为什么训练时准确性和损失保持不变？

Question

所以我尝试修改https://www.tensorflow.org/tutorials/keras/basic_classification中的入门教程，以使用我自己的数据。 目标是对狗和猫的图像进行分类。 代码非常简单，如下所示。 问题是网络似乎根本没有学习，训练损失和准确性在每个时代之后都保持不变。

图像（X_training）和标签（y_training）似乎具有正确的格式： X_training.shape返回： (18827, 80, 80, 3)

y_training是一个一维列表，其中的条目为{0,1}

我已经多次检查过， X_training中的“图像”被正确标记：假设X_training[i,:,:,:]代表一只狗，那么y_training[i]将返回1，如果X_training[i,:,:,:]表示一只猫，然后y_training[i]将返回0。

下面显示的是没有import语句的完整python文件。

#loading the data from 4 pickle files:
pickle_in = open("X_training.pickle","rb")
X_training = pickle.load(pickle_in)

pickle_in = open("X_testing.pickle","rb")
X_testing = pickle.load(pickle_in)

pickle_in = open("y_training.pickle","rb")
y_training = pickle.load(pickle_in)

pickle_in = open("y_testing.pickle","rb")
y_testing = pickle.load(pickle_in)


#normalizing the input data:
X_training = X_training/255.0
X_testing = X_testing/255.0


#building the model:
model = keras.Sequential([
    keras.layers.Flatten(input_shape=(80, 80,3)),
    keras.layers.Dense(128, activation=tf.nn.relu),
    keras.layers.Dense(1,activation='sigmoid')
])
model.compile(optimizer='adam',loss='mean_squared_error',metrics=['accuracy'])


#running the model:
model.fit(X_training, y_training, epochs=10)

该代码编制并训练了10个时代，但既没有损失也没有精确度提高，它们在每个时代之后保持完全相同。 该代码适用于本教程中使用的MNIST-fashion数据集，略有变化，考虑了多类与二元分类和输入形状的差异。

Answer 1

如果你想训练一个分类模型，你必须有丢失函数时的binary_crossentropy，而不是用于回归任务的mean_squared_error

更换

model.compile(optimizer='adam',loss='mean_squared_error',metrics=['accuracy'])

同

model.compile(optimizer='adam',loss='binary_crossentropy',metrics=['accuracy'])

此外，我建议不要在密集层上使用relu激活，而是使用linear

更换

keras.layers.Dense(128, activation=tf.nn.relu),

同

keras.layers.Dense(128),

为了更好地利用神经网络的力量，在你的flatten layer之前使用一些convolutional layers flatten layer

Answer 2

我找到了一个不同的实现，其中一个稍微复杂的模型可以工作。 这是没有import语句的完整代码：

#global variables:
batch_size = 32
nr_of_epochs = 64
input_shape = (80,80,3)


#loading the data from 4 pickle files:
pickle_in = open("X_training.pickle","rb")
X_training = pickle.load(pickle_in)

pickle_in = open("X_testing.pickle","rb")
X_testing = pickle.load(pickle_in)

pickle_in = open("y_training.pickle","rb")
y_training = pickle.load(pickle_in)

pickle_in = open("y_testing.pickle","rb")
y_testing = pickle.load(pickle_in)



#building the model
def define_model():
    model = Sequential()
    model.add(Conv2D(32, (3, 3), activation='relu', input_shape=input_shape))
    model.add(MaxPooling2D((2, 2)))
    model.add(Flatten())
    model.add(Dense(128, activation='relu'))
    model.add(Dense(1, activation='sigmoid'))
    # compile model
    model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
    return model
model = define_model()


#Possibility for image data augmentation
train_datagen = ImageDataGenerator(rescale=1.0/255.0)
val_datagen = ImageDataGenerator(rescale=1./255.) 
train_generator =train_datagen.flow(X_training,y_training,batch_size=batch_size)
val_generator = val_datagen.flow(X_testing,y_testing,batch_size= batch_size)



#running the model
history = model.fit_generator(train_generator,steps_per_epoch=len(X_training) //batch_size,
                              epochs=nr_of_epochs,validation_data=val_generator,
                              validation_steps=len(X_testing) //batch_size)