Search results

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

Status: Open.

#49 In facebookresearch/ToMe;

· Lopa07 opened on Nov 22, 2024

Why do we need this code snippet for training MAE?

Status: Open.

#44 In facebookresearch/ToMe;

· redagavin opened on Aug 25, 2024

About using ToMe in ImageEncoderViT within segment anything

Status: Open.

#42 In facebookresearch/ToMe;

· yangzijia opened on Jul 5, 2024

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

Status: Open.

#40 In facebookresearch/ToMe;

· Guoli-w opened on Jun 4, 2024

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

Status: Open.

#39 In facebookresearch/ToMe;

· rwshihhh opened on May 10, 2024

Video Classification off-the-shelf model

Status: Open.

#37 In facebookresearch/ToMe;

· JoakimHaurum opened on Oct 27, 2023

Does ToMe work for focal modulation networks?

Status: Open.

#23 In facebookresearch/ToMe;

· subneed opened on Mar 21, 2023

Is there any way to generate heatmap with ToMe

Status: Open.

#14 In facebookresearch/ToMe;

· Nightwatch-Fox11 opened on Dec 8, 2022

Adding support for HuggingFace vision Transformers

Status: Open.

#2 In facebookresearch/ToMe;

· NielsRogge opened on Oct 19, 2022