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
.
0.3.11
a1ff216d
·
eos should not be included in variable lengthed mesh decoding
·
Dec 16, 2023
0.3.10
231c658b
·
residual quantize dropout should be applied to lfq as well
·
Dec 15, 2023
0.3.9
ecddf071
·
handle early break during generate loop from all eos
·
Dec 15, 2023
0.3.8
0e04c430
·
able to pass in a function on generate that transforms the face coordinates...
·
Dec 15, 2023
0.3.7
6e2f12c3
·
start moving eng stuff to another package
·
Dec 15, 2023
0.3.6
7a12023a
·
return loss breakdown and expose reconstruction loss
·
Dec 15, 2023
0.3.5
83a86a55
·
able to make model dimensions of fine vertex transformer to be smaller, given hierarchy
·
Dec 15, 2023
0.3.4
08b76d89
·
address a bug with auto appending eos, thanks to @qixuema
·
Dec 15, 2023
0.3.3
83e2c008
·
address
https://github.com/lucidrains/meshgpt-pytorch/issues/14
·
Dec 14, 2023
0.3.2
06a67a64
·
make gateloop cache work
·
Dec 14, 2023
0.3.1
374a0079
·
default kv cache to true on generate
·
Dec 14, 2023
0.3.0
5529a010
·
finish fine transformer kv caching, 7x speedup now
·
Dec 14, 2023
0.2.25
1318cfd3
·
first take care of caching on the coarse transformer, 3x speedup
·
Dec 14, 2023
0.2.24
39cda2e3
·
remove text related keys returned from dataset, for sharing same dataset for...
·
Dec 14, 2023
0.2.23
cdb14f51
·
dropout still helpful for smaller transformers
·
Dec 14, 2023
0.2.22
5cd25bdd
·
nvm confused myself
·
Dec 14, 2023
0.2.21
d88fa95a
·
fix can_eos logic in generate
·
Dec 14, 2023
0.2.20
85e557d0
·
not needed
·
Dec 14, 2023
0.2.19
f6d02605
·
sos is autohandled
·
Dec 14, 2023
0.2.18
ed909153
·
add a masked mean squeeze excite
·
Dec 14, 2023
Prev
1
2
3
4
5
6
7
8
9
10
11
Next