admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-26 11:16:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

refactor(wandb): differentiate tags for reward model and policy training in cfg_to_group function

Browse Source
This commit is contained in:
Khalil Meftah
2026-04-23 13:50:45 +02:00
parent c6b8d36d48
commit 173e3495dc
1 changed files with 5 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/lerobot/common/wandb_utils.py
+5 -1
View File
@@ -41,8 +41,12 @@ def cfg_to_group(
return tag return tag
return tag[:max_tag_length] return tag[:max_tag_length]
if cfg.is_reward_model_training:
trainable_tag = f"reward_model:{cfg.reward_model.type}"
else:
trainable_tag = f"policy:{cfg.policy.type}"
lst = [ lst = [
f"policy:{cfg.policy.type}", trainable_tag,
f"seed:{cfg.seed}", f"seed:{cfg.seed}",
] ]
if cfg.dataset is not None: if cfg.dataset is not None: