Commit graph

267 commits

Author SHA1 Message Date
Forkoz
eaac7a7b76
Update example script to include desc_act
It will help with people unwittingly making incompatible models.
2023-05-15 11:26:22 +00:00
潘其威(William)
570867c109
Merge pull request #79 from oobabooga/main
support loading quantized model with .pt file extension
2023-05-15 16:08:44 +08:00
oobabooga
86c7021285
Look for .pt files 2023-05-15 00:00:05 -03:00
潘其威(William)
262669112b
Merge pull request #76 from PanQiWei/gptj_fused_attention
Gptj fused attention
2023-05-14 16:21:27 +08:00
PanQiWei
d5429441ef add GPTJ fused attention module 2023-05-14 16:17:21 +08:00
PanQiWei
e1c564ac0e compatible with older pytorch version 2023-05-14 16:17:03 +08:00
PanQiWei
4586b3f31f update setup.py 2023-05-14 16:16:20 +08:00
PanQiWei
5445c67190 add library version comparison help functions 2023-05-14 16:16:06 +08:00
潘其威(William)
7c248cebf6
Merge pull request #43 from PanQiWei/faster-llama
Faster llama
2023-05-14 13:09:10 +08:00
PanQiWei
e83c9fc8dd update setup.py 2023-05-14 13:08:26 +08:00
PanQiWei
de33d26d67 fix bugs 2023-05-14 13:07:18 +08:00
PanQiWei
2273f9ef39 refactor file structure for triton kernels 2023-05-14 11:49:10 +08:00
PanQiWei
fef1a4fe4b make code clean and extendable 2023-05-12 20:11:55 +08:00
PanQiWei
d718d63e9c add import_utils.py for commonly used module importation 2023-05-12 19:58:48 +08:00
潘其威(William)
6f887f666a
Update 02-Advanced-Model-Loading-and-Best-Practice.md 2023-05-12 19:47:05 +08:00
PanQiWei
c5ff195764 skip fused module injection instead of raising error if it's not supported yet. 2023-05-12 19:36:00 +08:00
PanQiWei
f159aeabb6 refactor .from_quantized api and improve model loading strategy 2023-05-12 18:09:50 +08:00
PanQiWei
69610329d2 add _fused_base.py 2023-05-12 18:09:23 +08:00
潘其威(William)
393a2fbac2
Update README.md 2023-05-12 13:47:30 +08:00
潘其威(William)
e5c267e289
Update README.md 2023-05-12 13:46:41 +08:00
潘其威(William)
d6d099a1d1
Merge branch 'main' into faster-llama 2023-05-12 13:39:24 +08:00
PanQiWei
4bb10fda49 groupsize -> group_size 2023-05-12 13:37:52 +08:00
潘其威(William)
560cf92d7d
Merge pull request #62 from lszxb/fix_incorrect_pack
fix incorrect pack while using cuda, desc_act and grouping
2023-05-08 10:35:30 +08:00
潘其威(William)
8b67f7de2f
Merge pull request #59 from Sciumo/setup_conda
Setup conda
2023-05-08 10:34:10 +08:00
lszxb
174ef81995 fix incorrect pack while using cuda, desc_act and grouping 2023-05-07 20:44:47 +08:00
Sciumo
ee4ca934aa add conda cuda include directory if found 2023-05-05 14:28:04 -04:00
Sciumo
81f3dfe39c add conda cuda include directory if found 2023-05-05 14:27:11 -04:00
qwopqwop200
3ff6ab18cb
Merge branch 'main' into faster-llama 2023-05-06 00:20:29 +09:00
潘其威(William)
7c33fa2fa4
Merge pull request #58 from TheBloke/TheBloke_faster-llama_groupsize_fix
Fix bug caused by 'groupsize' vs 'group_size' and change all code to use 'group_size' consistently
2023-05-05 23:00:59 +08:00
TheBloke
1b3329b399 Fix 'groupsize' -> 'group_size' in all other .py files. I haven't touched any CUDA kernels in case there's any complexity there I don't understand 2023-05-05 14:44:16 +01:00
TheBloke
f61ce12271 Change 'groupsize' to 'group_size' everywhere. Turns out this is easier than 'groupsize' due to dependencies in other files. 2023-05-05 13:36:00 +01:00
TheBloke
f64c71e779 Change referenes to 'group_size' to 'groupsize' to match rest of this file 2023-05-05 13:21:13 +01:00
PanQiWei
374ce21066 release v0.1.0 2023-05-05 00:18:50 +08:00
PanQiWei
753c261388 update README.md 2023-05-05 00:15:33 +08:00
PanQiWei
d79aec7bd0 update README.md 2023-05-04 23:06:32 +08:00
PanQiWei
fe3456100c bug fix from commit 3c108d4232 2023-05-04 22:34:16 +08:00
PanQiWei
e4d476be16 update README.md 2023-05-04 22:17:38 +08:00
PanQiWei
6cba6e7123 reformat code 2023-05-04 22:16:08 +08:00
PanQiWei
1c6bb69fae fix attribute name error 2023-05-04 22:10:33 +08:00
潘其威(William)
771b650a7c
Merge pull request #38 from PanQiWei/faster-cuda-no-actorder
Faster cuda no actorder
2023-05-04 21:47:19 +08:00
qwopqwop200
b19c59541b
fix bug 2023-05-04 13:17:10 +09:00
qwopqwop200
908248114e
fix bug 2023-05-04 13:15:52 +09:00
qwopqwop200
b14d42e68a
bug fix 2023-05-04 13:03:38 +09:00
qwopqwop200
b0bc0b0358
bug fix 2023-05-04 13:03:11 +09:00
qwopqwop200
208d660920
fix bug 2023-05-04 10:04:00 +09:00
qwopqwop200
f51a92ed79
support faster and model load strict 2023-05-04 09:53:28 +09:00
qwopqwop200
cc992c21bd
Merge branch 'faster-cuda-no-actorder' into faster-llama 2023-05-04 09:09:09 +09:00
qwopqwop200
d49281bc5d
support faster and model load strict 2023-05-04 09:07:34 +09:00
qwopqwop200
c8504f0660
support faster and model load strict 2023-05-04 09:06:52 +09:00
qwopqwop200
34201dbff9
support faster and model load strict 2023-05-04 09:05:07 +09:00