DDPG-Keras-Torcs icon indicating copy to clipboard operation
DDPG-Keras-Torcs copied to clipboard

Dimension mismatch for self.action_grads = tf.gradients(self.model.output, self.action)

Open calamont opened this issue 7 years ago • 0 comments

I feel like I must be missing something considering nobody has brought this up yet.. however I'm getting an error with trying to feed my states and a_for_grad into grads = critic.gradients(states, a_for_grad) .

Why is self.action_grads = tf.gradients(self.model.output, self.action) created with self.model.output if we are then feeding in the states, which are of different dimensions (and is what is throwing the issues for me)

calamont avatar Sep 29 '17 09:09 calamont