在 Keras 的 MNIST 数字识别中获得测试数据的不同准确度

Question

我正在使用 Keras 进行手写数字识别，我有两个文件：predict.py 和 train.py.

train.py 训练模型（如果尚未训练）并将其保存到一个目录，否则它只会从它所在的目录加载训练好的模型被保存到并打印 Test Loss 和 Test Accuracy.

def getData():
    (X_train, y_train), (X_test, y_test) = mnist.load_data()
    y_train = to_categorical(y_train, num_classes=10)
    y_test = to_categorical(y_test, num_classes=10)
    X_train = X_train.reshape(X_train.shape[0], 784)
    X_test = X_test.reshape(X_test.shape[0], 784)
    
    # normalizing the data to help with the training
    X_train /= 255
    X_test /= 255
    
 
    return X_train, y_train, X_test, y_test

def trainModel(X_train, y_train, X_test, y_test):
    # training parameters
    batch_size = 1
    epochs = 10
    # create model and add layers
    model = Sequential()    
    model.add(Dense(64, activation='relu', input_shape=(784,)))
    model.add(Dense(10, activation = 'softmax'))

  
    # compiling the sequential model
    model.compile(loss='categorical_crossentropy', metrics=['accuracy'], optimizer='adam')
    # training the model and saving metrics in history
    history = model.fit(X_train, y_train,
          batch_size=batch_size, epochs=epochs,
          verbose=2,
          validation_data=(X_test, y_test))

    loss_and_metrics = model.evaluate(X_test, y_test, verbose=2)
    print("Test Loss", loss_and_metrics[0])
    print("Test Accuracy", loss_and_metrics[1])
    
    # Save model structure and weights
    model_json = model.to_json()
    with open('model.json', 'w') as json_file:
        json_file.write(model_json)
    model.save_weights('mnist_model.h5')
    return model

def loadModel():
    json_file = open('model.json', 'r')
    model_json = json_file.read()
    json_file.close()
    model = model_from_json(model_json)
    model.load_weights("mnist_model.h5")
    return model

X_train, y_train, X_test, y_test = getData()

if(not os.path.exists('mnist_model.h5')):
    model = trainModel(X_train, y_train, X_test, y_test)
    print('trained model')
    print(model.summary())
else:
    model = loadModel()
    print('loaded model')
    print(model.summary())
    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    loss_and_metrics = model.evaluate(X_test, y_test, verbose=2)
    print("Test Loss", loss_and_metrics[0])
    print("Test Accuracy", loss_and_metrics[1])

这是输出（假设模型之前训练过并且这次模型将被加载）：

('Test Loss', 1.741784990310669)

('Test Accuracy', 0.414)

predict.py，另一方面，预测一个手写数字：

def loadModel():
    json_file = open('model.json', 'r')
    model_json = json_file.read()
    json_file.close()
    model = model_from_json(model_json)
    model.load_weights("mnist_model.h5")
    return model

model = loadModel()

model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())

(X_train, y_train), (X_test, y_test) = mnist.load_data()
y_test = to_categorical(y_test, num_classes=10)
X_test = X_test.reshape(X_test.shape[0], 28*28)


loss_and_metrics = model.evaluate(X_test, y_test, verbose=2)

print("Test Loss", loss_and_metrics[0])
print("Test Accuracy", loss_and_metrics[1])

在这种情况下，令我惊讶的是，得到以下结果：

('Test Loss', 1.8380377866744995)

('Test Accuracy', 0.8856)

在第二个文件中，我得到的 Test Accuracy 为 0.88（是我之前得到的两倍多）。

此外，model.summery() 在两个文件中是相同的：

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 64)                50240     
_________________________________________________________________
dense_2 (Dense)              (None, 10)                650       
=================================================================
Total params: 50,890
Trainable params: 50,890
Non-trainable params: 0
_________________________________________________________________

我无法弄清楚此行为背后的原因。正常吗？还是我遗漏了什么？

Answer 1

差异是由于您一次使用标准化数据调用 evaluate() 方法（即除以 255）而另一次（即在 "predict.py" 文件中）调用它使用非标准化数据。在推理时间（即测试时间），您应该始终使用与训练数据相同的预处理步骤。

进一步，首先将数据转换为浮点数，然后除以255（否则，使用/，真正的除法是在Python 2.x和Python 3.x 当运行 X_train /= 255 和 X_test /= 255 时你会得到错误：

X_train = X_train.astype('float32')
X_test = X_test.astype('float32')

X_train /= 255.
X_test /= 255.

在 Keras 的 MNIST 数字识别中获得测试数据的不同准确度

Getting different accuracy on test data in MNIST digit recognition in Keras

python

classification

machine-learning

mnist

keras