Yinzhicun

操作日志/readme.md

@@ -1,5 +1 @@
-每行数据共58行，分别是：
-| |单词出现频率|字符出现频率|平均词长|最长词长|总词长|分类|
- |----|----|----|----|----|----|----|
- |个数|48|6|1|1|1|1|
- |类型|实数|实数|实数|整数|整数|布尔|
+Answer:10001001110101110100111011011100100101111110000011001001

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/StduyTest/StudyOneHot.py

@@ -0,0 +1,23 @@
+import tensorflow as tf
+# import os
+# os.system('cls')
+
+NUM_CLASSES = 30 # 10分类
+
+labels = list(range(0,30,1)) # sample label
+print("labels",labels)
+
+batch_size = tf.size(labels) # get size of labels : 4
+
+labels = tf.expand_dims(labels, 1) # 增加一个维度
+
+indices = tf.expand_dims(tf.range(0, batch_size,1), 1) #生成索引
+concated = tf.concat([indices, labels] , 1) #作为拼接
+onehot_labels = tf.sparse_to_dense(concated, tf.stack([batch_size, NUM_CLASSES]), 1.0, 0.0) # 生成one-hot编码的标签
+
+with tf.Session() as ssess:
+    print("labels",labels.eval())
+    print("indices",indices.eval())
+    print("concated",concated.eval())
+    print(onehot_labels[0])
+

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/StduyTest/study_cifar10_cnn.py

@@ -0,0 +1,140 @@
+'''Train a simple deep CNN on the CIFAR10 small images dataset.
+
+It gets to 75% validation accuracy in 25 epochs, and 79% after 50 epochs.
+(it's still underfitting at that point, though).
+'''
+
+from __future__ import print_function
+import keras
+from keras.datasets import cifar10
+from keras.preprocessing.image import ImageDataGenerator
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Flatten
+from keras.layers import Conv2D, MaxPooling2D
+import os
+
+batch_size = 32     #批尺度——梯度下降法
+                    # batch_size=1 
+                    # batch_size=N ( N为总数量 ) 
+                    # batch_size=n ( n < N )
+
+num_classes = 10    #
+epochs = 100        #
+data_augmentation = True
+num_predictions = 20
+save_dir = os.path.join(os.getcwd(), 'saved_models')
+model_name = 'keras_cifar10_trained_model.h5'
+
+# The data, split between train and test sets: 
+# 数据分为训练集和测试集：
+(x_train, y_train), (x_test, y_test) = cifar10.load_data()
+print('x_train shape:', x_train.shape)
+print(x_train.shape[0], 'train samples')
+print(x_test.shape[0], 'test samples')
+
+# Convert class vectors to binary class matrices.
+# 将类向量转换为二进制类矩阵。
+y_train = keras.utils.to_categorical(y_train, num_classes)
+y_test = keras.utils.to_categorical(y_test, num_classes)
+
+model = Sequential()
+model.add(Conv2D(32, (3, 3), padding='same',
+                 input_shape=x_train.shape[1:]))
+model.add(Activation('relu'))
+model.add(Conv2D(32, (3, 3)))
+model.add(Activation('relu'))
+model.add(MaxPooling2D(pool_size=(2, 2)))
+model.add(Dropout(0.25))
+
+model.add(Conv2D(64, (3, 3), padding='same'))
+model.add(Activation('relu'))
+model.add(Conv2D(64, (3, 3)))
+model.add(Activation('relu'))
+model.add(MaxPooling2D(pool_size=(2, 2)))
+model.add(Dropout(0.25))
+
+model.add(Flatten())
+model.add(Dense(512))
+model.add(Activation('relu'))
+model.add(Dropout(0.5))
+model.add(Dense(num_classes))
+model.add(Activation('softmax'))
+
+# initiate RMSprop optimizer
+# 启动RMSprop优化器
+opt = keras.optimizers.rmsprop(lr=0.0001, decay=1e-6)
+
+# Let's train the model using RMSprop
+# 我们使用RMSprop来训练模型
+model.compile(loss='categorical_crossentropy',
+              optimizer=opt,
+              metrics=['accuracy'])
+
+x_train = x_train.astype('float32')
+x_test = x_test.astype('float32')
+x_train /= 255
+x_test /= 255
+
+if not data_augmentation:
+    print('Not using data augmentation.')
+    model.fit(x_train, y_train,
+              batch_size=batch_size,
+              epochs=epochs,
+              validation_data=(x_test, y_test),
+              shuffle=True)
+else:
+    print('Using real-time data augmentation.')
+    # This will do preprocessing and realtime data augmentation:
+    # 这将做预处理和实时数据增强：
+        # 在深度学习中，我们经常需要用到一些技巧(比如将图片进行旋转，翻转等)来进行data augmentation, 来减少过拟合。
+    datagen = ImageDataGenerator(
+        featurewise_center=False,  # set input mean to 0 over the dataset
+                                   # 在数据集上设置输入均值为0
+        samplewise_center=False,  # set each sample mean to 0
+                                  # 将每个样本均值设置为0
+        featurewise_std_normalization=False,  # divide inputs by std of the dataset
+                                              # 按照数据集的std(标准)分割输入
+        samplewise_std_normalization=False,  # divide each input by its std
+                                             #
+        zca_whitening=False,  # apply ZCA whitening
+                              #
+
+        rotation_range=0,  # randomly rotate images in the range (degrees, 0 to 180)
+                           # 整数，数据提升时图片随机转动的角度
+        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)
+                                # 浮点数，图片宽度的某个比例，数据提升时图片水平偏移的幅度`
+        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)
+                                 # 浮点数，图片高度的某个比例，数据提升时图片竖直偏移的幅度
+        horizontal_flip=True,  # randomly flip images
+                               # 布尔值，进行随机水平翻转
+        vertical_flip=False  # randomly flip images
+                             # 布尔值，进行随机竖直翻转
+    )
+
+    # Compute quantities required for feature-wise normalization
+    # 计算功能所需的标准化所需的数量
+    # (std, mean, and principal components if ZCA whitening is applied).
+    # (如果应用ZCA白化，则为标准，平均和主要成分).
+    datagen.fit(x_train)
+
+    # Fit the model on the batches generated by datagen.flow().
+    # 将模型放在由 datagen.flow() 生成的批处理上。
+    model.fit_generator(datagen.flow(x_train, y_train,
+                                     batch_size=batch_size),
+                        epochs=epochs,
+                        validation_data=(x_test, y_test),
+                        workers=4)
+
+# Save model and weights
+# 保存模型和权重
+if not os.path.isdir(save_dir):
+    os.makedirs(save_dir)
+model_path = os.path.join(save_dir, model_name)
+model.save(model_path)
+print('Saved trained model at %s ' % model_path)
+
+# Score trained model.
+# 评分训练模型
+scores = model.evaluate(x_test, y_test, verbose=1)
+print('Test loss:', scores[0])
+print('Test accuracy:', scores[1])

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/StduyTest/study_keras_cnn.py

@@ -0,0 +1,12 @@
+# -*- coding: utf-8 -*-
+
+
+
+from keras.datasets import mnist
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Flatten
+from keras.layers import Convolution2D, MaxPooling2D
+from keras.utils import np_utils
+from keras import backend as K
+
+

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/StduyTest/study_load_file.py

@@ -0,0 +1,41 @@
+# encoding: utf-8
+'''
+1、读取指定目录下的所有文件
+2、读取文件，正则匹配出需要的内容，获取文件名
+3、打开此文件(可以选择打开可以选择复制到别的地方去)
+'''
+import os.path
+import sys
+import re
+
+# 遍历指定目录，显示目录下的所有文件名
+def eachFile(filepath):
+    pathDir =  os.listdir(filepath)
+    for allDir in pathDir:
+            print(allDir)
+        # child = os.path.join('%s\%s' % (filepath, allDir))
+        # if os.path.isfile(child):
+        #     readFile(child)
+        #     # print child.decode('gbk') # .decode('gbk')是解决中文显示乱码问题
+        #     continue
+        # eachFile(child)
+
+# # 遍历出结果 返回文件的名字
+# def readFile(filenames):
+#         fopen = open(filenames, 'r', encoding='UTF-8') # r 代表read
+#         fileread = fopen.read()
+#         fopen.close()
+#         t=re.search(r'clearSpitValve',fileread)
+#         if t:
+#             # print "匹配到的文件是:"+filenames
+#             arr.append(filenames)    
+
+if __name__ == "__main__":
+    print('目前系统的编码为：',sys.getdefaultencoding()) 
+
+    filenames = "Data/logos/" # refer root dir
+    arr=[]
+    eachFile(filenames)
+    for i in arr:
+        print(i)
+

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/StduyTest/test.py

@@ -0,0 +1,79 @@
+# import json
+
+# def myJsonLoad(filePath):
+#     '''把文件打开从字符串转换成数据类型'''
+#     with open(filePath,'rb') as load_file:
+#         load_dict = json.load(load_file)
+#         return load_dict
+
+# logo_id =myJsonLoad('Data/logo_30/id_label.json') 
+# print(logo_id)
+
+
+
+
+# import re
+# allDir =  'aodi.500_0aic_QXOX3HGW_0DHP431L.jpg'
+# theTpye = re.split('\.',allDir )[0]
+# print(theTpye)
+
+
+# import tensorflow as tf
+# def myOneHot(num_classes):
+#     NUM_CLASSES = num_classes  # 分类个数
+
+#     labels = list(range(0,num_classes,1)) # sample label
+
+#     batch_size = tf.size(labels) # get size of labels : 4
+
+#     labels = tf.expand_dims(labels, 1) # 增加一个维度
+#     indices = tf.expand_dims(tf.range(0, batch_size,1), 1) #生成索引
+#     concated = tf.concat([indices, labels] , 1) #作为拼接
+#     onehot_labels = tf.sparse_to_dense(concated, tf.stack([batch_size, NUM_CLASSES]), 1.0, 0.0) # 生成one-hot编码的标签    
+#     with tf.Session() as ssess:
+#         # print(onehot_labels.eval())
+#         return onehot_labels.eval()
+
+# ontHot = myOneHot(30)
+# print(ontHot[1])
+
+
+
+# import tensorflow as tf
+# x = tf.placeholder(tf.float32,[None, 784*3])
+# data = tf.reshape(x, [-1, 28, 28, 3]) #最后一维代表通道数目，如果是rgb则为3 
+
+# print(x)
+# print(data)
+
+
+
+
+
+
+# import tensorflow as tf
+# data = tf.placeholder(tf.float32,[1, 5])
+# # with tf.Session() as sess:
+# print(data)
+
+
+
+
+import tensorflow as tf
+# python pkl 文件读写
+import pickle as pickle
+
+train_data_np = pickle.load(open('Model/train_data.plk', 'rb'))
+train_labels = pickle.load(open('Model/train_labels.plk', 'rb'))
+
+eval_data_np = pickle.load(open('Model/eval_data.plk', 'rb'))
+eval_labels = pickle.load(open('Model/eval_labels.plk', 'rb'))
+
+with tf.Session() as sess:
+    train_data = tf.convert_to_tensor(train_data_np)
+    eval_data = tf.convert_to_tensor(eval_data_np)
+
+    print(train_data.eval())
+# print(train_labels)
+    print(eval_data.eval())
+# print(eval_labels)

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/Utiliy/MyJson.py

@@ -0,0 +1,40 @@
+# -*- coding: utf-8 -*-
+
+import json
+'''
+    Json模块提供了四个功能：dumps、dump、loads、load
+        - dumps把python数据类型转换成字符串         json_str = json.dumps(test_dict)
+        - dump把数据类型转换成字符串并存储在文件中   json.dump(new_dict,file)
+        - loads把字符串转换成数据类型               new_dict = json.loads(json_str)
+        - load把文件打开从字符串转换成数据类型       load_dict = json.load(load_file)
+'''
+
+def myJsonDump(filePath,pyDict):
+    '''dump把数据类型转换成字符串并存储在文件中'''
+    with open(filePath,"w") as fileJson:
+        json.dump(pyDict,fileJson)
+        print("加载入文件完成...")
+
+def myJsonLoad(filePath):
+    '''把文件打开从字符串转换成数据类型'''
+    with open(filePath,'r') as load_file:
+        load_dict = json.load(load_file)
+        return load_dict
+'''
+写入json文件示例：
+    test_dict = {'bigberg': [7600, {1: [['iPhone', 6300], ['Bike', 800], ['shirt', 300]]}]}
+
+    with open("../config/record.json",'r') as load_f:
+        load_dict = json.load(load_f)
+        print(load_dict)
+
+
+    load_dict['smallberg'] = [8200,{1:[['Python',81],['shirt',300]]}]
+    print(load_dict)
+
+    with open("../config/record.json","w") as dump_f:
+        json.dump(load_dict,dump_f)
+
+'''
+
+

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/Utiliy/__init__.py

@@ -0,0 +1,2 @@
+# -*- coding: utf-8 -*-
+

第一次实验 尹志存 殷康宁 李嘉炜/Here is the code/Utiliy/studyJson.py

@@ -0,0 +1,45 @@
+# -*- coding: utf-8 -*-
+
+import json
+'''
+    Json模块提供了四个功能：dumps、dump、loads、load
+        - dumps把python数据类型转换成字符串         json_str = json.dumps(test_dict)
+        - dump把数据类型转换成字符串并存储在文件中  
+        - loads把字符串转换成数据类型
+        - load把文件打开从字符串转换成数据类型
+'''
+
+'''     dumps：将python中的 字典 转换为 字符串  '''
+test_dict = {'bigberg': [7600, {1: [['iPhone', 6300], ['Bike', 800], ['shirt', 300]]}]}
+print(test_dict)
+print(type(test_dict))
+#dumps 将数据转换成字符串
+json_str = json.dumps(test_dict)
+print(json_str)
+print(type(json_str))
+
+'''    loads: 将 字符串 转换为 字典    '''
+new_dict = json.loads(json_str)
+print(new_dict)
+print(type(new_dict))
+
+'''    dump: 将数据写入json文件中   '''
+with open("../config/record.json","w") as f:
+    json.dump(new_dict,f)
+    print("加载入文件完成...")
+
+'''     load:把文件打开，并把字符串变换为数据类型   '''
+# with open("../config/record.json",'r') as load_f:
+with open("record.json",'r') as load_f:
+    load_dict = json.load(load_f)
+    print(load_dict)
+
+
+load_dict['smallberg'] = [8200,{1:[['Python',81],['shirt',300]]}]
+print(load_dict)
+
+# with open("../config/record.json","w") as dump_f:
+with open("record.json","w") as dump_f:
+    json.dump(load_dict,dump_f)
+
+