Hi Arthur,
Rengarajan Seshadri

Hi Renga,

Sorry about that. There was an error with the notebook on github. I’ve re-uploaded a fixed version, and it should work fine.

As for A3C, it is current set to run indefinitely. The progress is displayed in Tensorboard, which you will need to run from the command line. The model periodically saves, so once you have obtained performance you are happy with, you simply need to end the process.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.