Question

我目前正在使用三种类型的车辆（货车/ SUV，轿车和卡车）来训练图像分类模型。我有1800张训练图像和210张验证图像。当我尝试插入数据时。我用keras.preprocessing.image.ImageDataGenerator()和Val_Data.flow(预处理数据。由于我的精度保持不变，因此似乎绝对正在发生。以下是我的代码和结果。我已经尝试解决了很长时间，似乎无法解决此问题。

代码：

    # Creating Training Data Shuffled and Organized
Train_Data = keras.preprocessing.image.ImageDataGenerator()

Train_Gen = Train_Data.flow(
        Train_Img, 
        Train_Labels,
        batch_size=BATCH_SIZE,
        shuffle=True)

# Creating Validation Data Shuffled and Organized
Val_Data = keras.preprocessing.image.ImageDataGenerator()

Val_Gen = Val_Data.flow(
        Train_Img, 
        Train_Labels,
        batch_size=BATCH_SIZE,
        shuffle=True)


print(Train_Gen)


###################################################################################
###################################################################################


#Outline the Model
hidden_layer_size = 300
output_size = 3

#Model Core
model = tf.keras.Sequential([

                             tf.keras.layers.Flatten(input_shape=(IMG_HEIGHT,IMG_WIDTH,CHANNELS)),
                             tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),
                             tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),
                             tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),
                             tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),
                             tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),
                             tf.keras.layers.Dense(output_size, activation = 'softmax')

                            ])

custom_optimizer = tf.keras.optimizers.SGD(learning_rate=0.001)

#Compile Model
model.compile(optimizer='adam', loss ='sparse_categorical_crossentropy', metrics = ['accuracy'])

#Train Model
NUM_EPOCHS = 15;
model.fit(Train_Gen, validation_steps = 10, epochs = NUM_EPOCHS, validation_data = Val_Gen, verbose = 2)

结果：

    180/180 - 27s - loss: 10.7153 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 2/15
180/180 - 23s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 3/15
180/180 - 23s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 4/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 5/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 6/15
180/180 - 21s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 7/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 8/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 9/15
180/180 - 23s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 10/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 11/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 12/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 13/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 14/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300
Epoch 15/15
180/180 - 22s - loss: 10.7454 - accuracy: 0.3333 - val_loss: 10.7991 - val_accuracy: 0.3300

Answer 1

我想您首先要研究的是卷积神经网络，因为我可以看到您正在尝试使用密集网络来解决基于图像的问题。它可以工作，但不如CNN好。

有很多原因导致模型卡在张量流中，这是我最常遇到的原因：

有一次我因为输入不是numpy数组而陷入模型阻塞
优化器学习率太高。

首先要使自定义优化器的学习率低于默认值，这意味着：

custom_optimizer = tf.keras.optimizers.Adam(lr=0.0001)
model.compile(optimizer=custom_optimizer, loss="sparse_categorical_crossentropy", metrics = ['acc'])

检查此链接，它是您附近的CNN实施 https://gist.github.com/RyanAkilos/3808c17f79e77c4117de35aa68447045

麻烦训练Tensor流神经网络，如何解决此问题？

1 个答案: