Sigmaban Playground

Installation

Install uv

curl -LsSf https://astral.sh/uv/install.sh | sh

Training

If you want to use the imitation reward, you can generate reference motion with this repo (in the branch `sigmaban)

Then copy polynomial_coefficients.pkl in playground/sigmaban2024/data/

You'll also have to set USE_IMITATION_REWARD=True in it's joystick.py file

Run a training:

uv run playground/sigmaban2024/runner.py --task flat_terrain_backlash --num_timesteps 300000000

Tensorboard

uv run tensorboard --logdir=<yourlogdir>

Inference

Infer mujoco

uv run playground/sigmaban_2024/mujoco_infer.py -o <path_to_.onnx> --model_path playground/sigmaban2024/xmls/scene_flat_terrain_backlash.xml

Mapper

Run the sampler:

uv run playground/sigmaban2024/footsteps_sampler.py -o <path_to_onnx>

This will generate footsteps.json

Run the mapper:

uv run playground/sigmaban2024/footsteps_mapper.py --plot --model_path ONNX_with_metadata.onnx