lunar
|
618a5f50ee
|
Add transpose operator when replace Conv1d with qlinear_cuda_old
|
2023-06-05 23:11:18 +08:00 |
|
潘其威(William)
|
bf521cbe7b
|
Merge pull request #134 from TheBloke/TB_benchmark
add command flags inject_fused_attention and inject_fused_mlp
|
2023-06-05 23:02:36 +08:00 |
|
潘其威(William)
|
8a1616c63c
|
Merge pull request #102 from PanQiWei/peft_integration
Peft integration
|
2023-06-05 22:55:34 +08:00 |
|
PanQiWei
|
129884c598
|
update version to 0.3.0.dev0
|
2023-06-05 22:53:50 +08:00 |
|
PanQiWei
|
b132d774e3
|
update README
|
2023-06-05 22:53:17 +08:00 |
|
TheBloke
|
edb13d493e
|
Default inject_fused_attention and mlp to True, matching defaults
|
2023-06-03 17:58:40 +01:00 |
|
TheBloke
|
4617629f0c
|
Support setting inject_fused_attention and inject_fused_mlp to False
|
2023-06-03 17:48:36 +01:00 |
|
PanQiWei
|
923fc87a11
|
Merge branch 'main' into peft_integration
|
2023-06-03 19:10:41 +08:00 |
|
潘其威(William)
|
023bb1c593
|
Merge pull request #125 from PanQiWei/support-32dim
Support 32dim
|
2023-06-03 19:08:29 +08:00 |
|
潘其威(William)
|
95a4381f50
|
Merge pull request #126 from PanQiWei/support-cuda-64dim
Support cuda 64dim
|
2023-06-03 19:08:12 +08:00 |
|
潘其威(William)
|
810ed4de66
|
Merge pull request #132 from EliEron/patch-1
Specify UTF-8 encoding for README.md in setup.py
|
2023-06-03 10:57:52 +08:00 |
|
qwopqwop200
|
f4820f2988
|
change qlinear cuda support 64dim
|
2023-06-03 07:30:34 +09:00 |
|
qwopqwop200
|
8951212ab3
|
change setup
|
2023-06-03 07:29:19 +09:00 |
|
qwopqwop200
|
e04c3b86cc
|
add cuda
|
2023-06-03 07:28:35 +09:00 |
|
qwopqwop200
|
5fc2064e1a
|
Rename autogptq_cuda_kernel.cu to autogptq_cuda_kernel_64.cu
|
2023-06-03 07:27:45 +09:00 |
|
qwopqwop200
|
446e12d3de
|
Rename autogptq_cuda.cpp to autogptq_cuda_64.cpp
|
2023-06-03 07:27:31 +09:00 |
|
EliEron
|
eeb8b78a55
|
Specify Encoding when reading README.md
Prevents UnicodeDecodeError from being raised in certain locals.
|
2023-06-02 20:54:32 +02:00 |
|
潘其威(William)
|
b4fdd8d264
|
Merge branch 'main' into peft_integration
|
2023-06-02 19:11:59 +08:00 |
|
PanQiWei
|
7206705456
|
set version to 0.2.1
|
2023-06-02 19:07:56 +08:00 |
|
qwopqwop200
|
2df7d7105d
|
support 64 cuda dim
|
2023-06-02 19:54:37 +09:00 |
|
qwopqwop200
|
b03f53294f
|
support 64dim cuda
|
2023-06-02 19:53:50 +09:00 |
|
qwopqwop200
|
90106d7c34
|
support cuda 64dim
|
2023-06-02 19:49:38 +09:00 |
|
PanQiWei
|
65c0115b86
|
update README
|
2023-06-02 18:18:11 +08:00 |
|
PanQiWei
|
0e609bec40
|
only append CUDA_VERSION to release version string when in github actions
|
2023-06-02 18:16:38 +08:00 |
|
qwopqwop200
|
0891ea4036
|
support 32dim triton]
|
2023-06-02 19:05:55 +09:00 |
|
qwopqwop200
|
b3654a68c3
|
support 32dim triton kernel
|
2023-06-02 19:04:12 +09:00 |
|
PanQiWei
|
50ac2ad4bc
|
update README
|
2023-06-02 10:59:36 +08:00 |
|
PanQiWei
|
113884d976
|
Merge remote-tracking branch 'origin/main'
|
2023-06-02 10:57:16 +08:00 |
|
PanQiWei
|
b248a2655a
|
update README
|
2023-06-02 10:56:57 +08:00 |
|
潘其威(William)
|
f948b56c07
|
Merge pull request #123 from jllllll/main
Fix and extend build_wheels.yml workflow
|
2023-06-02 10:12:20 +08:00 |
|
jllllll
|
0f1793b554
|
Revert "Remove workflow restriction for testing"
This reverts commit e62bda1c1e.
|
2023-06-01 20:49:42 -05:00 |
|
jllllll
|
3c6a002be5
|
Clean up workflow sdist creation
|
2023-06-01 20:35:30 -05:00 |
|
jllllll
|
e62bda1c1e
|
Remove workflow restriction for testing
|
2023-06-01 20:27:40 -05:00 |
|
jllllll
|
996382788b
|
Finalize workflow fix
|
2023-06-01 13:58:24 -05:00 |
|
jllllll
|
198e079da4
|
Restrict build_wheels.yml to minimum compute 6.0
|
2023-06-01 13:25:04 -05:00 |
|
jllllll
|
a0063fc9db
|
Add GitHub Actions bypass for cuda check to setup.py
|
2023-06-01 13:07:00 -05:00 |
|
jllllll
|
3084422095
|
Merge branch 'PanQiWei:main' into main
|
2023-06-01 12:50:34 -05:00 |
|
PanQiWei
|
b5db750c00
|
update setup.py
|
2023-06-02 01:39:56 +08:00 |
|
jllllll
|
2b96343e87
|
Update build_wheels.yml (#1)
|
2023-06-01 12:39:56 -05:00 |
|
PanQiWei
|
6a37f7c266
|
update setup.py
|
2023-06-02 00:03:44 +08:00 |
|
PanQiWei
|
bc61e51394
|
update README
|
2023-06-01 10:35:17 +08:00 |
|
PanQiWei
|
7ae89f282a
|
update build_wheels.yml
|
2023-06-01 01:48:29 +08:00 |
|
PanQiWei
|
31b8c1313e
|
update build_wheels.yml
|
2023-06-01 01:34:46 +08:00 |
|
PanQiWei
|
d53a30d351
|
update build_wheels.yml
|
2023-06-01 01:16:10 +08:00 |
|
PanQiWei
|
ac7dd9bc1f
|
update build_wheels.yml
|
2023-06-01 01:03:35 +08:00 |
|
潘其威(William)
|
a63f8fd523
|
Merge pull request #120 from PanQiWei/add_build_wheels_workflow
Add build wheels workflow
|
2023-06-01 00:43:38 +08:00 |
|
PanQiWei
|
d780ef5eef
|
update build_wheels.yml
|
2023-06-01 00:42:31 +08:00 |
|
PanQiWei
|
407e5d8133
|
add workflow to build wheels
|
2023-06-01 00:39:09 +08:00 |
|
PanQiWei
|
0ece40ca25
|
update setup.py
|
2023-06-01 00:38:35 +08:00 |
|
PanQiWei
|
402973259f
|
update setup.py
|
2023-06-01 00:18:43 +08:00 |
|