facebookresearch / facebookresearch/SlowFast
Does the num_sample > 1 equals to increasing the epoch?
Open
- Dominant language
- Python
- Stars
- 7.4k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
I saw the code of function __getitem__() in class Kinetics. The NUM_SAMPLE >1 just increases the batch size but keeps the iterations unchanged. Considering the training time, does the operation equal to increase the epoch?
Contributor guide
Assessment
This issue has not been assessed yet.