modelscope / modelscope/DiffSynth-Studio
Wan2.2-TI2V-5B lora训练崩溃，全是噪声
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.1k
- Forks
- 1.3k
- Avg merge
- 13h 12m
- Merged PRs (30d)
- 45
Description
各位大佬好，我在训练Wan2.2-TI2V-5B model的时候想初步尝试能否成功，就用了自己收集的十几个video,来训练，结果训练出来的model，其实也就两个epoch之后就产生出来全是噪声。 我不清楚这是否是数据集数量过少或者是lora的其他设置不合理导致的崩溃，但是原理上小批量数据训练几个epoch应该不会导致模式崩塌吧。
dataset_repeat=10
--learning_rate 1e-4
--num_epochs 5
--remove_prefix_in_ckpt "pipe.dit."
--output_path "./models/train/Wan2.2-TI2V-5B_lora"
--lora_base_model "dit"
--lora_target_modules "q,k,v,o,ffn.0,ffn.2"
--lora_rank 32 \
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No source files or tests are named. Start by reproducing the Wan2.2-TI2V-5B LoRA training with the reported dataset size, repeat count, learning rate, epochs, target modules, and rank, then inspect the training configuration and logs around the second epoch. Done means identifying whether the failure is caused by the dataset or settings and documenting a reproducible fix.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100