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/meshgpt-pytorch
. Pull mirroring updated
Sep 19, 2024
.
0.5.9
5dc77d9b
·
add ability to checkpoint the residaul quantizer module, addressing...
·
Jan 04, 2024
0.5.8
194f77c5
·
handle hierarchical kv cache correctly when prompted with mesh codes,...
·
Jan 04, 2024
0.5.7
99b9539c
·
handle hierarchical kv cache correctly when prompted with mesh codes,...
·
Jan 04, 2024
0.5.6
b912f140
·
default heinsen formulation to false
·
Jan 03, 2024
0.5.5
282567ef
·
0.5.5
·
Dec 31, 2023
0.5.4
16b14a07
·
patch
·
Dec 30, 2023
0.5.3
6e270f20
·
allow for mesh codes to be precomputed for training the transformer
·
Dec 22, 2023
0.5.2
f29f6d25
·
augmentation fn
·
Dec 22, 2023
0.5.1
b366fba6
·
kv cache now works with classifier free guidance
·
Dec 21, 2023
0.5.0
ddead0c7
·
quick patch, minor bump
·
Dec 21, 2023
0.4.23
822aa300
·
dry up some wandb + accelerate stuff
·
Dec 21, 2023
0.4.22
c08d4a69
·
can do a first pass over the dataset to cache, then set the file mode to read
·
Dec 21, 2023
0.4.21
5dc79c66
·
address
https://github.com/lucidrains/meshgpt-pytorch/issues/33
·
Dec 21, 2023
0.4.20
37858671
·
address
https://github.com/lucidrains/meshgpt-pytorch/issues/30
·
Dec 20, 2023
0.4.19
55be0b51
·
address
https://github.com/lucidrains/meshgpt-pytorch/issues/30
·
Dec 20, 2023
0.4.18
b338eaff
·
latest torch
·
Dec 20, 2023
0.4.17
4cdeb310
·
logcumsumexp should be available in torch 2.0 and onwards
·
Dec 20, 2023
0.4.16
9f03299a
·
0.4.16
·
Dec 20, 2023
0.4.15
34189ee0
·
use efficient associative scan based on heinsen sequence
·
Dec 20, 2023
0.4.14
8c390dc6
·
automatically transform the data_kwargs, if set on the dataset instance
·
Dec 20, 2023
Prev
1
2
3
4
5
6
7
8
…
11
Next