| --- |
| language: |
| - en |
| license: mit |
| tags: |
| - ACT |
| - LeRobot |
| - robotics |
| - imitation-learning |
| - behavior-cloning |
| - so101 |
| pipeline_tag: reinforcement-learning |
| library_name: lerobot |
| --- |
| |
| # LeRobot SO101 ACT task2-mixed_bs32_s60000 |
|
|
| ## Summary |
|
|
| This repository contains the final checkpoint for a ACT policy trained on `aswinkumar99/task2-mixed` for SO101 sponge pick-and-place experiments. |
|
|
| Dataset meaning: Task 2: Multiple Sponges - No Distractors (mixed layouts). |
|
|
| This ACT policy was trained for this dataset configuration and was not initialized from a published ACT base checkpoint. The visual backbone uses ImageNet-initialized ResNet-18 weights (`ResNet18_Weights.IMAGENET1K_V1`) as recorded in the training config. |
|
|
| ## Training Setup |
|
|
| - Dataset repo: `aswinkumar99/task2-mixed` |
| - Local dataset root during training: `/home/riftuser/datasets_combined/aswinkumar99/task2-mixed` |
| - Output directory during training: `/home/riftuser/outputs_matrix/act/task2-mixed_bs32_s60000` |
| - Batch size: `32` |
| - Training steps: `60000` |
| - Checkpoint save frequency: `15000` |
| - Data loader workers: `8` |
| - WandB project: `so101-layout-generalization` |
| - GPU: `NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition` |
| - Python: `CPython 3.12.13` |
| - CUDA: `12.9` |
| - Training start: `2026-04-23T21:07:01.112937+00:00` |
| - Training end: `2026-04-24T00:17:53` |
| - Approximate training duration: `3h 10m 51s` |
|
|
| - Vision backbone: `resnet18` |
| - Observation cameras: `observation.images.overhead`, `observation.images.wrist` |
| - Action chunk size: `100` |
| - Action steps predicted: `100` |
|
|
| ## Exact Training Command |
|
|
| ```bash |
| lerobot-train \ |
| --dataset.repo_id=aswinkumar99/task2-mixed \ |
| --dataset.root=/home/riftuser/datasets_combined/aswinkumar99/task2-mixed \ |
| --dataset.video_backend=torchcodec \ |
| --output_dir=/home/riftuser/outputs_matrix/act/task2-mixed_bs32_s60000 \ |
| --job_name=act_task2-mixed_bs32 \ |
| --batch_size=32 \ |
| --steps=60000 \ |
| --log_freq=200 \ |
| --save_freq=15000 \ |
| --save_checkpoint=true \ |
| --num_workers=8 \ |
| --wandb.enable=true \ |
| --wandb.project=so101-layout-generalization \ |
| --wandb.mode=online \ |
| --wandb.disable_artifact=true \ |
| --policy.type=act \ |
| --policy.device=cuda \ |
| --policy.push_to_hub=false |
| ``` |
|
|
| ## Repository Contents |
|
|
| - `pretrained_model/`: final downloadable model artifacts for inference/loading |
| - `training_state/`: optimizer, RNG, scheduler/state, and step information for resuming or auditability |
|
|
| ## Notes |
|
|
| - This repo stores the final checkpoint that was uploaded from the cloud training workspace. |
| - The checkpoint was trained with LeRobot tooling via `lerobot-train`. |
| - For SO101 experiments in this workspace, the dataset source was created by Aswinkumar. |
|
|
| ## Creator |
|
|
| Aswinkumar |
|
|
| - Website: [aswinkumar.me](https://aswinkumar.me) |
| - Hugging Face repo: <https://huggingface.co/aswinkumar99/LeRobot-SO101-ACT-task2-mixed_bs32_s60000> |
|
|
