search

awjuliani / Meta-RL

Implementation of Meta-RL A3C algorithm
MIT License
401 stars 110 forks source link

Init AC Network #2

Closed Golly closed 7 years ago

Golly commented 7 years ago

Hi, thank you for your code.

I have a issue with init AC Network for meta bandit: hidden = tf.concat(1, [self.prev_rewards, self.prev_actions_onehot, self.timestep])

It gave me a error: TypeError: Expected int32, got list containing Tensors of type '_Message' instead.

Thank you very much for replay

awjuliani commented 7 years ago

Hi Golly,

That error is the result of the wrong argument order for tf.concat. In Tensorflow 1.0 they rearranged a number of the function arguments. It should be

tf.concat([self.prev_rewards, self.prev_actions_onehot, self.timestep],1)

I just updated the three notebooks, and they should work correctly now.