facebookresearch / facebookresearch/SlowFast
Only can train several epochs ?
Open
- Dominant language
- Python
- Stars
- 7.4k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
Every time I train the model for about ten epochs，procedure finally goes to fail and it shows me that...
In .yaml，I set the DECODING_BACKEND:pyav，TRAIN.BATCH_SIZE: 2 , NUM_WORKERS:0. It terribly weried!
my gpu is only one GeFore RTX 3060 Laptop, and my train data is just eight about 10s video clips.
Can anyone help me?
Contributor guide
Assessment
This issue has not been assessed yet.