深度学习中取代.load_data()使用自制数据集的方法

在学习mnist数据集时，我们使用.load_data方法实现打开，但如何实现打开训练自己制作的数据集呢？

def hanshu (path,txt):#定义读取本地数据集函数，其作用是代替之前的load data功能
    file = open(txt,'r')#文件以只读形式打开
    neirong = file.readlines()#读出所有行
    file.close()#文件关闭


    x,y_ = [],[]
    for content in neirong:#逐行读出
        values = content.split()#将每行的内容以空格分开形成两列 图片路径为value【0】 标签路径为value【1】
        image_path =path + values[0]#读取路径等于图片路径＋图片特征
        img = Image.open(image_path)#打开图片
        img = np.array(img.convert('L'))#转化为0-255灰度值的np.array格式
        img= img/255#归一化
        x.append(img)#将值填入x
        y_.append(values[1])#将值填入y_
        print('loading : '+ content)#打印信息

完整代码如下：

import tensorflow as tf
from PIL import Image
import  numpy as np
import  os
trainpath='./mnist_image_label/mnist_train_jpg_60000/'#训练图片路径
traintxt='./mnist_image_label/mnist_train_jpg_60000.txt'#签路径
x_train_savepath = './mnist_image_label/mnist_x_train.npy'#训练结果保存路径
y_train_savepath = './mnist_image_label/mnist_y_train.npy'#训练结果保存路径（标签）


testpath = './mnist_image_label/mnist_test_jpg_10000/'#测试图片路径
testtxt = '/mnist_image_label/mnist_test_jpg_10000.txt'#测试标签路径
x_test_savepath = './mnist_image_label/mnist_x_test.npy'#测试结果保存路径
y_test_savepath = './mnist_image_label/mnist_y_test.npy'#测试结果保存路径（标签）




def hanshu (path,txt):#定义读取本地数据集函数，其作用是代替之前的load data功能
    file = open(txt,'r')#文件以只读形式打开
    neirong = file.readlines()#读出所有行
    file.close()#文件关闭


    x,y_ = [],[]
    for content in neirong:#逐行读出
        values = content.split()#将每行的内容以空格分开形成两列 图片路径为value【0】 标签路径为value【1】
        image_path =path + values[0]#读取路径等于图片路径＋图片特征
        img = Image.open(image_path)#打开图片
        img = np.array(img.convert('L'))#转化为0-255灰度值的np.array格式
        img= img/255#归一化
        x.append(img)#将值填入x
        y_.append(values[1])#将值填入y_
        print('loading : '+ content)#打印信息



    x = np.array(x)
    y_=np.array(y_)
    y_=y_.astype(np.int64)#转换数据类型为64位整形
    return x,y_#返回输入特征x，标签y_


if os.path.exists(x_train_savepath) and os.path.exists(y_train_savepath) and os.path.exists(
        x_test_savepath) and os.path.exists(y_test_savepath):
    print('------load datasets------')
    x_train_save = np.load(x_train_savepath)
    y_train = np.load(y_train_savepath)
    x_test_save = np.load(x_test_savepath)
    y_test = np.load(y_test_savepath)
    x_train = np.reshape(x_train_save,(len(x_train_save),28,28))
    x_test = np.reshape(x_test_save, (len(x_test_save), 28, 28))
else:
    print('------making datasets------')
    x_train , y_train = hanshu(trainpath,traintxt)
    x_test , y_test = hanshu(testpath,testtxt)


    print('------saving datasets------')
    x_train_save = np.reshape(x_train,(len(x_train),-1))
    x_test_save = np.reshape(x_test,(len(x_test),-1))
    np.save(x_train_savepath,x_train_save)
    np.save(y_train_savepath,y_train)
    np.save(x_test_savepath,x_test_save)
    np.save(y_test_savepath,y_test)

model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dense(10, activation='softmax')
])



model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
              metrics=['sparse_categorical_accuracy'])

model.fit(x_train, y_train, batch_size=32, epochs=5, validation_data=(x_test, y_test), validation_freq=1)
model.summary()