Commit Graph

14 Commits (c2e62c4b75e4334e54ab661c4e166f1ea3f3a344)

Author SHA1 Message Date
vipcxj c2e62c4b75 fix tensor not in same device error 2026-05-11 22:17:12 +08:00
jingyaogong 0d1ae504f2 [update] requirement 2026-05-11 13:44:12 +08:00
jingyaogong 4cfd48ebf3 [update] train pipeline 2026-05-08 15:04:39 +08:00
jingyaogong 0b85f814ed [fix] existing bugs 2026-05-07 19:07:26 +08:00
jingyaogong 2607758417
Merge pull request #3 from zhuxiaoxuhit/fix/multiple-bugs
Fix bugs: label leak, repetition penalty, stop condition, O(n²) generation
2026-05-07 19:04:09 +08:00
zhuxiaoxuhit c06144907e Fix 4 bugs in training and inference
1. Fix EOS label leak in multi-turn masking (omni_dataset.py)
2. Fix repetition penalty for negative logits (model_minimind.py, model_omni.py)
3. Fix stream_generate stop condition (model_omni.py)
4. Fix O(n^2) tolist in generation loop (model_omni.py)
2026-05-07 17:50:44 +00:00
jingyaogong 09b6315ee9 [update] mimi device 2026-05-07 13:11:55 +08:00
jingyaogong 60442b425e [update] readme 2026-05-07 10:23:39 +08:00
jingyaogong 3a29069104 [update] readme 2026-05-07 08:58:18 +08:00
jingyaogong d00782515b [update] readme 2026-05-06 22:37:33 +08:00
jingyaogong 3ef557f77a [update] training pipeline params 2026-05-06 15:56:08 +08:00
jingyaogong e12d136c8c [update] readme 2026-05-06 14:05:18 +08:00
jingyaogong c009567d06 [update] minimind-3o 2026-05-06 13:53:28 +08:00
jingyaogong 30b9d1c96b [init] LICENSE 2026-05-06 13:52:48 +08:00