tryolabs/luminoth

[Evaluation] - Support multiple splits

Open

#244 opened on Nov 13, 2018

View on GitHub
 (1 comment) (0 reactions) (0 assignees)Python (2,395 stars) (411 forks)batch import
Type: Feature Requestgood first issue

Description

Feature Request

I want to obtain the evaluation of different datasets of a single job lumi eval -h && lumi eval -c config_ssd.yml --split valid --split train --split test --watch --from-global-step 0

I understand that it is not so trivial because it was being kept as a summary of work scalars like this image

sounds very crazy?

Contributor guide

Tech stack
python
Domain
machine learningcli
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
mostly clear
Prerequisites
PythonDeep learning basics
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
The issue requests adding support for multiple splits (e.g., split valid split train split test) to the 'lumi eval' command. The user notes that currently evaluation aggregates scalars across splits, and implementing this may require reworking the summary of work scalars. Investigate the existing evaluation code in the repository, particularly the command line interface and the scalar aggregation logic. Check for any related PRs or comments that discuss implementation approaches. The maintainer may need to clarify the desired behavior for overlapping scalars across splits.
[Evaluation] - Support multiple splits · tryolabs/luminoth#244 | Good First Issue