Issues

• Explanation on why off-the-self MAE models do not need proportional attention in ToMeAttention.forward()?

  facebookresearch/ToMe#49 · Filter by author Lopa07 opened on Nov 22, 2024

• Why do we need this code snippet for training MAE?

  facebookresearch/ToMe#44 · Filter by author redagavin opened on Aug 25, 2024
  1

• About using ToMe in ImageEncoderViT within segment anything

  facebookresearch/ToMe#42 · Filter by author yangzijia opened on Jul 5, 2024
  2

• AttributeError: 'ToMeVisionTransformer' object has no attribute 'blocks'

  facebookresearch/ToMe#40 · Filter by author Guoli-w opened on Jun 4, 2024
  3

• End-to-end inference doesn't be accelerated.

  facebookresearch/ToMe#39 · Filter by author rwshihhh opened on May 10, 2024
  3

• Video Classification off-the-shelf model

  facebookresearch/ToMe#37 · Filter by author JoakimHaurum opened on Oct 27, 2023

• Does ToMe work for focal modulation networks?

  facebookresearch/ToMe#23 · Filter by author subneed opened on Mar 21, 2023
  2

• Is there any way to generate heatmap with ToMe

  facebookresearch/ToMe#14 · Filter by author Nightwatch-Fox11 opened on Dec 8, 2022

• Adding support for HuggingFace vision Transformers

  facebookresearch/ToMe#2 · Filter by author NielsRogge opened on Oct 19, 2022
  3