add microbatch
Open
enhancement
- Dominant language
- C++
- Stars
- 20
- Forks
- 8
- PR merge metrics
- No merged PRs in 30d
Description
To avoid memory usage bump on validation and reduce mem usage in case of conv2d
-sum gradient in microbatches,
-optimize and apply in minibatches
Contributor guide
No contributing guide indexed for this repository
Research direction
The issue does not identify files, tests, entry points, or a precise microbatching design. Start by locating the validation, Conv2D sum-gradient, and minibatch training paths, then clarify the intended memory behavior and scope. Done should include agreed microbatch behavior and validation that memory usage no longer increases in the reported cases.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100