Commit Graph

275 Commits

Author SHA1 Message Date
Phil Wang
2a6828e6fb fix comment 2023-03-22 09:56:50 -07:00
Phil Wang
978ba2db57 another tab/spaces fix 2023-03-22 09:33:47 -07:00
Phil Wang
916000c8bf fix consistent tabs / spaces 2023-03-22 09:27:13 -07:00
Phil Wang
aa9b939edd add some comments, and fix use of g_val 2023-03-22 09:22:19 -07:00
Phil Wang
a43cd2008d add some code in test_optim.py, although it seems to be failing 2023-03-22 09:14:05 -07:00
Phil Wang
9b656f461a follow advice of Tim to fix update of momentum vs parameters in blockwise 8 bit 2023-03-22 07:52:59 -07:00
Phil Wang
369a51c432 switch all eps to beta2 2023-03-10 14:08:35 -08:00
Phil Wang
6c377b39b6 always pass beta2 into all the 1state functions 2023-03-10 13:00:59 -08:00
Phil Wang
abbe65adfc beta2 is actually accessible in kOptimizerStatic8bit1StateBlockwise 2023-03-10 12:50:14 -08:00
Phil Wang
19b9ef34b9 whoops 2023-03-10 08:59:49 -08:00
Phil Wang
c99b44f774 do the epsilon beta2 switcharoo within the cuda code, and not within the python class (so that the state dict still makes sense) 2023-03-10 08:57:59 -08:00
Phil Wang
8618bed001 swap the order in which momentum and parameters are updated in ops.cu 2023-03-10 08:39:06 -08:00
Phil Wang
c5582724d5 missed adagrad 2023-03-09 14:05:45 -08:00
Phil Wang
af03430992 fix weight decay for lion to be decoupled, using a switch 2023-03-09 14:03:07 -08:00
Phil Wang
ead570a43e remove something rmsprop specific 2023-03-09 11:58:31 -08:00
Phil Wang
c83888aa1a use epsilon as beta2 for lion, complete most of the logic in kernel.cu for all functions 2023-03-09 11:54:54 -08:00
Phil Wang
64bb1ae8d1 add a sign function, for lion 2023-03-09 11:10:28 -08:00
Phil Wang
8de29fc364 forget about tests for now, will test live on local enwik8 training 2023-03-09 10:11:32 -08:00
Phil Wang
cb4c3c8c66 do a bunch of typical bookkeeping before getting to main lion logic 2023-03-09 10:10:19 -08:00
Phil Wang
d43ea9722c make sure interface is correct 2023-03-09 09:45:33 -08:00
Phil Wang
7247cb4554 initial commit, slowly work from interface into the kernel 2023-03-09 08:08:46 -08:00
Tim Dettmers
0f5c394870 Added version 0.37.0. 2023-02-01 20:27:01 -08:00
Tim Dettmers
de53588934 Added Int8 matmul support for all GPUs. Full backward support. 2023-02-01 20:09:31 -08:00
Tim Dettmers
92ab6a8d5f
Merge pull request #119 from stas00/patch-1
improve install instructions
2023-02-01 19:21:36 -08:00
Stas Bekman
c5372a8567
improve install instructions 2023-01-05 13:34:51 -08:00
Tim Dettmers
1341fb44ad Fixed issue where the CUDA SETUP was not printed. 2023-01-04 03:50:53 -08:00
Tim Dettmers
3901ebf7ae Added CUDA 12.0 support; removed CC 3.0 support. 2023-01-04 02:28:33 -08:00
Tim Dettmers
b3de19218e Added error message for unexpected CUDA exception. 2023-01-03 06:57:07 -08:00
Tim Dettmers
81990491ff
Merge pull request #113 from Borzik/fix-warnings
Import missing warn function
2023-01-03 15:46:58 +01:00
Tim Dettmers
9180b4cc11 Added additional error message for cudart error #85 2023-01-03 06:44:11 -08:00
Tim Dettmers
dfb049f8e4 Added Python >= 3.8 requirement. 2023-01-03 06:20:06 -08:00
Tim Dettmers
211ad594df Added error+instructions for unsupported CUDA 10.0 version #82 2023-01-03 06:07:35 -08:00
Felix Borzik
f3800bab75 import warn function 2023-01-03 13:23:34 +00:00
Tim Dettmers
9d353ca786
Merge pull request #87 from lostmsu/main
Add `device` and `dtype` parameters to `StableEmbedding`
2023-01-02 13:22:45 +01:00
Tim Dettmers
7a6563b6c8 Default to CPU library on CUDA error+small refactor. 2023-01-02 03:47:09 -08:00
Tim Dettmers
d9112dc55b
Merge pull request #110 from BlackHC/cublaslt_version
Improve cc version detection for cublaslt
2023-01-02 12:35:53 +01:00
Tim Dettmers
336e24696c CUDASetup only executed once + fixed circular import. 2023-01-02 03:31:43 -08:00
Tim Dettmers
df9a9b0c4c
Merge pull request #77 from Cyberes/main
Allow hiding of the welcome message
2023-01-02 11:28:17 +01:00
Tim Dettmers
be5cecb88f
Merge branch 'main' into main 2023-01-02 11:23:17 +01:00
Tim Dettmers
f0ec93d016
Merge pull request #76 from tomaarsen/cleanup
Cleanup involving a handful of failures, some optimization and a lot of code quality improvements
2023-01-02 11:19:28 +01:00
Tim Dettmers
c91f592ad7
Merge branch 'main' into cleanup 2023-01-02 11:19:16 +01:00
blackhc
ed17aa9a31 Don't mark it as failure though. 2022-12-29 23:50:48 +00:00
blackhc
7b39a5511d Fix issue #97 2022-12-29 23:47:21 +00:00
Tim Dettmers
c059bd2848 Added additional blocksizes: {64, 128, 256}. 2022-11-20 14:18:15 -08:00
Tim Dettmers
eb028e6ebc Fixed k-bit quantization maps. 2022-11-19 07:24:03 -08:00
Tom Aarsen
b104ce3b62
Merge branch 'main' into cleanup 2022-11-17 15:22:29 +01:00
Tim Dettmers
08fa2e7b01 Fixed bug in cpu quant; faster GPU dequant. 2022-11-07 18:06:18 -08:00
Tim Dettmers
62a333ac40 Added pre/post calls do quantize_blockwise. 2022-11-06 17:17:51 -08:00
Tim Dettmers
e0e697b150 Fixed blockwise test and logic. 2022-11-06 16:36:31 -08:00
Tim Dettmers
6bc2b992be Added blocksizes 2048, 1024, and 512 to blockwise quant. 2022-11-06 16:27:48 -08:00