TensorFlow2.1 模型训练使用

发布于:2023-09-22 ⋅ 阅读:(63) ⋅ 点赞:(0)

1、环境安装搭建

链接: Windows 安装Tensorflow2.1、Pycharm开发环境

2、神经网络

1、传统方式解决问题
在这里插入图片描述
2、机器学习解决方式
在这里插入图片描述
在这里插入图片描述

2.1、解决线性问题

下面通过两组数据推导出公式:
-1.0, 0.0, 1.0, 2.0, 3.0, 4.0
-3.0, -1.0, 1.0, 3.0, 5.0, 7.0
很明显是一个线性问题,y=2x-1,下面我们通过tensorflow来解决这个问题,输入当x=10的时候求y的值?

import tensorflow as tf
from tensorflow import keras
import numpy as np


def tensor_test1():
    # layers表示的是一层神经元,units表示这一层里面只有一个。input_shape输入值
    model = keras.Sequential([keras.layers.Dense(units=1, input_shape=[1])])
    # 指定优化和损失函数
    model.compile(optimizer='sgd', loss='mean_squared_error')
    xs = np.array([-1.0, 0.0, 1.0, 2.0, 3.0, 4.0], dtype=float)
    ys = np.array([-3.0, -1.0, 1.0, 3.0, 5.0, 7.0], dtype=float)

    # epochs 表示训练次数
    model.fit(xs, ys, epochs=500)

    # y = 2x-1
    # 通过模型去检测x=10的时候,y等于多少
    print(model.predict([10.0]))


if __name__ == '__main__':
    tensor_test1()

通过结果可以看出,是一个很接近的值

在这里插入图片描述

2.2、FAshion MNIST数据集使用

在这里插入图片描述

700000张图片
10个类别
28*28
训练神经元网络
通过tensorflow进行模型构建,通过构建出来的模型对图片进行识别

import tensorflow as tf
from tensorflow import keras
import numpy as np
import matplotlib.pyplot as plt

# 使用fashion数据集
# 自动终止
# 深度学习是不是训练的次数越多越好呢,不是次数太多会出现一些过拟合问题,就是做的题目都认识,但是新题目不会
# 所以我们需要通过callback来对他进行终止
class myCallbcak(tf.keras.callbacks.Callback):
    def on_epoch_end(self, epoch, logs={}):
        if (logs.get('loss') < 0.4):
            print("\nloss is low so cancelling training!")
            self.model.stop_training = True


def tensor_Fashion():
    callbacks = myCallbcak()
    fashion_mnist = keras.datasets.fashion_mnist
    # 训练数据集,每张图片对应的标签   测试用的图片  测试用的标签
    (train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()
    # print(train_images.shape)
    # plt.imshow(train_images[0])

    # 构造模型
    # 构造一个三层结构,第一层用来接收输入,中间层有512个神经元,这个是任意的,最后层,我们要分的类别有10

    model = keras.Sequential([
        keras.layers.Flatten(input_shape=(28, 28)),
        keras.layers.Dense(512, activation=tf.nn.relu),
        keras.layers.Dense(10, activation=tf.nn.softmax)
    ])
    model.summary()
    # 归一化,更准确
    train_images_scaled = train_images / 255.0
    # 指定优化
    model.compile(optimizer='adam', loss=tf.losses.sparse_categorical_crossentropy
                  , metrics=['accuracy'])
    model.fit(train_images_scaled, train_labels, epochs=100, callbacks=[callbacks])

    test_images_scaled = test_images / 255.0
    model.evaluate(test_images_scaled, test_labels)

    # 判断单张图片的属于哪个类别
    print(model.predict([[test_images[0] / 255]]))
    # 打印出标签
    print(np.argmax(model.predict([[test_images[0] / 255]])))
    print(test_labels[0])

3、卷积神经网络

3.1、卷积神经网络使用

通过卷积神经网络对FAshion MNIST数据集进行训练,得出的准确率比神经网络的更准确,当时也更耗时

import tensorflow as tf
from tensorflow import keras
import numpy as np
import matplotlib.pyplot as plt

def convolution_nerve():
    fashion_mnist = keras.datasets.fashion_mnist
    (train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()

    # 构造模型
    model = keras.Sequential([
        keras.layers.Conv2D(64, (3, 3), activation='relu', input_shape=(28, 28, 1)),
        keras.layers.MaxPooling2D(2, 2),
        keras.layers.Conv2D(64, (3, 3), activation='relu'),
        keras.layers.MaxPooling2D(2, 2),

        keras.layers.Flatten(),
        keras.layers.Dense(128, activation='relu'),
        keras.layers.Dense(10, activation='softmax')
    ])
    model.summary()
    # 归一化
    train_images_scaled = train_images / 255.0
    # 指定优化
    model.compile(optimizer='adam', loss=tf.losses.sparse_categorical_crossentropy
                  , metrics=['accuracy'])
    model.fit(train_images_scaled.reshape(-1, 28, 28, 1), train_labels, epochs=5)

if __name__ == '__main__':
    convolution_nerve()

模型结构

1层卷积层
输入是2828,过滤器是33,最后会去掉两个像素,所以是2626,64是过滤器,经过第一次卷积就变成64张图片了,(33+1)64=640
2池化层
尺寸减少原来的1/4,长宽各自减去一半
2层卷积层
(3
3*64+1)*64=36928

在这里插入图片描述
第一层卷积层
在这里插入图片描述
max pooling
在这里插入图片描述

3.2、ImageDataGenerator使用

1、 真实数据做处理
2、图片尺寸大小不一,需要裁成一样大小
3、数据量比较大,不能一下载装入内容
4、经常需要修改参数,列入尺寸
使用ImageDataGenerator对图片做处理

from tensorflow.keras.preprocessing.image import ImageDataGenerator

# 创建两个数据生成器,指定scaling否为0-1
train_datagen = ImageDataGenerator(rescale=1 / 255)
validation_datagen = ImageDataGenerator(rescale=1 / 255)

# 指向训练数据文件夹
train_genrator = train_datagen.flow_from_directory(
    '/',  # 训练数据所在文件夹
    target_size=(300, 300),  # 指定输出尺寸
    batch_size=32,  # 每次提取多少
    class_mode='binary'  # 指定二分类
)

validation_genrator = validation_datagen.flow_from_directory(
    '/',  # 训练数据所在文件夹
    target_size=(300, 300),  # 指定输出尺寸
    batch_size=32,  # 每次提取多少
    class_mode='binary'  # 指定二分类
)

3.3、猫狗识别案例

图片资源下载:https://download.csdn.net/download/weixin_45715405/88226536

from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.optimizers import RMSprop
import os
import tensorflow as tf
from tensorflow import keras
import numpy as np
def dogs_cats():
    base_dir = 'E:\\BaiduNetdiskDownload\\06.TensorFlow框架课件资料\\Tensorflow课件资料\\猫狗识别项目实战\\猫狗识别\\猫狗识别\data\\cats_and_dogs'
    train_dir = os.path.join(base_dir, 'train')
    validation_dir = os.path.join(base_dir, 'validation')

    # 训练集
    train_cats_dir = os.path.join(train_dir, 'cats')
    train_dogs_dir = os.path.join(train_dir, 'dogs')

    # 验证集
    validation_cats_dir = os.path.join(validation_dir, 'cats')
    validation_dogs_dir = os.path.join(validation_dir, 'dogs')

    model = tf.keras.models.Sequential([
        tf.keras.layers.Conv2D(16, (3, 3), activation='relu', input_shape=(64, 64, 3)),
        tf.keras.layers.MaxPooling2D(2, 2),
        tf.keras.layers.Conv2D(32, (3, 3), activation='relu'),
        tf.keras.layers.MaxPooling2D(2, 2),
        tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),
        tf.keras.layers.MaxPooling2D(2, 2),
        tf.keras.layers.Flatten(),
        tf.keras.layers.Dense(512, activation='relu'),
        tf.keras.layers.Dense(1, activation='sigmoid')  # 如果是多分类用softmax,2分类用sigmoid就可以了
    ])
    # 设置损失函数,优化函数
    model.compile(loss='binary_crossentropy', optimizer=RMSprop(0.001)
                  , metrics=['acc'])


    # 数据预处理
    # 都进来的数据会被自动转换成tensor(float32)格式,分别准备训练和验证
    # 图像数据归一化(0-1)区间
    train_datagen = ImageDataGenerator(
        rescale=1. / 255,
        rotation_range=40,
        width_shift_range=0.2,
        height_shift_range=0.2,
        shear_range=0.2,
        zoom_range=0.2,
        horizontal_flip=True,
        fill_mode='nearest'
    )
    test_datagen = ImageDataGenerator(rescale=1. / 255)

    train_generator = train_datagen.flow_from_directory(
        train_dir,  # 文件夹路径
        target_size=(64, 64),  # 指定resize的大小
        batch_size=20,
        # 如果one-hot就是categorical,二分类用binary就可以
        class_mode='binary'
    )
    validation_generator = test_datagen.flow_from_directory(
        validation_dir,
        target_size=(64, 64),
        batch_size=20,
        class_mode='binary'
    )

    # 训练网络模型
    # 直接fit也可以,但是通常不能把所有数据全部放入内存,fit_generator相当于一个生成器,动态产生所需的batch数据
    # steps_per_epoch相当给定一个停止条件,因为生成器会不断产生batch数据,说白了就是它不知道一个epoch里需要执行多少个step
    history = model.fit_generator(
        train_generator,
        steps_per_epoch=100,
        epochs=5,
        validation_data=validation_generator,
        validation_steps=50,
        verbose=2)

3.4、参数优化

安装

pip3 install keras-tuner
tensorFlow需要更新到2.2以上版本,不然会出现以下错误
优化之后的参数版本
在这里插入图片描述

from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.optimizers import RMSprop
import os
from kerastuner.tuners import Hyperband
from kerastuner.engine.hyperparameters import HyperParameters

# 创建两个数据生成器,指定scaling否为0-1
# train_datagen = ImageDataGenerator(rescale=1 / 255)
# validation_datagen = ImageDataGenerator(rescale=1 / 255)
#
# # 指向训练数据文件夹
# train_genrator = train_datagen.flow_from_directory(
#     'E:\\BaiduNetdiskDownload\\06.TensorFlow框架课件资料\\Tensorflow课件资料\\猫狗识别项目实战\\猫狗识别\\猫狗识别\data\\cats_and_dogs\\train',  # 训练数据所在文件夹
#     target_size=(300, 300),  # 指定输出尺寸
#     batch_size=32,  # 每次提取多少
#     class_mode='binary'  # 指定二分类
# )
#
# validation_genrator = validation_datagen.flow_from_directory(
#     'E:\\BaiduNetdiskDownload\\06.TensorFlow框架课件资料\\Tensorflow课件资料\\猫狗识别项目实战\\猫狗识别\\猫狗识别\data\\cats_and_dogs\\validation',  # 训练数据所在文件夹
#     target_size=(300, 300),  # 指定输出尺寸
#     batch_size=32,  # 每次提取多少
#     class_mode='binary'  # 指定二分类
# )

hp = HyperParameters()


def dogs_cats(hp):
    model = tf.keras.models.Sequential()

    # values 指定范围
    model.add(tf.keras.layers.Conv2D(hp.Choice('num_filters_layer0', values=[16, 64], default=16),
                                     (3, 3), activation='relu',
                                     input_shape=(64, 64, 3)))
    model.add(tf.keras.layers.MaxPooling2D(2, 2))

    for i in range(hp.Int('num_conv_layers', 1, 3)):
        model.add(tf.keras.layers.Conv2D(hp.Choice(f'num_filters_layer{i}', values=[16, 64], default=16), (3, 3),
                                         activation='relu'))
        model.add(tf.keras.layers.MaxPooling2D(2, 2))

    model.add(tf.keras.layers.Flatten())
    model.add(tf.keras.layers.Dense(hp.Int('hidde_units', 128, 512, step=32), activation='relu'))
    model.add(tf.keras.layers.Dense(1, activation='sigmoid'))  # 如果是多分类用softmax,2分类用sigmoid就可以了

    # 设置损失函数,优化函数
    model.compile(loss='binary_crossentropy', optimizer=RMSprop(0.001)
                  , metrics=['acc'])
    return model


base_dir = 'E:\\BaiduNetdiskDownload\\06.TensorFlow框架课件资料\\Tensorflow课件资料\\猫狗识别项目实战\\猫狗识别\\猫狗识别\data\\cats_and_dogs'
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')

# 训练集
train_cats_dir = os.path.join(train_dir, 'cats')
train_dogs_dir = os.path.join(train_dir, 'dogs')

# 验证集
validation_cats_dir = os.path.join(validation_dir, 'cats')
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
# 数据预处理
# 都进来的数据会被自动转换成tensor(float32)格式,分别准备训练和验证
# 图像数据归一化(0-1)区间
train_datagen = ImageDataGenerator(
    rescale=1. / 255,
    rotation_range=40,
    width_shift_range=0.2,
    height_shift_range=0.2,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True,
    fill_mode='nearest'
)
test_datagen = ImageDataGenerator(rescale=1. / 255)

train_generator = train_datagen.flow_from_directory(
    train_dir,  # 文件夹路径
    target_size=(64, 64),  # 指定resize的大小
    batch_size=20,
    # 如果one-hot就是categorical,二分类用binary就可以
    class_mode='binary'
)
validation_generator = test_datagen.flow_from_directory(
    validation_dir,
    target_size=(64, 64),
    batch_size=20,
    class_mode='binary'
)

# 训练网络模型
# 直接fit也可以,但是通常不能把所有数据全部放入内存,fit_generator相当于一个生成器,动态产生所需的batch数据
# steps_per_epoch相当给定一个停止条件,因为生成器会不断产生batch数据,说白了就是它不知道一个epoch里需要执行多少个step
# history = model.fit_generator(
#     train_generator,
#     steps_per_epoch=100,
#     epochs=5,
#     validation_data=validation_generator,
#     validation_steps=50,
#     verbose=2)

tuner = Hyperband(
    dogs_cats,
    objective='val_acc',
    max_epochs=15,
    directory='dog_cats_params',
    hyperparameters=hp,
    project_name='my_dog_cat_project'
)
tuner.search(train_generator, epochs=10, validation_data=validation_generator)

# 查看参数情况
best_hps = tuner.get_best_hyperparameters(1)[0]
print(best_hps.values)
# 通过参数将模型构建出来
model = tuner.hypermodel.build(best_hps)
model.summary()

3.5、石头剪刀布案例

wget --no-check-certificate \
    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/rps.zip \
    -O /tmp/rps.zip
  
wget --no-check-certificate \
    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/rps-test-set.zip \
    -O /tmp/rps-test-set.zip
import os
import zipfile

local_zip = '/tmp/rps.zip'
zip_ref = zipfile.ZipFile(local_zip, 'r')
zip_ref.extractall('/tmp/')
zip_ref.close()

local_zip = '/tmp/rps-test-set.zip'
zip_ref = zipfile.ZipFile(local_zip, 'r')
zip_ref.extractall('/tmp/')
zip_ref.close()

rock_dir = os.path.join('/tmp/rps/rock')
paper_dir = os.path.join('/tmp/rps/paper')
scissors_dir = os.path.join('/tmp/rps/scissors')

print('total training rock images:', len(os.listdir(rock_dir)))
print('total training paper images:', len(os.listdir(paper_dir)))
print('total training scissors images:', len(os.listdir(scissors_dir)))

rock_files = os.listdir(rock_dir)
print(rock_files[:10])

paper_files = os.listdir(paper_dir)
print(paper_files[:10])

scissors_files = os.listdir(scissors_dir)
print(scissors_files[:10])

%matplotlib inline

import matplotlib.pyplot as plt
import matplotlib.image as mpimg

pic_index = 2

next_rock = [os.path.join(rock_dir, fname) 
                for fname in rock_files[pic_index-2:pic_index]]
next_paper = [os.path.join(paper_dir, fname) 
                for fname in paper_files[pic_index-2:pic_index]]
next_scissors = [os.path.join(scissors_dir, fname) 
                for fname in scissors_files[pic_index-2:pic_index]]

for i, img_path in enumerate(next_rock+next_paper+next_scissors):
  #print(img_path)
  img = mpimg.imread(img_path)
  plt.imshow(img)
  plt.axis('Off')
  plt.show()

import tensorflow as tf
import keras_preprocessing
from keras_preprocessing import image
from keras_preprocessing.image import ImageDataGenerator

TRAINING_DIR = "/tmp/rps/"
training_datagen = ImageDataGenerator(
      rescale = 1./255,
	  rotation_range=40,
      width_shift_range=0.2,
      height_shift_range=0.2,
      shear_range=0.2,
      zoom_range=0.2,
      horizontal_flip=True,
      fill_mode='nearest')

VALIDATION_DIR = "/tmp/rps-test-set/"
validation_datagen = ImageDataGenerator(rescale = 1./255)

train_generator = training_datagen.flow_from_directory(
	TRAINING_DIR,
	target_size=(150,150),
	class_mode='categorical'
)

validation_generator = validation_datagen.flow_from_directory(
	VALIDATION_DIR,
	target_size=(150,150),
	class_mode='categorical'
)

model = tf.keras.models.Sequential([
    # Note the input shape is the desired size of the image 150x150 with 3 bytes color
    # This is the first convolution
    tf.keras.layers.Conv2D(64, (3,3), activation='relu', input_shape=(150, 150, 3)),
    tf.keras.layers.MaxPooling2D(2, 2),
    # The second convolution
    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    # The third convolution
    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    # The fourth convolution
    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    # Flatten the results to feed into a DNN
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dropout(0.5),
    # 512 neuron hidden layer
    tf.keras.layers.Dense(512, activation='relu'),
    tf.keras.layers.Dense(3, activation='softmax')
])


model.summary()

model.compile(loss = 'categorical_crossentropy', optimizer='rmsprop', metrics=['accuracy'])

history = model.fit_generator(train_generator, epochs=25, validation_data = validation_generator, verbose = 1)

model.save("rps.h5")


在这里插入图片描述

4、词条化

from tensorflow.keras.preprocessing.text import Tokenizer

from tensorflow.keras.preprocessing.text import Tokenizer
# 保持句子长度性一致
from tensorflow.keras.preprocessing.sequence import pad_sequences
import json


def hand_identi():
    sentences = [
        'I love my dog',
        'I love my cat',
        'You love my dog',
        'Do you think my dog is amazing?'
    ]
    # 100个单词的词典
    # 通过oov_token替代没有在词典里面的词
    tokenizer = Tokenizer(num_words=100, oov_token="<OOV>")
    # 按照每个数据英文单词,出现的频率次数进行排序,排序在前面100个的进行编码
    tokenizer.fit_on_texts(sentences)
    word_index = tokenizer.word_index
    # 将句子替换成对应的数字
    sequences = tokenizer.texts_to_sequences(sentences)

    # 的内容从后面开始
    # maxlen 设置内容最长多少
    # truncating = 'post' 设置超过最长长度,从后面开始截取
    padded = pad_sequences(sequences, padding='post')
    print(word_index)
    print(sequences)
    print(padded)

4.1、讽刺数据集的词条化和序列化

数据集下载地址:https://www.kaggle.com/datasets/rmisra/news-headlines-dataset-for-sarcasm-detection
在这里插入图片描述在这里插入图片描述

def json_parse():
    file_path = 'E:\\chrome_down\\archive (1)\\Sarcasm_Headlines_Dataset.json'
    sentences = []
    labels = []
    urls = []

    with open(file_path, 'r') as file:
        data = file.readlines()
    # 解析文件中的json数据
    for line in data:
        item = json.loads(line)
        sentences.append(item['headline'])
        labels.append(item['is_sarcastic'])
        urls.append(item['article_link'])

    tokenizer = Tokenizer(oov_token="<OOV>")
    tokenizer.fit_on_texts(sentences)
    word_index = tokenizer.word_index
    print(len(word_index))
    sequences = tokenizer.texts_to_sequences(sentences)
    padded = pad_sequences(sequences, padding='post')
    print(sentences[2])
    print(padded[2])
    print(padded.shape)

4.2、词嵌入

安装tensorflow数据集:pip3 install tensorflow-datasets
https://projector.tensorflow.org

def word_model():
    # 加载词条数据集
    imdb, info = tfds.load("imdb_reviews", with_info=True, as_supervised=True)
    train_data, test_data = imdb['train'], imdb['test']
    training_sentences = []
    training_labels = []
    testing_sentences = []
    testing_labels = []

    for s, l in train_data:
        training_labels.append(str(s.numpy()))
        training_labels.append(l.numpy())

    for s, l in test_data:
        testing_sentences.append(str(s.numpy()))
        testing_labels.append(l.numpy())

    training_labels_final = np.array(training_labels)
    testing_labels_final = np.array(testing_labels)

    vocab_size = 10000
    embedding_dim = 16
    max_length = 120
    trunc_type = 'post'
    oov_tok = "<OOV>"
    tokenizer = Tokenizer(num_words=vocab_size, oov_token=oov_tok)
    tokenizer.fit_on_texts(training_sentences)
    word_index = tokenizer.word_index
    sequences = tokenizer.texts_to_sequences(training_sentences)
    padded = pad_sequences(sequences, maxlen=max_length, truncating=trunc_type, padding='post')

    testing_sequences = tokenizer.texts_to_sequences(testing_sentences)
    testing_padded = pad_sequences(testing_sequences, maxlen=max_length)

    # 神经网络结构
    model = tf.keras.Sequential([
        # 嵌入层 维度为16
        tf.keras.layers.Embedding(vocab_size, embedding_dim, input_length=max_length),
        tf.keras.layers.Flatten(),
        tf.keras.layers.Dense(6, activation='relu'),
        tf.keras.layers.Dense(1, activation='sigmoid')
    ])
    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
    model.summary()

    model.fit(padded, training_labels_final, epochs=10, validation_data=(testing_padded, testing_labels_final))

    e = model.layers[0]
    weights = e.get_weights()[0]
    print(weights.shape)

    # out_v = io.open('vecs.tsv', 'w', encoding='utf-8')
    # out_m = io.open('meta.tsv', 'w', encoding='utf-8')
    # for word_num in range(1, vocab_size):
    #     word = reverse_word_index[word_num]
    #     embeddings = weights[word_num]
    #     out_m.write(word + "\n")
    #     out_v.write('\t'.join([str(x) for x in embeddings]) + "\n")
    # out_v.close()
    # out_m.close()

网站公告

今日签到

点亮在社区的每一天
去签到