lzhengning
/
JittorMirror
mirror of https://github.com/Jittor/Jittor
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
2,034 Commits 57 Branches 331 Tags 26 MiB
Commit Graph

1596 Commits

Author SHA1 Message Date
uyzhang 2c2e8abe59 fix slice setitem 2024-09-23 13:18:49 +08:00
uyzhang 0d5035443e fix setitem not in graph 2024-09-23 03:26:12 +08:00
uyzhang fa288cb4d9 Refactor acl_op.h to use __fp16 for alphaValue in the case of ACL_FLOAT16 dtype 2024-09-22 18:06:38 +08:00
uyzhang 9ff62acf7d Refactor acl_op.h to use __fp16 for alphaValue in the case of ACL_FLOAT16 dtype 
Refactor grad method for improved performance and synchronization
Index indices to int32
Fix getitem bug
Add getitem&setitem mask
 2024-09-22 16:41:43 +08:00
lidongyang 8888b25ea7 fix getitem bug 2024-09-22 02:30:16 +08:00
lidongyang 464009af42 add getitem&setitem mask 2024-09-21 22:57:54 +08:00
uyzhang a357a7913d Refactor acl_op.h to use __fp16 for alphaValue in the case of ACL_FLOAT16 dtype 2024-09-21 17:17:47 +08:00
uyzhang 631a9a3aaa Refactor grad method for improved performance and synchronization 2024-09-21 14:20:10 +08:00
lidongyang 0705ed9d8f index indices to int32 2024-09-20 22:10:15 +08:00
uyzhang 015bd10210 Refactor flip and squeeze operations for improved performance and synchronization 2024-09-20 21:54:49 +08:00
lidongyang 898ec600b4 polish getitem&setitem 2024-09-20 21:44:47 +08:00
lidongyang babd92a002 polish getitem&setitem -1 2024-09-20 20:01:45 +08:00
lidongyang cdad66c01d polish output dtype 2024-09-20 19:43:56 +08:00
张仪 18afb843ad Fix synchronization issue in acl_op.h 2024-09-19 19:52:25 +08:00
张仪 4006f242de fixed bugs 2024-09-18 17:33:23 +08:00
张仪 e47a74a497 Fix broadcasting issue in acl_compiler.py and add support for setting item in jt.Var 2024-09-14 16:00:15 +08:00
lidongyang 651b24e634 add sigmoid embedding silu 2024-09-13 03:19:25 +08:00
lidongyang 0641a50a5d change op file to acl_op.h 2024-09-12 22:29:20 +08:00
lidongyang e00e4f099c add getitem&setitem 2024-09-12 20:25:48 +08:00
张仪 c55d49a8de add new aclop 2024-09-12 20:14:22 +08:00
张仪 3beeec78b1 add new aclop & fixed some bugs 2024-09-12 17:11:23 +08:00
张仪 eb89ae19ed add new aclop 2024-09-07 22:11:39 +08:00
张仪 21580ce80e update aclnn 2024-09-07 18:18:00 +08:00
范孙奇 2c141fa996 fix dim=3 error 2024-09-05 20:14:36 +08:00
DongYang Li 4b907d493c Merge pull request #584 from liylo/module 
Make forward hook modifiy the inputs and outputs
 2024-09-04 16:33:04 +08:00
DongYang Li 79527c40e9 Merge pull request #583 from liylo/func 
Add support for block diag function
 2024-09-04 16:32:45 +08:00
DongYang Li a1fcd0f337 Merge pull request #503 from 514flowey/attention_mask 
add attention mask
 2024-09-04 16:19:46 +08:00
DongYang Li 96b97ccf55 Merge pull request #549 from fansunqi/bilinear 
check input1 and input2 shape in jt.nn.Bilinear()
 2024-09-04 16:16:31 +08:00
DongYang Li 818edc962e Merge pull request #558 from fansunqi/Upsample 
check input shape and scale factor's positiveness in jt.nn.Upsample
 2024-09-04 16:16:18 +08:00
DongYang Li 60d4f5a2ef Merge pull request #582 from liylo/master 
fix load_parameter for Parameterlist issue Jittor#581
 2024-09-04 16:15:49 +08:00
lidongyang 30b8a637de remove compatibility 2024-09-04 16:11:51 +08:00
liylo df442516ab forward hooks now could modifiy inputs and outputs 2024-08-28 21:35:12 +08:00
liylo 949c6ed676 init 2024-08-28 21:27:02 +08:00
liylo 1c5519acf2 simple implementation for block diag with proper grad 2024-08-28 21:18:56 +08:00
liylo c8ca6d30eb simple implementation for block diag 2024-08-28 21:13:00 +08:00
liylo ddaf3520e3 fix load 2024-08-28 20:50:37 +08:00
514flowey 1fbd56bb6d fix unique bug 2024-08-22 12:53:03 +08:00
张仪 b4244090ae first commit 2024-08-21 22:15:12 +08:00
514flowey 822955ac00 add several ffunctions 2024-08-20 15:08:19 +08:00
Yuxuan Han 1c0cf4c2e4 complement of test_aclop: error of scatter()-multiple and where() 2024-08-12 19:50:29 +08:00
Yuxuan Han b46264b9f8 complement of test_aclop 2024-08-12 19:28:01 +08:00
Yuxuan Han f353b18472 complement of test_aclop 2024-08-01 16:00:00 +08:00
Yuxuan Han 4deb69c4e5 Merge pull request #1 from Jittor/master 
Fixed the BUG of ACL op memory
 2024-07-26 21:20:43 +08:00
Yuxuan Han 550ca96a75 complement of test_aclop 2024-07-26 21:16:09 +08:00
张仪 c25ac3a4e8 Fixed the BUG of ACL op memory 2024-07-25 15:54:57 +08:00
hanyx 69b6dd3b42 Merge remote-tracking branch 'upstream/master' 2024-07-24 21:18:52 +08:00
Yi Zhang 496b771211 Update acl_compiler.py 2024-07-24 16:20:02 +08:00
Yi Zhang 29f2fbd853 Update compile_extern.py 2024-07-24 15:43:19 +08:00
张仪 53327feff2 feat: enable ACL optimization in split function 2024-07-24 15:25:10 +08:00
邓一轩 a755d64f9e FEAT! add floor_int 2024-07-24 15:13:57 +08:00
Yi Zhang 279e4113f3 Update compile_extern.py 2024-07-24 15:01:30 +08:00
Yi Zhang 140b17b824 Update acl_compiler.py 2024-07-24 14:53:04 +08:00
Yi Zhang 67d79a66d4 Merge pull request #572 from dengyx21/dev-dyx 
FEAT! add aclop unittest
 2024-07-19 17:05:00 +08:00
Yi Zhang 8a9c10d615 Format test_aclop.py 2024-07-19 17:04:42 +08:00
邓一轩 2b12e55447 FEAT! add aclop unittest 2024-07-19 17:01:17 +08:00
CHEN Xinsheng 9758b18c7d fix dtype mismatch in `nn.cross_entropy_loss` 2024-07-19 16:42:55 +08:00
邓一轩 8f6563cba9 FEAT! where,scatter,cumsum,gather,flip 2024-07-18 20:04:40 +08:00
lidongyang 121fee583d add no gpu device error 2024-07-12 15:07:35 +08:00
Jiapeng Zhang f7bc197200 fix load bugs 
fix load bugs of state
 2024-07-10 19:58:07 +08:00
hanyx fa8b332f32 ComplexNumber:polar,view_as_complex,view_as_real 2024-07-09 22:27:28 +08:00
Yi Zhang 3f0814b482 Update acl_compiler.py 2024-07-09 21:48:35 +08:00
张仪 2ae2f1d453 update acl 2024-07-09 19:50:35 +08:00
lidongyang 3b2ca1c2c0 Merge branch 'master' of https://github.com/Jittor/jittor 2024-07-09 14:28:57 +08:00
lidongyang a58c8c7988 polish nn.Sequential __getattr__ 2024-07-09 14:28:17 +08:00
DongYang Li 914cd170b4 Merge pull request #548 from fansunqi/binary_cross_entropy_with_logits 
check target and output shape in jt.nn.binary_cross_entropy_with_logits
 2024-07-08 17:16:26 +08:00
DongYang Li 6736ce68e3 Merge pull request #553 from fansunqi/conv_transpose3d 
modify stride positive check in jt.nn.conv_transpose3d/jt.nn.conv_transpose; add input shape check in jt.nn.conv_transpose3d/jt.nn.conv_transpose
 2024-07-08 17:15:38 +08:00
DongYang Li dde745407e Merge pull request #554 from fansunqi/ConvTranspose 
check stride positiveness and input shape in jt.nn.ConvTranspose
 2024-07-08 17:14:05 +08:00
DongYang Li 133307627e Update nn.py 2024-07-08 17:13:32 +08:00
DongYang Li 9983779d7a Merge pull request #551 from fansunqi/Conv1d_sp 
check input shape in jt.nn.Conv1d_sp
 2024-07-08 17:08:33 +08:00
DongYang Li bdd6bb6de5 Merge pull request #550 from fansunqi/Conv1d 
check input shape in jt.nn.Conv1d
 2024-07-08 17:07:57 +08:00
DongYang Li 2b57b2d988 Merge pull request #555 from fansunqi/Dropout2d 
check input shape in nn.Dropout2d
 2024-07-08 17:05:49 +08:00
DongYang Li c669b1219a Merge pull request #556 from fansunqi/zeroPad2d 
check input shape in jt.nn.ZeroPad2d
 2024-07-08 17:05:22 +08:00
JittorRepos 596368ae7c Merge pull request #557 from fansunqi/ReplicationPad2d 
check input shape in jt.nn.ReplicationPad2d
 2024-07-08 17:04:21 +08:00
JittorRepos 98d7c2d0fa Merge pull request #562 from fansunqi/unfold 
check parameter's positive in jt.nn.Unfold
 2024-07-08 17:03:02 +08:00
lidongyang c47549e673 add isin 2024-07-05 18:12:43 +08:00
DongYang Li dcd6c6b2be update version 2024-07-02 20:02:02 +08:00
fansunqi f4d4c9d55c check parameter's positive in jt.nn.unfold 2024-07-01 15:41:24 +08:00
Sunqi Fan c45dac35e6 Merge branch 'Jittor:master' into fold 2024-07-01 12:26:35 +08:00
fansunqi 45ccf3d2ac check parameters' positive in jt.nn.fold 2024-07-01 12:23:03 +08:00
DongYang Li 4196cb8154 update version 2024-06-25 16:49:44 +08:00
范孙奇 f358fb7518 check input shape and scale factor's positiveness in jt.nn.Upsample 2024-06-10 19:27:29 +08:00
范孙奇 969d810f55 resume 2024-06-10 19:26:40 +08:00
范孙奇 78b7cf091b check input shape and scale factor's positiveness in jt.nn.Upsample 2024-06-10 19:25:53 +08:00
范孙奇 c4480b7e3b check input shape in jt.nn.ReplicationPad2d 2024-06-10 19:08:53 +08:00
范孙奇 d31b0a244d check input shape in jt.nn.ZeroPad2d 2024-06-10 19:05:49 +08:00
范孙奇 1fba329474 check input shape in nn.Dropout2d 2024-06-10 17:02:05 +08:00
范孙奇 958708ed60 modify error information 2024-06-10 16:48:27 +08:00
范孙奇 e6e5949765 add stride check in jt.nn.ConvTranspose 2024-06-10 16:45:53 +08:00
范孙奇 2266d21a8b remove 3D(unbatch) description 2024-06-10 16:43:22 +08:00
范孙奇 db8fcb33da modify stride positive check in jt.nn.conv_transpose; add input shape check in jt.nn.conv_transpose 2024-06-10 16:39:05 +08:00
范孙奇 baf6b45cf1 add input shape check in jt.nn.transpose3d 2024-06-10 16:29:58 +08:00
范孙奇 8fd834465c modify stride positive check in jt.nn.transpose3d 2024-06-10 16:08:48 +08:00
范孙奇 ae0e52dca5 check input shape in jt.nn.ConvTranspose 2024-06-06 20:55:46 +08:00
范孙奇 d895cb9d36 jt.nn.Conv1d in_channels and out_channels must be positive 2024-06-06 20:39:10 +08:00
范孙奇 b4155d8021 jt.nn.Conv1d_sp in_channels and out_channels must be positive 2024-06-06 20:35:05 +08:00
范孙奇 a42198705b check input shape in jt.nn.Conv1d_sp 2024-06-06 20:25:29 +08:00
范孙奇 4d11325634 check input shape in jt.nn.Conv1d 2024-06-06 20:18:31 +08:00
范孙奇 7f6beb58b9 check input1 and input2 shape in jt.nn.Bilinear() 2024-06-06 20:04:35 +08:00
范孙奇 2f11e3bbbe check target shape and output shape in jt.nn.binary_cross_entropy_with_logits 2024-06-06 18:02:29 +08:00
lidongyang 393684f196 polish nn.Sequential attribute 2024-06-05 22:31:20 +08:00
Hanyuxuan f20ea9dcf1 fix illegal parameters of ConvTranspose and Pool,issue #478,#480,#481,#482,#483 2024-05-31 15:23:42 +08:00
Hanyuxuan 26963bc70f fix Pad2d with illegal padding,issue #464,#465,#466,#467 2024-05-31 14:50:47 +08:00
Hanyuxuan 64c6400070 check x.shape and kernel_size of Pool and Pool3d,issue #461,#463 2024-05-31 14:33:07 +08:00
Hanyuxuan c7e31604c2 fix illegal parameters of PixelShuffle of issue #458,fix validity of concat of issue #459 2024-05-30 19:03:17 +08:00
Hanyuxuan 981f60c381 fix illegal parameters of Conv2d issue #471,#472,#473,#474,#475,#476,#477 2024-05-30 16:10:06 +08:00
Hanyuxuan 102a689fee fix illegal parameters of Pool and Pool3d of issue #451,#453,#456,#457 2024-05-30 15:19:19 +08:00
Hanyuxuan 681174a606 a ValueError fix of issue #450 2024-05-30 14:29:34 +08:00
Hanyuxuan fe8fb30136 a IndexError fix of issue #448 2024-05-30 14:00:27 +08:00
Hanyuxuan 317defa7a1 fix: jt.Var.expand with valid index -1 2024-05-29 11:05:53 +08:00
zhc7 fce14c8d9d fix: a minimal quick fix for issue #544 2024-05-22 11:08:28 +08:00
Yi Zhang 2c1a5e14f1 Merge branch 'master' into master 2024-05-21 12:48:58 +08:00
DongYang Li f8fde94c3f Update version to 1.3.9.8 2024-05-20 21:43:48 +08:00
DongYang Li 4e41e6b070 Merge pull request #539 from fansunqi/issue523_branch 
fix issue 523;update jt.nn.Conv1d/Conv3d/conv2d/conv3d
 2024-05-20 21:42:16 +08:00
DongYang Li 21eaa919b9 Merge pull request #540 from fansunqi/issue522_branch 
fix issue 522,520,519,516; update jt.Pool/Pool3d
 2024-05-20 21:41:52 +08:00
DongYang Li f3744aa47d Merge pull request #541 from fansunqi/issue521_branch 
fix issue 521;update jt.nn.MaxUnpool2d/MaxUnpool3d
 2024-05-20 21:41:34 +08:00
lidongyang 9190180d8d polish rocm support 2024-05-20 21:34:26 +08:00
范孙奇 1bd014ca9b fix issue 521;update jt.nn.MaxUnpool2d/MaxUnpool3d 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-17 11:34:46 +00:00
范孙奇 e4981653e3 fix issue 522;update jt.Pool/Pool3d 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-17 11:22:12 +00:00
范孙奇 dd9ac69eec fix issue 523;update jt.nn.Conv1d/Conv3d/conv2d/conv3d 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-17 09:35:12 +00:00
uyzhang 871ed92fc4 fix: Add conditional import for change_function in __init__.py 2024-05-16 16:03:09 +08:00
DongYang Li 4efbbbf75c Merge pull request #443 from co63oc/patch-1 
Update mnist.py
 2024-05-16 15:20:07 +08:00
DongYang Li 9943ddf8de Merge pull request #536 from fansunqi/issue528_branch 
fix issue 528;update conv_transpose
 2024-05-16 15:19:19 +08:00
DongYang Li 5934b20720 Merge pull request #535 from fansunqi/issue529_branch 
fix issue 529;update contrib.argmax_pool()
 2024-05-16 15:18:14 +08:00
DongYang Li 9370896b35 Merge pull request #537 from fansunqi/issue527_branch 
fix issue 527,526;update jt.zeros/ones/full/randn/randint/random
 2024-05-16 14:47:53 +08:00
DongYang Li ee3c68ce7a Merge pull request #538 from fansunqi/issue525_branch 
fix issue 525;update jt.nn.Reflection2d/Replication2d
 2024-05-16 14:43:28 +08:00
DongYang Li 136a710775 polish PixelShuffle in nn.py 2024-05-16 14:35:21 +08:00
DongYang Li 75429f83b7 Merge pull request #534 from fansunqi/master 
fix issue 531,530;update jt.nn.PixelShuffle/jt.histc
 2024-05-16 14:27:24 +08:00
514flowey 96fda6dee5 Merge pull request #518 from 514flowey/complex 
add complex matmul, inv, qr, eig, and svd
 2024-05-16 13:38:53 +08:00
范孙奇 82595dc766 fix issue 525;update jt.nn.Reflection2d/Replication2d 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 13:49:26 +00:00
范孙奇 69fc229912 fix issue 526;update jt.randn/random/randint 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 13:34:06 +00:00
范孙奇 9f4c156e12 fix issue 527;update jt.zeros/ones/full 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 13:23:30 +00:00
范孙奇 9baccaed4d fix issue 528;update conv_transpose 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 12:21:21 +00:00
范孙奇 fc252af9a2 fix issue 529;update contrib.argmax_pool()-2 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 12:04:38 +00:00
范孙奇 72f72900d6 fix issue 529;update contrib.argmax_pool() 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 12:01:08 +00:00
范孙奇 0a1a8d738b fix issue 530; update jt.histc 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 11:31:28 +00:00
范孙奇 ab39f8283c fix issue 531 
Signed-off-by: 范孙奇 <fansq20@mails.tsinghua.edu.cn>
 2024-05-15 11:14:24 +00:00
uyzhang d587961209 Update ACL library and fix bugs in ACL integration 2024-05-15 19:03:04 +08:00
514flowey 290f2aec60 add complex matmul, inv, qr, eig, and svd 2024-05-08 21:09:07 +08:00
DongYang Li 4808184bee
Update version to 1.3.9.6 2024-05-07 10:33:25 +08:00
lidongyang 4c048d185a polish stack 2024-05-06 14:44:14 +08:00
DongYang Li ce783d2950
Merge branch 'Jittor:master' into master 2024-05-06 01:38:45 +08:00
lidongyang 6945c035c5 fix stack grad bug 2024-05-06 01:38:10 +08:00
uyzhang 82b3f99f51 Fix memory leak in TempAllocator::free() 2024-04-24 18:17:43 +08:00
lidongyang 3092bba175 polish atomicCAS use 2024-04-23 00:43:53 +08:00
lidongyang 185ae5d7eb add new cuda12.2 for g++11 2024-04-11 00:39:57 +08:00
DongYang Li aab6b0f784
Update version 1.3.9.5 2024-04-10 02:07:32 +08:00
lidongyang 8c8c5007d8 fix fp16 include 2024-04-10 02:03:00 +08:00
514flowey e5bfdfb162 add attention mask 2024-04-08 19:26:47 +08:00
MenghaoGuo af07705726
Merge pull request #502 from LDYang694/master 
merge jdiff
 2024-04-08 19:10:58 +08:00
DongYang Li c9c08f5cde
Update __init__.py 2024-04-08 18:56:08 +08:00
zjp_shadow 4f2fe9c4d1 Update compatibility 2024-04-02 16:38:12 +08:00
zhangjiapeng f045e196c1 Update Load 2024-04-01 18:00:34 +08:00
Dun Liang 2744992946 add fallback_func for static graph compiler 2024-01-03 03:57:14 +08:00
Dun Liang 4b9d777570 add doc 2023-12-29 03:55:56 +08:00
lidongyang 4ae9171578 fix merge conflict 2023-12-28 16:10:57 +08:00
Dun Liang 2a995d6cc4 update version to 1.3.9.2 2023-12-27 23:20:56 +08:00
Dun Liang fa40284b60 Merge branch 'master' of https://github.com/Jittor/jittor 2023-12-27 23:18:17 +08:00
Dun Liang 6e1327ba07 fix vcompiler bug 2023-12-27 15:16:06 +08:00
Dun Liang e260f2b6bd fitten code 2023-12-05 01:34:26 +08:00
lidongyang 9f5224819e Merge branch 'master' of github.com:Jittor/jittor 2023-12-04 15:47:49 +08:00
lidongyang 1f30881c6e polish GroupNorm&fix different op 2023-12-04 15:46:03 +08:00
Dun Liang 2bd05ce04e fix pyobj memleak 2023-11-29 14:10:16 +08:00
Dun Liang 881f769ba7 add tile and ne 2023-10-06 06:59:55 +08:00
Dun Liang e4272cdd4a polish computing graph liveness 2023-09-25 14:48:05 +08:00
Dun Liang ea9026aa6b fix memleak 2023-09-25 10:43:07 +08:00
lzhengning 6e78553016 remove unnecessary prints 2023-09-18 02:41:27 +08:00
lzhengning d21dfe65e6 Add SkipFirstBatchesSampler 2023-09-18 02:06:49 +08:00
Dun Liang 77a5f37f12 fix random sampler 2023-09-16 01:56:48 +08:00
Dun Liang 47d1483ca8 dont deep copy optim state dict 2023-09-15 13:49:13 +08:00
Dun Liang 40368453e8 polish bfloat16 2023-09-15 13:12:42 +08:00
Dun Liang c485fdc07b madd bf16 2023-09-08 22:43:31 +08:00
Dun Liang 2457c5ce4a mem opt for llm 2023-09-03 02:14:09 +08:00
lzhengning 9c69783f05 Merge branch 'master' of github.com:Jittor/jittor 2023-08-25 18:04:58 +08:00
Dun Liang 88582fa854 add reuse_np_array 2023-08-25 17:50:22 +08:00
zhouwy19 ca05be86f0
Update vcompiler.cc free unused output directly 2023-08-15 18:15:44 +08:00
Dun Liang dea40b30a3 bug fix for getitem 2023-08-10 03:19:57 +08:00
Dun Liang a6fca86633 polish nccl wrapper 2023-08-08 01:03:22 +08:00
Dun Liang c48216fb75 Merge branch 'master' of github.com:Jittor/jittor 2023-08-06 18:27:53 +08:00
Dun Liang 3a65eb3c43 add mpi barrier 2023-08-06 18:26:50 +08:00
yang guo ye 21b551411e
Merge pull request #486 from 514flowey/complex_number 
add complex matmul
 2023-08-06 17:05:52 +08:00
514flowey 14c39a7a53 add complex matmul 2023-08-06 16:58:50 +08:00
Dun Liang aa2dc39317 Merge branch 'master' of github.com:Jittor/jittor into HEAD 2023-08-02 00:15:43 +08:00
Dun Liang f03556533c MPI half type 2023-08-02 00:14:39 +08:00
lzhengning 6df06213e3 polish mpi dataset 2023-08-01 21:35:03 +08:00
Gword 7ea3c5530d save&load 2023-07-27 21:03:05 +08:00
Dun Liang f65c600508 remove print 2023-07-26 14:45:32 +08:00
Dun Liang 54112e0e2a Merge branch 'master' of github.com:Jittor/jittor 2023-07-26 00:35:39 +08:00
Dun Liang 39cff908a7 add lock for mpi compile_module 2023-07-26 00:34:34 +08:00
cxjyxxme 89ef365c1a install cutlass 2023-07-25 15:27:27 +08:00
Dun Liang 3f2d0313ed support detial nan checker 2023-07-24 12:41:24 +08:00
Dun Liang 4b35e70b65 polish code 2023-07-17 13:47:58 +08:00
Dun Liang 9a73639a09 polish acl interface 2023-07-13 16:07:48 +08:00
Dun Liang b8854f93d1 add acl support [version 1.3.8.7] 2023-07-13 14:36:57 +08:00
Dun Liang c99546586d update version to 1.3.8.6 2023-07-07 22:09:15 +08:00
Dun Liang 5b6c26ede7 Merge commit '1018e3b2' into HEAD 2023-07-07 22:08:03 +08:00
Dun Liang 1018e3b237 add save pytorch format 2023-07-07 19:18:51 +08:00
Dun Liang 192a11c6b6 add ccec support 2023-06-30 23:31:34 +08:00
Dun Liang f2fdf89fdc add fp16 acc 2023-06-27 22:22:00 +08:00
li-xl 640bfd9c6c fix save checkpoint bugs 2023-06-26 18:36:48 +08:00
Dun Liang a92051ec23 polish version 3.9 2023-06-25 22:58:20 +08:00