feat(envs): add RoboCasa365 benchmark integration

Add RoboCasa365 (arXiv:2603.04356) as a new simulation benchmark with
365 everyday kitchen manipulation tasks across 2,500 diverse environments.

New files:
- src/lerobot/envs/robocasa.py: gym.Env wrapper with deferred env creation,
  flat 12D action / 16D state vectors, 3-camera support
- docs/source/robocasa.mdx: user-facing documentation
- docker/Dockerfile.benchmark.robocasa: CI benchmark image

Modified files:
- src/lerobot/envs/configs.py: RoboCasaEnv config (--env.type=robocasa)
- pyproject.toml: robocasa optional dependency group
- docs/source/_toctree.yml: sidebar entry
- .github/workflows/benchmark_tests.yml: integration test job

Refs: https://arxiv.org/abs/2603.04356, https://robocasa.ai
Related: huggingface/lerobot#321

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Pepijn
2026-04-14 11:25:25 +02:00
parent a656a982af
commit 69ed70edeb
7 changed files with 705 additions and 0 deletions
+90
View File
@@ -310,3 +310,93 @@ jobs:
name: metaworld-metrics
path: /tmp/metaworld-artifacts/metrics.json
if-no-files-found: warn
# ── ROBOCASA365 ──────────────────────────────────────────────────────────
# Isolated image: lerobot[robocasa] only (robocasa, robosuite, mujoco chain)
robocasa-integration-test:
name: RoboCasa365 — build image + 1-episode eval
runs-on:
group: aws-g6-4xlarge-plus
env:
HF_USER_TOKEN: ${{ secrets.LEROBOT_HF_USER }}
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: false
lfs: true
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3 # zizmor: ignore[unpinned-uses]
with:
cache-binary: false
- name: Login to Docker Hub
uses: docker/login-action@v3 # zizmor: ignore[unpinned-uses]
with:
username: ${{ secrets.DOCKERHUB_LEROBOT_USERNAME }}
password: ${{ secrets.DOCKERHUB_LEROBOT_PASSWORD }}
- name: Build RoboCasa365 benchmark image
uses: docker/build-push-action@v6 # zizmor: ignore[unpinned-uses]
with:
context: .
file: docker/Dockerfile.benchmark.robocasa
push: false
load: true
tags: lerobot-benchmark-robocasa:ci
- name: Run RoboCasa365 smoke eval (1 episode)
if: env.HF_USER_TOKEN != ''
run: |
docker run --name robocasa-eval --gpus all \
--shm-size=4g \
-e HF_HOME=/tmp/hf \
-e HF_USER_TOKEN="${HF_USER_TOKEN}" \
-e HF_HUB_DOWNLOAD_TIMEOUT=300 \
-e MUJOCO_GL=egl \
lerobot-benchmark-robocasa:ci \
bash -c "
hf auth login --token \"\$HF_USER_TOKEN\" --add-to-git-credential 2>/dev/null || true
lerobot-eval \
--policy.path=pepijn223/smolvla_robocasa \
--env.type=robocasa \
--env.task=CloseFridge \
--eval.batch_size=1 \
--eval.n_episodes=1 \
--eval.use_async_envs=false \
--policy.device=cuda \
--output_dir=/tmp/eval-artifacts
"
- name: Copy RoboCasa365 artifacts from container
if: always()
run: |
mkdir -p /tmp/robocasa-artifacts
docker cp robocasa-eval:/tmp/eval-artifacts/. /tmp/robocasa-artifacts/ 2>/dev/null || true
docker rm -f robocasa-eval || true
- name: Parse RoboCasa365 eval metrics
if: always()
run: |
python3 scripts/ci/parse_eval_metrics.py \
--artifacts-dir /tmp/robocasa-artifacts \
--env robocasa \
--task CloseFridge \
--policy pepijn223/smolvla_robocasa
- name: Upload RoboCasa365 rollout video
if: always()
uses: actions/upload-artifact@v4 # zizmor: ignore[unpinned-uses]
with:
name: robocasa-rollout-video
path: /tmp/robocasa-artifacts/videos/
if-no-files-found: warn
- name: Upload RoboCasa365 eval metrics
if: always()
uses: actions/upload-artifact@v4 # zizmor: ignore[unpinned-uses]
with:
name: robocasa-metrics
path: /tmp/robocasa-artifacts/metrics.json
if-no-files-found: warn