Views
No views yet
goal_gen/upload_hf_checkpoints.sh.
Each subfolder contains the deployment-ready pretrained_model/ payload
(model.safetensors + config.json + pre/postprocessor + train_config.json).| Subfolder | Train step | Final train loss |
|---|---|---|
checkpoint-050000 | 50,000 | 0.032 |
1from huggingface_hub import snapshot_download
2ckpt_dir = snapshot_download(
3 "JayCao99/smolvla-rm65b-toy-v0.0",
4 allow_patterns="checkpoint-050000/*",
5)
6from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy
7policy = SmolVLAPolicy.from_pretrained(f"{ckpt_dir}/checkpoint-050000")