# KrishKrosh/combo-krish-e7-pi05-base-step-2500

OpenPI checkpoint exported from Modal training.

## Checkpoint

- config: `pi05_trlc`
- experiment: `combo_krish_e7_pi05_base_h100x2_bs128_v2`
- uploaded checkpoint step: `2500`
- variant: `custom`
- includes `train_state`: `true`

## Dataset

- dataset: `Gongsta/krish-simpler-tshirt,Gongsta/e7-tshirt-folding`
- default prompt: "Pick up the t-shirt, stretch out the corners to lay it flat, and smooth out any wrinkles. Perform precise, sequential folds to condense it into a neat, perfectly aligned rectangle, matching the quality of a professional laundromat."

## Files

- `params/`: inference-ready OpenPI parameters
- `assets/`: normalization stats required by OpenPI serving
- `_CHECKPOINT_METADATA`: checkpoint metadata
- `checkpoint_info.json`: upload metadata
- `train_state/`: resume state for further training

## W&B

- run: `combo_krish_e7_pi05_base_h100x2_bs128_v2`
- url: https://wandb.ai/krish-life21/openpi/runs/rupr4km9
- run final step: `2400`

## Usage

Download the repo locally, then serve it with the OpenPI TRLC wrapper:

```bash
cd /workspace/openpi
uv run /workspace/recap/scripts/serve_trlc_openpi.py --variant custom --checkpoint-dir /path/to/downloaded/repo --port 8000
```