How can I get the output file of the policy network and use it?

[蔡文天]

I run the experiment "pr2_tensorflow_example", after the training process ended, I didn't find the output file of tensorflow in the folder, so how can I get the training result of tensorflow?

If I get the result, how can I use it to lead the agent to perform the optimal action?

Thanks!