Skip to content
GitLab
Explore
Sign in
Tags
Tags give the ability to mark specific points in history as being important
This project is mirrored from
https://github.com/lucidrains/video-diffusion-pytorch
. Pull mirroring updated
Sep 19, 2024
.
0.7.0
2b0b75ff
·
remove group norms
https://arxiv.org/abs/2312.02696
·
May 03, 2024
0.6.3
f68f31ea
·
0.6.3
·
Nov 28, 2023
0.6.2
9630374a
·
0.6.2
·
Nov 26, 2023
0.6.1
9a71ea7a
·
try adding hf transformers as dep
·
Nov 26, 2023
0.6.0
601aceec
·
move first residual after initial temporal attention
·
Jul 24, 2022
v0.5.3
e4d9e9d2
·
fix a bug with bert encoding
·
Jul 14, 2022
v0.5.2
98d64b43
·
one more residual
·
Jun 16, 2022
v0.5.1
4590f60b
·
fix a bug where the highest resolution unet residual was not being concatted...
·
Jun 15, 2022
v0.5.0
485186f5
·
fix a bug where the highest resolution unet residual was not being concatted...
·
Jun 15, 2022
v0.4.2
275998c6
·
move to film-like conditioning for time
·
May 24, 2022
v0.4.1
89d3c229
·
move to film-like conditioning for time
·
May 24, 2022
v0.4.0
9aecf568
·
bring in dynamic thresholding, from Imagen paper, which purportedly improves...
·
May 24, 2022
0.3.19
7e5686e0
·
higher default lr
·
May 16, 2022
0.3.18
492ec084
·
take care of normalization and inverse normalization of image for ddpm, so...
·
May 14, 2022
0.3.17
cab9ec29
·
for numerical accuracy
·
May 10, 2022
0.3.16
cbd19780
·
allow for mask to be passed in for directly specifying which batch element...
·
May 10, 2022
0.3.15
6ed37a69
·
attempt to fix long description problem
·
May 10, 2022
0.3.14
fbb0ef5e
·
fix focus-present masking, thanks to @CiaoHe yet again!
·
May 10, 2022
0.3.12
aded599b
·
allow for non-strict load_state_dict
·
May 09, 2022
0.3.11
01f7f82b
·
initial temporal attention should be prenorm residual
·
May 09, 2022
Prev
1
2
3
Next