appo-mujoco-swimmer

A(n) APPO model trained on the mujoco_swimmer environment.

This model was trained using Sample-Factory 2.0: https://github.com/alex-petrenko/sample-factory. Documentation for how to use Sample-Factory can be found at https://www.samplefactory.dev/

Downloading the model

After installing Sample-Factory, download the model with:

python -m sample_factory.huggingface.load_from_hub -r MattStammers/appo-mujoco-swimmer

Using the model

To run the model after download, use the enjoy script corresponding to this environment:

python -m sf_examples.mujoco.enjoy_mujoco --algo=APPO --env=mujoco_swimmer --train_dir=./train_dir --experiment=appo-mujoco-swimmer

You can also upload models to the Hugging Face Hub using the same script with the --push_to_hub flag. See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details

Training with this model

To continue training with this model, use the train script corresponding to this environment:

python -m sf_examples.mujoco.train_mujoco --algo=APPO --env=mujoco_swimmer --train_dir=./train_dir --experiment=appo-mujoco-swimmer --restart_behavior=resume --train_for_env_steps=10000000000

Note, you may have to adjust --train_for_env_steps to a suitably high number as the experiment will resume at the number of steps it concluded at.

Adjustments

This is the best one I have managed to train with sample_factory. I have been training two policies at once and invariably one of the policies is weaker than the other. By increasing the gamma to 0.9999 and using ppo I was able to get this model but it is still far off the ppo benchmark.