search

young-geng / CQL

Conservative Q Learning on top of SAC
MIT License
118 stars 24 forks source link

Make checkpoints public #7

Closed ezhang7423 closed 1 year ago

ezhang7423 commented 1 year ago

Hi, would it be possible to release the checkpoints for this implementation? Would be very grateful for this.

young-geng commented 1 year ago

Unfortunately I don't have saved checkpoints for this. However, it takes less than a day to train one from scratch on a single GPU. If that is still too long, you might want to check out my JAX implementation of CQL, which can train a CQL agent on a single GPU in around 2 hours.

ezhang7423 commented 1 year ago

Thank you for the response!