encoder 训练 loss不降问题
Open
- Dominant language
- Python
- Stars
- 36.9k
- Forks
- 5.2k
- PR merge metrics
- No merged PRs in 30d
Description
请问训练enconder 为什么loss一直都不降？跑了10k可以，loss一直保持在4.1多。然而，synthesizer的训练就很有效果。4000步就看到明显loss的下降。有高手给给意见吗。感谢
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by reproducing the encoder training run described in the issue and compare its loss behavior with the synthesizer training run. Inspect the encoder and synthesizer training entry points and configurations; done means identifying a reproducible cause or documenting the missing information needed to diagnose the loss plateau.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100