| --- |
| language: |
| - en |
| tags: |
| - robotics |
| - reinforcement-learning |
| - diffusion-policy |
| - hire |
| --- |
| |
| # HiRE: Base Policies for Finetuning |
|
|
| Behavior-cloning checkpoints for HiRE policy finetuning on MimicGen. Each task includes the selected checkpoint and its original Hydra configuration, which the finetuning loader uses to restore the policy architecture. |
|
|
| | Task | Demonstrations | BC seed | Checkpoint | |
| | --- | --- | --- | --- | |
| | Stack Three | 200 | 42 | `stack_three/checkpoint/state_75.pt` | |
| | Threading | 30 | 42 | `threading/checkpoint/state_100.pt` | |
| | Three Piece Assembly | 200 | 42 | `three_piece_assembly/checkpoint/state_75.pt` | |
|
|
| These are diffusion policies with a ResNet-18 visual encoder, a 1D UNet action head, and an eight-step action horizon. The selected epochs are 75, 100, and 75, respectively. The original checkpoint payloads are preserved, including both `model` and `ema` weights. The HiRE finetuning loader selects `model`. |
|
|
| ## Files |
|
|
| ```text |
| stack_three/ |
| .hydra/config.yaml |
| checkpoint/state_75.pt |
| threading/ |
| .hydra/config.yaml |
| checkpoint/state_100.pt |
| three_piece_assembly/ |
| .hydra/config.yaml |
| checkpoint/state_75.pt |
| manifest.json |
| ``` |
|
|
| Keep each `.hydra/config.yaml` with its checkpoint. `manifest.json` records the original repository revision, file paths, training overrides, and SHA256 hashes. |
|
|
| ## Finetuning |
|
|
| From the [HiRE code repository](https://github.com/JimmyZtHan/HiRE-Dice_RL), after installing dependencies and preparing demonstrations: |
|
|
| ```bash |
| source scripts/set_path.sh |
| python scripts/launch.py finetune stack_three \ |
| --checkpoint /absolute/path/to/HiRE-release/stack_three/checkpoint/state_75.pt \ |
| --seed 42 |
| ``` |
|
|
| Select `threading` or `three_piece_assembly` and the corresponding checkpoint path for the other tasks. The code handles the Python class paths in the original Hydra configurations. |
|
|
| Processed demonstrations and their matching `normalization.npz` files are required separately. Use the same task, camera order, delta-action convention, and demonstration subset as the BC policy. Tool Hang weights are not included in this bundle. |
|
|
| ## Source |
|
|
| The checkpoint files and configurations are copied without modification from [JimmyHan2004/MimicGen-pretrain](https://huggingface.co/datasets/JimmyHan2004/MimicGen-pretrain), at revision `e0d681881f64771604353b5d2535c71d87e5d4a1`. |
|
|
| [Project website](https://hire-project.github.io/) |
|
|
