ericboniface

The compatibility table becomes a pivotal information element for Substra users and enthusiasts, super useful. 1. However one might wonder which line he/she should consider. I suggest to add a...

Add a test on saving results

1

Follow-up of PR #337 , adding a unit test to it.

Elaborate a test scenario coherent with fertility context

3

- [x] Elaborate the scenario - [x] Identify a public dataset of choice for running this experimental scenario - [ ] Adapt library for working with this public dataset (in...

And/or also add saving results of scenarios run alone

Update sections to make a clearer split between data scientist activities vs. org/processes

1

So far the 6 sections follow roughly the data scientist activities: data, bias/modeling, evaluation, usage in production. However that is not that clear, as: - section 2 and 3 cover...

Improvements and additions to modelisation and evaluation elements

1

List of topics to study to improve/complement the existing eval elements and answer items: [ME] Modélisation, éval de perf. ME1- Qualité de l’apprentissage : limiter l’overfitting, l’évaluer ME2- Si biais...

List of topics to study to improve/complement the existing eval elements and answer items: [GCI] Gouvernance des projets, carte d’identité des modèles GCI1- Spécifications/Objectifs du modèle ou système d’IA :...

List of topics to study to improve/complement the existing eval elements and answer items: [U] Usages en production U1- Si usage en production, quelle procédure de validation pour passage en...

- https://www.coursera.org/learn/data-science-ethics#syllabus - https://www.coursera.org/specializations/ethics-in-ai#courses - https://courses.openmined.org - https://training.linuxfoundation.org/resources/free-courses/ethics-in-ai-and-big-data