我试图在keras模型中使用huber loss(编写DQN),但是我的结果很糟糕,我认为我做错了.我的代码如下.
model = Sequential()
model.add(Dense(output_dim=64, activation='relu', input_dim=state_dim))
model.add(Dense(output_dim=number_of_actions, activation='linear'))
loss = tf.losses.huber_loss(delta=1.0)
model.compile(loss=loss, opt='sgd')
return model