Edward Beeching

Hosting model checkpoints on the Hugging Face hub

6

Hello, thank you for sharing these checkpoints with the community. Would you be interested in hosting these checkpoints on the [Hugging Face Model Hub?](https://huggingface.co/models) In addition, we could also collaborate...

Hello, I only see the environment in this repository. Do you intend to provide the code for training and testing the agent? Thanks.

Initial implementation of an automatic model builder. Adds a default mode, which will generate encoder(s) automatically based on the shape(s) of the input observation(s). Adds a decoder option after the...

Thanks for you work on splerger, I just moved a project to Godot 4 and needed to update the addon. Find the changes in this PR. The changes work on...

[sf2] envpool integration

1

Adds envpool integration for Atari and Mujoco. The integration works as follows, when batched_sampling=True envpool is used, when false it uses the normal method.

Grad accumulation and memory bugfix

2

- Adds command line arg passing to sentiment and toxicity examples - Adds gradient accumulation as a command line arg #218 - Fixes tensors not being detached from graph when...

Add gradient accumulation

1

With larger sequences and batches, we quickly go out of memory when the batch size is greater than 1.

add accelerate example

1

## Description ## Types of changes - [ ] Bug fix - [x] New feature - [ ] New algorithm - [ ] Documentation ## Checklist: - [ ] I've...

[Enhancement] Add logging

1

Add logging to file instead of terminal (requested on discord)