Keras入门--Mnist手写数字识别(源自Keras官方样例)

1. Keras官方Mnist例子

https://keras.io/examples/vision/mnist_convnet/

2. 改造模型

2.1 加载keras相关模块

import os
import tempfile
import numpy as np
# from tensorflow import keras
# from tensorflow.keras import layers
# 以上注释掉的，是基于tensorflow 2.x版本的高级接口，因我的tensorflow版本为1.x的，所以改用如下接口
from tensorflow.contrib.keras.python.keras.datasets.mnist import load_data
from tensorflow.contrib.keras.python.keras.utils import to_categorical
from tensorflow.contrib.keras.python import keras

2.2 准备训练和测试Mnist数据

keras有load_data接口下载(https://s3.amazonaws.com/img-datasets/mnist.npz)数据。但是本地网络接口下载数据太慢，半天没有反应。所以先把mnist.npz下载下来，然后用numpy解析。

# Model / data parameters
num_classes = 10
input_shape = (28, 28, 1)

# the data, split between train and test sets
# (x_train, y_train), (x_test, y_test) = load_data() #此接口下载太慢，提前下载用网页，然后用如下方式处理数据。
with np.load('../datasets/Keras/mnist.npz') as data:
    x_train = data['x_train']
    y_train = data['y_train']
    x_test = data['x_test']
    y_test = data['y_test']

# Scale images to the [0, 1] range
x_train = x_train.astype("float32") / 255
x_test = x_test.astype("float32") / 255
# Make sure images have shape (28, 28, 1)
x_train = np.expand_dims(x_train, -1)
x_test = np.expand_dims(x_test, -1)
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
y_train = to_categorical(y_train, num_classes)
y_test = to_categorical(y_test, num_classes)

2.3 构建模型

官方源码使用keras.models.Input(shape=input_shape)方式将输入数据的shape传入模型，但是编译老报错，说：必须是layer类的。于是将输入数据的shape放到二维卷积Conv2D类属性中。

model = keras.models.Sequential(
    [
        # keras.models.Input(shape=input_shape), # 源码使用的
        keras.layers.Conv2D(32, kernel_size=(3, 3), activation="relu", input_shape=input_shape), # 进行了改装，添加input_shape参数值
        keras.layers.MaxPooling2D(pool_size=(2, 2)),
        keras.layers.Conv2D(64, kernel_size=(3, 3), activation="relu"),
        keras.layers.MaxPooling2D(pool_size=(2, 2)),
        keras.layers.Flatten(),
        keras.layers.Dropout(0.5),
        keras.layers.Dense(num_classes, activation="softmax"),
    ]
)
model.summary()

• keras.models.Sequential
  Linear stack of layers. list of layers to add to the model.
  Note: The first layer passed to a Sequential model should have a defined input shape. 这个例子，我把input shape放到了Conv2D中。
• keras.layers.Flatten
  将输入数据拉直，当然不影响batch size。即(batch_size, s1, s2, …)拉直成(batch_size, s1s2…)
• keras.layers.Dropout
  和tensorflow一样，避免过拟合。概率性位置变为0，其他位置除以传入的系数。
• keras.layers.Dense
  全连接。output = activation(dot(input, kernel) + bias)，activation为传入的激活函数

2.4 训练模型并保存模型

保存模型是自己加的，为了拿实际图片进行测试，以验证识别的准确性。更接近生活中的实际应用。

# train the model
batch_size = 128
epochs = 15

# loss function and Gradient descent， get the optimal parameters value
model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])

# the same as scikit-learn, predict interface input the data, predict the class
out = model.predict(x_train)

# start iteration to train
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)

# Evaluate the trained model
score = model.evaluate(x_test, y_test, verbose=0)

# save model to .h5 file
fname = "E:/work/src/Keras/Minist/model/model-new.h5"
keras.models.save_model(model, fname)

print("Test loss:", score[0])
print("Test accuracy:", score[1])

3. 测试

import os
import cv2
import numpy as np
from tensorflow.contrib.keras.python import keras

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

def img_prepare(file_name):
    """handle the png image"""
    image_raw = cv2.imread(file_name)
    rgb_to_gray = cv2.cvtColor(image_raw, cv2.COLOR_RGB2GRAY)
    data_type_convert = rgb_to_gray.astype(np.float32)
    data_normalized = (data_type_convert) * 1.0/255
    data = np.reshape(data_type_convert, [1,28,28,1])
    return data

if __name__ == '__main__':
    # load model
    model_file = "../src/Keras/Minist/model/model-new.h5"
    new_model = keras.models.load_model(model_file)

    # prepare image
    file_name = "../src/Keras/Minist/00007.png"
    test_x = img_prepare(file_name)
    out2 = new_model.predict(test_x)
    print(out2)

打印是形如：
[0 0 0 0 0 0 0 1 0 0]第八个类别