NillionNetwork/nada-by-example

Implement the MaxPool operation

Offen

#16 geöffnet am 14.08.2024

 (0 Kommentare) (0 Reaktionen) (0 zugewiesene Personen)Python (10 Forks)auto 404
enhancementhacktoberfesthelp wanted

Repository-Metriken

Stars
 (11 Sterne)
PR-Merge-Metriken
 (PR-Metriken ausstehend)

Beschreibung

Description

Much like PyTorch, our MPC AI/ML framework nada-ai implements reusable standard AI modules. Pooling operations are one of these modules. Average pooling is already implemented here but max pooling is not.

Use Case

Max pooling is a very common operation in AI models - particularly computer vision models.

Specific Requirements

Our work on Average pooling should already provide solid template to performing kernel-style operations. The bit that would need to change is the actual operation that the kernel performs (average -> max) which can be found here.

Contributor Guide