Project-MONAI / Project-MONAI/MetricsReloaded

Evaluation suite - output

Open
#36 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
104
Forks
18
PR merge metrics
No merged PRs in 30d

Description

Context
Following the evaluation process on a set of references / output pairs, an aggregated report should be produced

Describe the solution you'd like
2 separate aggregations should be proposed:

  • Raw aggregation - All the calculated metrics for each reference/output pair should be made available
  • Population aggregation - Core statistics of the metrics over the population of reference/output pair should be produced including
    • central tendency and variation
    • indication of range
    • indication of cases at 5, 25, 50, 75, 95 percentile of performance for each metric
  • The reports should be made available as pandas data frame with the option to save in a specified format (csv, xlsx, html)

Additional context
https://github.com/Project-MONAI/MONAI/wiki/Evaluation-metrics-task-force
@jenspetersen
@danieltudosiu
@AReinke

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or implementation entry points are named. Start by reviewing the evaluation process and the linked Evaluation Metrics Task Force context, then identify where reference/output metrics are calculated; done means raw and population reports are available as pandas data frames and can be saved as CSV, XLSX, and HTML.

Written by the indexing model from the issue text.

Assessment

Tech stack
pandas, python
Domain
analytics, data
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.