type
status
date
slug
summary
tags
category
icon
password
按照步骤看模块
VIT-VAE
1. Attention
这里用了两种 TemporalAxialAttention (Time dim in sequence of video)
和 SpatialAxialAttention
D → h(heads) * d
The attention mechanism is applied along the temporal axis (
T
), meaning that the model learns how the information at different time steps relates to each other两种 rotary_emb 使用方法不同
rotary_emb
module is used to embed the queries (q
) and keys (k
) to incorporate the frequency information, which is helpful in capturing periodic or sequential dependencies.
is_causal == True: Only attend to past time steps
Spatial —> Attention computed on H, W
dependencies between different spatial locations
is_causal == False: Attend to the whole image
推荐: diffusion-forcing 的作者 这里的rotary_emb attention代码来源
‣
关键词: 生成式3d
- Author:ran2323
- URL:https://www.blueif.me//article/14a71a79-6e22-8064-ae3a-e9732f58b13b
- Copyright:All articles in this blog, except for special statements, adopt BY-NC-SA agreement. Please indicate the source!