Case safe chem by LJones-alt · Pull Request #5 · aarch-uol/IsaacLab-Laura

LJones-alt · 2026-03-17T16:47:01Z

Description

Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context.
List any dependencies that are required for this change.

Fixes # (issue)

Type of change

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (existing functionality will not work without user modification)
Documentation update

Screenshots

Please attach before and after screenshots of the change if applicable.

Checklist

I have read and understood the contribution guidelines
I have run the pre-commit checks with ./isaaclab.sh --format
I have made corresponding changes to the documentation
My changes generate no new warnings
I have added tests that prove my fix is effective or that my feature works
I have updated the changelog and the corresponding version in the extension's config/extension.toml file
I have added my name to the CONTRIBUTORS.md or my name already exists there

…-Laura into skillgen_demo

Copilot

Pull request overview

This PR primarily adds a collection of experiment artifacts (robomimic config snapshots), developer utility scripts, ROS2/DDS configuration changes, and Docker image/compose adjustments (including additional CUDA tooling, device passthrough, and OptiX-related mounting).

Changes:

Added many docs/training_runs/**/config.json snapshots for multiple models / datasets.
Added documentation-side utilities (checkpoint selection, URDF cleanup, experiment runner) plus LFS-tracked mesh assets.
Updated Dockerfiles / docker-compose and ROS2 DDS configs to support additional ROS2 + hardware/VPN workflows.

Reviewed changes

Copilot reviewed 93 out of 255 changed files in this pull request and generated 14 comments.

Show a summary per file

File	Description
docs/training_runs/good_data/bc_low_demos/20251007171556/config.json	Adds a BC training run config snapshot for low-dim demos.
docs/training_runs/Lift_med/model9/bc_med_data_/20251211123105/config.json	Adds a BC training run config snapshot for Lift_med model9 (med data).
docs/training_runs/Lift_med/model9/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model9 (med data).
docs/training_runs/Lift_med/model9/bc_high_data_/20251009101852/config.json	Adds a BC training run config snapshot for Lift_med model9 (high data).
docs/training_runs/Lift_med/model8/bc_med_data_/20251211123105/config.json	Adds a BC training run config snapshot for Lift_med model8 (med data).
docs/training_runs/Lift_med/model8/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model8 (med data).
docs/training_runs/Lift_med/model8/bc_high_data_/20251009101852/config.json	Adds a BC training run config snapshot for Lift_med model8 (high data).
docs/training_runs/Lift_med/model7/bc_med_data_/20251211123105/config.json	Adds a BC training run config snapshot for Lift_med model7 (med data).
docs/training_runs/Lift_med/model7/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model7 (med data).
docs/training_runs/Lift_med/model7/bc_high_data_/20251009101852/config.json	Adds a BC training run config snapshot for Lift_med model7 (high data).
docs/training_runs/Lift_med/model6/bc_med_data_/20251211123104/config.json	Adds a BC training run config snapshot for Lift_med model6 (med data).
docs/training_runs/Lift_med/model6/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model6 (med data).
docs/training_runs/Lift_med/model6/bc_high_data_/20251009101852/config.json	Adds a BC training run config snapshot for Lift_med model6 (high data).
docs/training_runs/Lift_med/model5/bc_med_data_/20251211123104/config.json	Adds a BC training run config snapshot for Lift_med model5 (med data).
docs/training_runs/Lift_med/model5/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model5 (med data).
docs/training_runs/Lift_med/model5/bc_high_data_/20251009101852/config.json	Adds a BC training run config snapshot for Lift_med model5 (high data).
docs/training_runs/Lift_med/model4/bc_med_data_/20251211123104/config.json	Adds a BC training run config snapshot for Lift_med model4 (med data).
docs/training_runs/Lift_med/model4/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model4 (med data).
docs/training_runs/Lift_med/model3/bc_med_data_/20251211123103/config.json	Adds a BC training run config snapshot for Lift_med model3 (med data).
docs/training_runs/Lift_med/model3/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model3 (med data).
docs/training_runs/Lift_med/model2/bc_med_data_/20251211123103/config.json	Adds a BC training run config snapshot for Lift_med model2 (med data).
docs/training_runs/Lift_med/model2/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model2 (med data).
docs/training_runs/Lift_med/model14/bc_med_data_/20251211123107/config.json	Adds a BC training run config snapshot for Lift_med model14 (med data).
docs/training_runs/Lift_med/model13/bc_med_data_/20251211123106/config.json	Adds a BC training run config snapshot for Lift_med model13 (med data).
docs/training_runs/Lift_med/model12/bc_med_data_/20251211123106/config.json	Adds a BC training run config snapshot for Lift_med model12 (med data).
docs/training_runs/Lift_med/model11/bc_med_data_/20251211123106/config.json	Adds a BC training run config snapshot for Lift_med model11 (med data).
docs/training_runs/Lift_med/model10/bc_med_data_/20251211123106/config.json	Adds a BC training run config snapshot for Lift_med model10 (med data).
docs/training_runs/Lift_med/model1/bc_med_data_/20251211123103/config.json	Adds a BC training run config snapshot for Lift_med model1 (med data).
docs/training_runs/Lift_med/model1/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model1 (med data).
docs/training_runs/Lift_med/model0/bc_med_data_/20251211123102/config.json	Adds a BC training run config snapshot for Lift_med model0 (med data).
docs/training_runs/Lift_med/model0/bc_med_data_/20251015111922/config.json	Adds a BC training run config snapshot for Lift_med model0 (med data).
docs/training_runs/Dev-IK-Rel-v0/model4/bc_new_demos/20250925163039/config.json	Adds a BC training run config snapshot for Dev-IK-Rel-v0 model4.
docs/training_runs/Dev-IK-Rel-v0/model3/bc_new_demos/20250925163039/config.json	Adds a BC training run config snapshot for Dev-IK-Rel-v0 model3.
docs/training_runs/Dev-IK-Rel-v0/model2/bc_new_demos/20250925163039/config.json	Adds a BC training run config snapshot for Dev-IK-Rel-v0 model2.
docs/training_runs/Dev-IK-Rel-v0/model1/bc_new_demos/20250925163039/config.json	Adds a BC training run config snapshot for Dev-IK-Rel-v0 model1.
docs/training_runs/Dev-IK-Rel-v0/model0/bc_new_demos/20250925163039/config.json	Adds a BC training run config snapshot for Dev-IK-Rel-v0 model0.
docs/sample_vial.obj	Adds an OBJ mesh asset via Git LFS pointer.
docs/ika_plate.obj	Adds an OBJ mesh asset via Git LFS pointer.
docs/conical_flask.obj	Adds an OBJ mesh asset via Git LFS pointer.
docs/beaker.obj	Adds an OBJ mesh asset via Git LFS pointer.
docs/run_experiments.sh	Adds a helper shell script to run experiment rollouts.
docs/macros_private.py	Adds a macros file for robomimic configuration in docs.
docs/find_best_checkpoints.py	Adds a script to locate/copy best validation checkpoints across ensemble models.
docs/clean_urdf.py	Adds a helper script to add inertial/collision defaults to URDF links.
docker/docker-compose.yaml	Adds nvoptix bind mount + hidraw device passthrough; adds a new named volume entry.
docker/Dockerfile.ros2	Adjusts ROS2 image base and expands apt dependencies for networking/control + vcstool.
docker/Dockerfile.base	Comments out singularity NVIDIA placeholders; adds CUDA toolkit installation and ML deps installs.
docker/.ros/fastdds.xml	Replaces default FastDDS profile with a Tailscale-unicast focused configuration.
docker/.ros/cyclonedds.xml	Removes CycloneDDS config file.
docker/.env.ros2	Comments out DDS env vars and adds ROS_DOMAIN_ID.
docker/.env.base	Adds a trailing whitespace line.
debug_ci.py	Adds a small local script to print the generated `docker compose` command for debugging.

+WANDB_ENTITY = "aarch-uol"
+
+# wandb api key (obtain from https://wandb.ai/authorize)
+# alternatively, set up wandb from terminal with `wandb login`
+WANDB_API_KEY = "f1bb6b7a94c7431472882c94835036d0156eb7b6"


+RUN set -euo pipefail && \
+    . /etc/os-release && \
+    case "$ID" in \
+      ubuntu) \
+        case "$VERSION_ID" in \
+          "20.04") cuda_repo="ubuntu2004";; \
+          "22.04") cuda_repo="ubuntu2204";; \
+          "24.04") cuda_repo="ubuntu2404";; \
+          *) echo "Unsupported Ubuntu $VERSION_ID"; exit 1;; \
+        esac ;; \
+      *) echo "Unsupported base OS: $ID"; exit 1 ;; \
+    esac && \


+# Detect Ubuntu version and install CUDA 12.8 via NVIDIA network repo (cuda-keyring)
+RUN set -euo pipefail && \
+    . /etc/os-release && \
+    case "$ID" in \
+      ubuntu) \
+        case "$VERSION_ID" in \
+          "20.04") cuda_repo="ubuntu2004";; \
+          "22.04") cuda_repo="ubuntu2204";; \
+          "24.04") cuda_repo="ubuntu2404";; \
+          *) echo "Unsupported Ubuntu $VERSION_ID"; exit 1;; \
+        esac ;; \
+      *) echo "Unsupported base OS: $ID"; exit 1 ;; \
+    esac && \
+    apt-get update && apt-get install -y --no-install-recommends wget gnupg ca-certificates && \
+    wget -q https://developer.download.nvidia.com/compute/cuda/repos/${cuda_repo}/x86_64/cuda-keyring_1.1-1_all.deb && \
+    dpkg -i cuda-keyring_1.1-1_all.deb && \
+    rm -f cuda-keyring_1.1-1_all.deb && \
+    wget -q https://developer.download.nvidia.com/compute/cuda/repos/${cuda_repo}/x86_64/cuda-${cuda_repo}.pin && \
+    mv cuda-${cuda_repo}.pin /etc/apt/preferences.d/cuda-repository-pin-600 && \
+    apt-get update && \
+    apt-get install -y --no-install-recommends cuda-toolkit-12-8 && \
+    apt-get -y autoremove && apt-get clean && rm -rf /var/lib/apt/lists/*
+
+
+ENV CUDA_HOME=/usr/local/cuda-12.8
+ENV PATH=${CUDA_HOME}/bin:${PATH}
+ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
+ENV TORCH_CUDA_ARCH_LIST=8.0+PTX
+


 ARG DOCKER_NAME_SUFFIX=""

-FROM isaac-lab-base${DOCKER_NAME_SUFFIX} AS ros2
+FROM isaac-lab-base AS ros2


+  - type: bind
+    source: /usr/share/nvidia/nvoptix.bin
+    target: /usr/share/nvidia/nvoptix.bin
+    read_only: true


+                            <address>100.71.76.112</address>
+                        </udpv4>


+    # Define paths - handle both host and Docker container paths
+    script_dir = Path(__file__).parent.resolve()
+    base_dir = script_dir / "/workspace/isaaclab/robomimic/Dev-IK-Rel-Insert-v1"


+
+
+## script to run experiments in the background 
+
+task="Dev-IK-Rel-Insert-v1"


+./isaaclab.sh -p $run_file --task $task --horizon $horizon --num_rollouts $num_rollouts --ensemble_size $ensemble_size --seed $seed --exp_name $exp_name --headless
+
+seed=107
+exp_name="${task}_ensemble_${ensemble_size}_seed_${seed}_${exp_type}"
+./isaaclab.sh -p $run_file --task $task --horizon $horizon --num_rollouts $num_rollouts --ensemble_size $ensemble_size --seed $seed --exp_name $exp_name --headless
+
+seed=115
+exp_name="${task}_ensemble_${ensemble_size}_seed_${seed}_${exp_type}"
+./isaaclab.sh -p $run_file --task $task --horizon $horizon --num_rollouts $num_rollouts --ensemble_size $ensemble_size --seed $seed --exp_name $exp_name --headless


+# NOTE: These are only needed for Singularity, not Docker
+# For Docker, comment these out as they override the real nvidia binaries
+# RUN touch /bin/nvidia-smi && \
+#     touch /bin/nvidia-debugdump && \
+#     touch /bin/nvidia-persistenced && \
+#     touch /bin/nvidia-cuda-mps-control && \
+#     touch /bin/nvidia-cuda-mps-server && \
+#     touch /etc/localtime && \
+#     mkdir -p /var/run/nvidia-persistenced && \
+#     touch /var/run/nvidia-persistenced/socket


LJones-alt and others added 18 commits September 25, 2025 10:55

93% policy

9273bde

Merge branch 'isaac-sim:main' into skillgen_demo

1a95d61

Merge branch 'isaac-sim:main' into skillgen_demo

0038a5b

some additional updates

148f1e7

Merge branch 'skillgen_demo' of https://github.com/aarch-uol/IsaacLab…

b0f68ae

…-Laura into skillgen_demo

added source docs to gitignore

cbd95ba

backup controller borked

a804d8e

fixed the DiffIK Rel Controller

971732b

with backup controller!

6d8cb19

backup controller update

4051543

addec copilot instructions

24a4ed2

idk how i messed that up, teleop fixed now

e85da64

upgraded for difusion

e031516

now with automated experiment running

03791a6

minor changes

750b20f

fixed comms

a8a7c5d

ready for case!

ceaa45b

case ready

526d0e7

Copilot AI review requested due to automatic review settings March 17, 2026 16:47

LJones-alt merged commit e4cd4f2 into aarch-uol:main Mar 17, 2026
2 of 7 checks passed

github-actions Bot added documentation Improvements or additions to documentation infrastructure isaac-mimic asset labels Mar 17, 2026

Copilot AI reviewed Mar 17, 2026

View reviewed changes

Navigation Menu

Case safe chem - #5

Case safe chem#5
LJones-alt merged 18 commits into
aarch-uol:mainfrom
LJones-alt:case_safe_chem

LJones-alt commented Mar 17, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants



		## script to run experiments in the background

		task="Dev-IK-Rel-Insert-v1"

Conversation

LJones-alt commented Mar 17, 2026

Description

Type of change

Screenshots

Checklist

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants