search

nicrusso7 / rex-gym

OpenAI Gym environments for an open-source quadruped robot (SpotMicro)
Apache License 2.0
973 stars 130 forks source link

Training from command line example #9

Closed jackwilkinson255 closed 4 years ago

jackwilkinson255 commented 4 years ago

Can you give an example of training a policy directly from the command line?

I've tried this but it doesn't seem to produce anything in the logging directory:

rex-gym train --playground True --env walk --log-dir ~/rex-gym/logging

Am I missing some arguments?

nicrusso7 commented 4 years ago

The agent will save a checkpoint every 500000 steps in the default configuration.

By the way, the playground flag is used to spin up a rendered single agent training session. You may want to run a batch session removing it (this will speed up the training).

nicrusso7 commented 4 years ago

Please feel free to reopen this issue if you incur in any other problem.