edeforas / edeforas/BeeDNN

add microbatch

Open
#67 0 comments 0 reactions 0 assignees View on GitHub
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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.