Commit graph

38 commits

Author SHA1 Message Date
Julien Demouth a079758fb9
Update README.md 2021-06-16 22:34:05 +02:00
bhsueh fd13bbdad1 [FT] Remove the source codes of FT and add the link of new repo 2021-04-16 20:03:43 -07:00
byshiue a64351f3f1
[FT] Add the deprecation in readme 2021-04-05 15:49:19 +08:00
menggeliu1205 596d11a72c
Fix the bug of mismatching datatype in print_to_screen (#866)
Co-authored-by: liumg <mengge.liu@mobvoi.com>
2021-03-11 07:02:07 +08:00
Kazuhiro Yamasaki c33299b322 Update URLs for GLUE downloader. 2021-03-07 13:27:22 +09:00
feifeibear 0f5ff94854
fix two obvious bugs. (#635) 2021-02-16 22:20:03 +08:00
byshiue 3d0d45b409
Byshiue patch 2 (#805)
* fix: fix the problem that we do not destroy the cublaslt Desc and lead to memory leak
2021-01-08 14:18:26 +08:00
byshiue ae76b894b9
Byshiue patch 2 (#788)
[FasterTransformer] feat: Update FasterTransformer v3.1
2020-12-14 07:28:11 +08:00
byshiue a095658e44
Fix: Fix the bugs of allocating workspace (#746,#747) 2020-11-09 10:24:56 +08:00
byshiue b2e89e6e80
[FT] FasterTransformer 3.0 Release (#696)
[FT] feat: Add FasterTransformer v3.0

1. Add supporting of INT8 quantization of cpp and TensorFlow op.
2. Provide the tools to quantize the model.
3. Fix the bugs that cmake 3.15 and 3.16 cannot build this project. 
4. Deprecate the FasterTransformer v1
2020-09-23 10:03:37 +08:00
BO-YANG HSUEH 1aa6813450
[FT] 1. Fix the bug of TensorRT plugin of FasterTransformer encoder. (#640)
* [FT] 1. Fix the bug of TensorRT plugin of FasterTransformer encoder.
2020-08-06 20:15:49 +08:00
372046933 108a783a77
fix typo (#605) 2020-07-21 23:10:01 +08:00
BO-YANG HSUEH 4972eb1f91
[FT] 1. Push the FasterTransformer v2.1 (#582) 2020-06-30 21:38:30 +08:00
BO-YANG HSUEH b50add4ada
Update README.md
Fix typo of FasterTransformer README.md
2020-06-23 22:10:47 +08:00
BO-YANG HSUEH 4bcf4269d1
Update README.md
Fix typo of FasterTransformer README.md
2020-06-23 22:06:42 +08:00
372046933 fe2cef59f5
Update translate_sample.py
fix typo
2020-06-19 17:47:47 +08:00
bhsueh 13e601c6e4 1. Fix the bugs of trt sample codes 2020-05-13 09:38:33 +00:00
BO-YANG HSUEH bee3ddfa0e
[FasterTransformer] Fix the bug of Readme. 2020-04-29 16:44:22 +08:00
bhsueh 5ee9b2ec03 1. [FasterTransformer] Fix the bug of encoder trt plugin. 2020-04-29 08:42:43 +00:00
bhsueh f0809df478 1. Update the README. 2020-04-03 09:07:20 +00:00
Bo Yang Hsueh 9519b9678c 1. Add License to FT. 2020-03-26 17:13:11 +08:00
BO-YANG HSUEH 4cbd73fcb9
Update encoder_gemm.h 2020-03-18 19:22:39 +08:00
bhsueh 77505ce75c 1. Fix LGTM alerts, remove useless module from python files. 2020-03-06 12:09:50 +00:00
BO-YANG HSUEH e7c94040e8
Fix the LGTM alert. 2020-03-06 19:38:42 +08:00
BO-YANG HSUEH 4b421a9656
Fix the LGTM alert 2020-03-06 19:37:14 +08:00
bhsueh ecfad37240 1. Fix bug of README. 2020-03-06 01:48:19 +00:00
bhsueh b69fec3bc8 1. Fix bug of README. 2020-03-06 01:46:18 +00:00
bhsueh 9560a304fa 1. Using install the OpenNMT-tf to replace clone the OpenNMT-tf repo. 2020-03-06 01:41:43 +00:00
bhsueh bd89bca344 [FasterTransformer] Add translation_sample, fix some bugs. 2020-03-05 11:21:22 +08:00
bhsueh b162523c38 1. Update the readme. 2020-03-03 09:11:01 +08:00
Przemek Strzelczyk 2ebdf718eb [FasterTransformer] Adding v2 2020-03-02 14:10:33 +01:00
xjia 61d96c2020 refine assert/cmake 2019-09-18 03:37:27 +00:00
xjia 392d6bee1c refine softmax/diff 2019-09-11 07:32:50 +00:00
xjia fd852b56a0 fix softmax max_value 2019-09-11 07:09:53 +00:00
xjia 488cc11967 refine softmax 2019-08-26 09:08:59 +00:00
xjia 5777ab97b7 delete duplicate files 2019-07-20 00:17:18 +08:00
xjia 9f00666591 support arbitrary sequence 2019-07-19 21:23:57 +08:00
Xipeng Li 75502be814 Adding FasterTransformer: A faster transformer layer inference implementation for BERT and other transformer based models. 2019-07-14 00:29:45 +08:00