt
 
 from tensorflow.examples.tutorials.mnist import input_data

lab-13-1-mnist_using_scope.py

@@ -1,8 +1,6 @@
 # Lab 7 Learning rate and Evaluation
 import tensorflow as tf
-import numpy as np
 import random
-import matplotlib.pyplot as plt
 
 from tensorflow.examples.tutorials.mnist import input_data
 

lab-13-2-mnist_tensorboard.py

@@ -1,8 +1,6 @@
 # Lab 7 Learning rate and Evaluation
 import tensorflow as tf
-import numpy as np
 import random
-import matplotlib.pyplot as plt
 
 from tensorflow.examples.tutorials.mnist import input_data
 

mxnet/mxlab-04-3-file_input_linear_regression.py

@@ -32,8 +32,8 @@
                     data_names=['data'],
                     label_names=['target'],
                     context=mx.gpu(0))
-net.bind(data_shapes = [mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
-         label_shapes= [mx.io.DataDesc(name='target', shape=(batch_size, 1), layout='NC')])
+net.bind(data_shapes=[mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
+         label_shapes=[mx.io.DataDesc(name='target', shape=(batch_size, 1), layout='NC')])
 net.init_params(initializer=mx.init.Normal(sigma=0.01))
 net.init_optimizer(optimizer='sgd', optimizer_params={'learning_rate': 1E-4, 'momentum': 0.9})
 
@@ -58,4 +58,4 @@
 '''
 input = [60, 70, 110], score = [[ 182.48858643]]
 input = [90, 100, 80], score = [[ 175.24279785]]
-'''
+'''

mxnet/mxlab-05-2-logistic_regression_diabetes.py

@@ -1,6 +1,5 @@
 # Lab 5 Logistic Regression Classifier
 import mxnet as mx
-import mxnet.ndarray as nd
 import numpy as np
 import logging
 import sys
@@ -30,10 +29,10 @@
                     data_names=['data'],
                     label_names=['target'],
                     context=mx.gpu(0))
-net.bind(data_shapes = [mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
-         label_shapes= [mx.io.DataDesc(name='target', shape=(batch_size, 1), layout='NC')])
+net.bind(data_shapes=[mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
+         label_shapes=[mx.io.DataDesc(name='target', shape=(batch_size, 1), layout='NC')])
 net.init_params(initializer=mx.init.Normal(sigma=0.01))
-net.init_optimizer(optimizer='sgd', optimizer_params={'learning_rate': 1E-3, 'momentum':0.9})
+net.init_optimizer(optimizer='sgd', optimizer_params={'learning_rate': 1E-3, 'momentum': 0.9})
 
 # 4. Train the model
 # First constructing the training iterator and then fit the model
@@ -84,4 +83,4 @@
 [ 1.]
 [ 1.]
 [ 1.]]
-'''
+'''

mxnet/mxlab-06-2-softmax_zoo_classifier.py

@@ -1,6 +1,5 @@
 # Lab 6 Softmax Classifier
 import mxnet as mx
-import mxnet.ndarray as nd
 import numpy as np
 import logging
 import sys
@@ -32,10 +31,10 @@
                     data_names=['data'],
                     label_names=['target'],
                     context=mx.gpu(0))
-net.bind(data_shapes = [mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
-         label_shapes= [mx.io.DataDesc(name='target', shape=(batch_size,), layout='NC')])
+net.bind(data_shapes=[mx.io.DataDesc(name='data', shape=(batch_size, dimension), layout='NC')],
+         label_shapes=[mx.io.DataDesc(name='target', shape=(batch_size,), layout='NC')])
 net.init_params(initializer=mx.init.Normal(sigma=0.01))
-net.init_optimizer(optimizer='sgd', optimizer_params={'learning_rate': 1E-1, 'momentum':0.9})
+net.init_optimizer(optimizer='sgd', optimizer_params={'learning_rate': 1E-1, 'momentum': 0.9})
 
 # 4. Train the model
 # First constructing the training iterator and then fit the model
@@ -91,4 +90,4 @@
   0.  0.  1.  1.  1.  1.  3.  3.  2.  0.  0.  0.  0.  0.  0.  0.  0.  1.
   6.  3.  0.  0.  2.  6.  1.  1.  2.  6.  3.  1.  0.  6.  3.  1.  5.  4.
   2.  2.  3.  0.  0.  1.  0.  5.  0.  6.  1.]
-'''
+'''

mxnet/mxlab-11-2-mnist_deep_cnn.py

@@ -16,8 +16,8 @@
 X, y = mnist.data, mnist.target
 X = X.astype(np.float32) / 255.0
 X_train, X_valid, X_test = X[:55000].reshape((-1, 1, 28, 28)),\
-                           X[55000:60000].reshape((-1, 1, 28, 28)),\
-                           X[60000:].reshape((-1, 1, 28, 28))
+    X[55000:60000].reshape((-1, 1, 28, 28)),\
+    X[60000:].reshape((-1, 1, 28, 28))
 y_train, y_valid, y_test = y[:55000], y[55000:60000], y[60000:]
 
 # hyper parameters
@@ -92,7 +92,7 @@
         # Slice the data batch and label batch.
         # Note that we use np.take to ensure that the batch will be padded correctly.
         data_npy = np.take(X_train,
-                           indices=np.arange(i * batch_size, (i+1) * batch_size),
+                           indices=np.arange(i * batch_size, (i + 1) * batch_size),
                            axis=0,
                            mode="clip")
         label_npy = np.take(y_train,
@@ -115,7 +115,7 @@
 total_num = 0
 for i in range(total_batch):
     num_valid = batch_size if (i + 1) * batch_size <= X_test.shape[0]\
-                           else X_test.shape[0] - i * batch_size
+        else X_test.shape[0] - i * batch_size
     data_npy = np.take(X_test,
                        indices=np.arange(i * batch_size, (i + 1) * batch_size),
                        axis=0,
@@ -137,7 +137,7 @@
 test_net.reshape(data_shapes=[mx.io.DataDesc(name='data', shape=(1, 1, 28, 28), layout='NCHW')],
                  label_shapes=None)
 r = np.random.randint(0, X_test.shape[0])
-test_net.forward(data_batch=mx.io.DataBatch(data=[nd.array(X_test[r:r+1])],
+test_net.forward(data_batch=mx.io.DataBatch(data=[nd.array(X_test[r:r + 1])],
                                             label=None))
 logits_nd = test_net.get_outputs()[0]
 print("Label: ", int(y_test[r]))

mxnet/mxlab-11-5-mnist_cnn_ensemble_layers.py

@@ -16,8 +16,8 @@
 X, y = mnist.data, mnist.target
 X = X.astype(np.float32) / 255.0
 X_train, X_valid, X_test = X[:55000].reshape((-1, 1, 28, 28)),\
-                           X[55000:60000].reshape((-1, 1, 28, 28)),\
-                           X[60000:].reshape((-1, 1, 28, 28))
+    X[55000:60000].reshape((-1, 1, 28, 28)),\
+    X[60000:].reshape((-1, 1, 28, 28))
 y_train, y_valid, y_test = y[:55000], y[55000:60000], y[60000:]
 
 # hyper parameters
@@ -26,6 +26,7 @@
 batch_size = 100
 num_models = 2
 
+
 def build_symbol():
     data = mx.sym.var(name="data")
     label = mx.sym.var(name="label")
@@ -134,7 +135,7 @@ def get_batch(p, batch_size, X, y):
 total_num = 0
 for i in range(total_batch):
     num_valid = batch_size if (i + 1) * batch_size <= X_test.shape[0]\
-                           else X_test.shape[0] - i * batch_size
+        else X_test.shape[0] - i * batch_size
     data_npy, label_npy, num_valid = get_batch(i, batch_size, X_test, y_test)
     prob_ensemble = nd.zeros(shape=(label_npy.shape[0], 10), ctx=mx.gpu())
     for i, test_net in enumerate(test_nets):

mxnet/mxlab-12-4-rnn_deep_prediction.py

@@ -9,7 +9,6 @@
 # brew install graphviz
 # pip3 install graphviz
 # pip3 install pydot-ng
-from mxnet.visualization import plot_network
 import matplotlib.pyplot as plt
 
 logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)  # Config the logging
@@ -125,8 +124,8 @@ def train_eval_net(use_cudnn):
 print("Done!")
 
 
-print("CUDNN time spent: %g, test mse: %g" %(cudnn_time_spent, cudnn_mse))
-print("NoCUDNN time spent: %g, test mse: %g" %(normal_time_spent, normal_mse))
+print("CUDNN time spent: %g, test mse: %g" % (cudnn_time_spent, cudnn_mse))
+print("NoCUDNN time spent: %g, test mse: %g" % (normal_time_spent, normal_mse))
 
 plt.close('all')
 fig = plt.figure()
@@ -138,4 +137,4 @@ def train_eval_net(use_cudnn):
 '''
 CUDNN time spent: 10.0955, test mse: 0.00721571
 NoCUDNN time spent: 38.9882, test mse: 0.00565724
-'''
+'''