KrishKrosh/krish-simpler-tshirt-from-pi05-trlc-23k-step-4500

OpenPI checkpoint exported from Modal training.

Checkpoint

• config: pi05_trlc
• experiment: krish_simpler_tshirt_from_pi05_trlc_23k_h100x2_bs128_v2
• uploaded checkpoint step: 4500
• variant: custom
• includes train_state: true

Dataset

• dataset: Gongsta/krish-simpler-tshirt
• 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: krish_simpler_tshirt_from_pi05_trlc_23k_h100x2_bs128_v2
• url: https://wandb.ai/krish-life21/openpi/runs/v553sx68
• uploaded step val/loss: 0.06015676259994507
• uploaded step train/loss: 0.011019379831850529
• run val/best_loss: 0.04028164967894554
• run final step: 4700

Usage

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

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