search

devsisters / DQN-tensorflow

Tensorflow implementation of Human-Level Control through Deep Reinforcement Learning
MIT License
2.46k stars 765 forks source link

why the Breakout-v0's action_size is 4 ,but in the checkpoint is 6 #66

Open JUZI1 opened 4 years ago

JUZI1 commented 4 years ago

when i run it ,get this wrong,is it the version of gym ? tensorflow.python.framework.errors_impl.InvalidArgumentError: Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Assign requires shapes of both tensors to match. lhs shape= [4] rhs shape= [6] [[node save/Assign_9 (defined at \Users\tang\Desktop\DQN-tensorflow-master\dqn\agent.py:328) ]]

ltcs11 commented 2 years ago

same problem

the saved checkpoint is ‘m2’ in mode type but 'm2' its not in this repo