---
library_name: pytorch
pipeline_tag: robotics
tags:
- robotics
- imitation-learning
- diffusion-policy
---

# WRM_delta

Complete checkpoint series for `WRM_delta` trained on the local
`WRM_grasp_cylinder_different_sizes_lero_tightness` LeRobot dataset.

- GPULab job: `5bffdea4-5db8-4fd8-825f-37974aa66219`
- Training steps: through `100000` using the run's configured save interval
- Checkpoints: all numbered milestones plus final EMA-enabled `last.pt`
- SHA-256: `68cf84c725e41d2d98b9cc7050a4697016eeb30e774e1aa8197e28bacbc3e2ae`
- Source configuration: `resolved_config.yaml`
- Metrics: `metrics.jsonl`

The repository is private and intended for AIRO-Doffy real-robot evaluation.