Commit Graph

  • 9c5acb452c Accelerate ROIPooling layer (#14894) Wang Jiajun 2019-07-12 18:17:28 -05:00
  • 2565fa2460 fix nightly CI failure (#15452) Lai Wei 2019-07-12 13:17:59 -07:00
  • 9ff6c46eda Add -R option to ci/build.py to avoid rebuilding containers (#15426) Pedro Larroy 2019-07-12 11:04:47 -07:00
  • 6a564bec3e cuda/cuDNN lib version checking. Force cuDNN v7 usage. (#15449) Dick Carter 2019-07-12 09:53:58 -07:00
  • 554b196559 Rebase #13757 to master (#15189) Yimin Jiang 2019-07-12 06:18:35 +08:00
  • 5ffd598499 CI: upgrade Julia version from 1.0.3 to 1.0.4 (#15502) Iblis Lin 2019-07-12 05:23:35 +08:00
  • 1f3195f2bd Julia docs (#15454) Aaron Markham 2019-07-11 14:23:20 -07:00
  • 68460a927d [Doc] Add MKL install method apt/yum into tutorial (#15491) Xinyu Chen 2019-07-11 11:29:36 +08:00
  • 7a83883ec3 Improve docs for AMP (#15455) Anirudh Subramanian 2019-07-10 20:15:29 -07:00
  • 6191dd7083 fix the bug on Scala Sparse (#15500) Lanking 2019-07-10 15:22:07 -07:00
  • 7d4d1bc26d Two fixes for info_gan.md example Code (#15323) Konrad Heidler 2019-07-10 04:32:39 +02:00
  • d82c89a9df Opperf: Support Python<3.6 (#15487) kshitij12345 2019-07-10 07:58:04 +05:30
  • 6b2b927871 [Perl] - simplify aliasing strategy (#15395) Robert Stone 2019-07-09 17:38:04 -07:00
  • e02c0a290e Upgrade MKL-DNN submodule to v0.20 release (#15422) Tao Lv 2019-07-09 10:07:25 +08:00
  • b62aaf3d98 fix comment (#15481) kshitij12345 2019-07-09 02:48:25 +05:30
  • 09c71bf314 Add Sparse NDArray support for Scala (#15378) Lanking 2019-07-08 11:48:30 -07:00
  • 1ae73de691 fix doc for sort and argsort (#15317) suyz526 2019-07-08 20:30:34 +02:00
  • 091fece543 fix fp32 flatten issue (#15351) Wuxun Zhang 2019-07-08 10:07:37 +08:00
  • a3ae309799 [MXNET-978] Higher Order Gradient Support reciprocal, abs. (#15413) kshitij12345 2019-07-08 02:35:09 +05:30
  • a6ed12fe4f Had a few PRs merged. Hope to become an official contributor and potentially a commiter. (#15451) Disi A 2019-07-07 07:07:20 -04:00
  • bc570dbd94 [OP] Add a new arange_like operator to contrib (#15400) Tao Lv 2019-07-06 14:14:04 +08:00
  • dfe923ac25 Update fp16 docs: Block.cast is inplace (#15458) Serge Panev 2019-07-06 03:08:33 +09:00
  • 728b8db572 Revise Symbol tutorial (#15343) Sergey Sokolov 2019-07-05 03:44:30 -07:00
  • 74dbadbc3f [TUTORIAL] Revise Naming tutorial (#15365) Sergey Sokolov 2019-07-05 03:43:16 -07:00
  • 612b9d1ed4 Update Horovod docs links in README (#15366) Serge Panev 2019-07-05 10:09:31 +09:00
  • 0ec48864f6 Expose get_all_registered_operators and get_operator_arguments in the… (#15364) Pedro Larroy 2019-07-04 14:53:51 -07:00
  • 8ebaa5c038 REAME MTCNN Link URL Error in original website (#15020) sunrongda 2019-07-04 20:02:17 +08:00
  • d7c542af3b [TUTORIAL] Gluon and Sparse NDArray (#15396) Thom Lane 2019-07-03 20:13:32 -07:00
  • faccc59bc0 Fix memory leak in NaiveEngine (#15405) Pedro Larroy 2019-07-03 17:36:56 -07:00
  • fc54781e08 enable TensorRT integration with cpp api (#15335) Haohuan Wang 2019-07-03 17:27:45 -07:00
  • 5078853650 remove comments from nano instructions (#15433) Aaron Markham 2019-07-03 12:40:52 -07:00
  • d49445fcdc Updating profiler tutorial to include new custom operator profiling (#15403) Zhaoqi Zhu 2019-07-03 10:26:04 -07:00
  • 3df3e2c616 [TUTORIAL] Gluon performance tips and tricks (#15427) Thom Lane 2019-07-03 09:31:49 -07:00
  • 6a8d9eb5fd [MXNET-978] Higher order gradient for sigmoid (#15288) Lin Yuan 2019-07-02 22:53:39 -07:00
  • 15475788ce Remove mhard-float option. This is already deprecated by Google. (#15435) Disi A 2019-07-03 00:05:24 -04:00
  • 512a491547 Temporarily Commenting out Flaky Test (#15436) Zhaoqi Zhu 2019-07-02 20:55:18 -07:00
  • c6bb2cec91 Use omp threads for cpu data loader (#15379) Zhennan Qin 2019-07-03 01:56:52 +08:00
  • 7210cc459d nano instructions (#15117) Aaron Markham 2019-07-01 16:20:39 -07:00
  • e368ea2489 Exclude external dependencies from MXNet JAR (and fix Clojure build) (#15361) v1.4.x tklein23 2019-07-01 23:51:32 +02:00
  • d74b9930ed Revert default return type for indices in argsort() and topk() back to float32 (#15360) Rohit Kumar Srivastava 2019-07-01 09:46:57 -07:00
  • b869ecd65f [Clojure] Add fastText example (#15340) Alexander Chalk 2019-06-30 15:58:35 -04:00
  • 06df38c5e6 point fix the vector declaration in MultiBoxDetection (#15300) Zhennan Qin 2019-06-29 15:06:04 +08:00
  • ca565a0028 Conversion from FP32 model to Mixed Precision model (#15118) Anirudh Subramanian 2019-06-28 14:22:04 -07:00
  • 11e6d4519a [AMP] Move topk from FP16_FP32_FUNCS to FP32_FUNCS (#15342) Haibin Lin 2019-06-28 14:21:32 -07:00
  • 8aaacde929 Update sparse_retain Documentation (#15394) Haibin Lin 2019-06-28 11:42:01 -07:00
  • e8f3e91a71 [Opperf] Make module/namespace of the operator parameterized (#15226) Sandeep Krishnamurthy 2019-06-28 11:02:32 -07:00
  • 92fce90671 Custom Operator Profiling Enhancement (#15210) Zhaoqi Zhu 2019-06-28 08:39:06 -07:00
  • cd19367dcb add 'asnumpy' dtype option to check_symbolic_backward (#15186) Hao Jin 2019-06-28 11:23:46 +08:00
  • 75a9e187d0 [backport 1.5.x]Fix Cached_op with static_shape=true (#15298) (#15380) 1.5.0 Lai Wei 2019-06-27 09:46:32 -07:00
  • 582489cebc Fix Cached_op with static_shape=true (#15298) Zhennan Qin 2019-06-27 11:08:26 +08:00
  • ba30644612 [DOC] Clarify that global pooling is going to reset padding (#15269) Pedro Larroy 2019-06-26 17:13:53 -07:00
  • 009907a8ed [C++] Improve inference script to support benchmark on Imagenet (#15164) Wuxun Zhang 2019-06-27 06:50:05 +08:00
  • 51acd4d328 [MXNET-1086] added sub and mul to ONNX->TensorRT conversion (#15344) Maxim Berman 2019-06-25 16:11:13 -07:00
  • 7fe478afc4 Fix build_ccache_wrappers: (#14631) Pedro Larroy 2019-06-24 01:20:33 -07:00
  • f44f6cfbe7 Extend Clojure BERT example (#15023) Dave Liepmann 2019-06-22 19:35:45 +02:00
  • 0340536394 indent changes (#15321) nihui 2019-06-22 16:44:29 +08:00
  • e6fad30e45 Efficient MXNet sampling in the multinomial distribution (#15311) Zixuan Wei 2019-06-22 14:26:27 +08:00
  • b4ce4e714a improve layernorm CPU performance (#15313) pengxin99 2019-06-22 07:30:33 +08:00
  • 8b5f376d17 [MXNET-1413] Adding Large Tensor support for sort operators (#15170) Rohit Kumar Srivastava 2019-06-21 13:21:20 -07:00
  • 3f8fd00540 Fixing duplication in operator profiling (#15240) Zhaoqi Zhu 2019-06-21 13:18:28 -07:00
  • 4a9e9f67a5 Typo fix in plan_memory relase -> release. (#15299) Disi A 2019-06-21 04:02:38 -04:00
  • c45d23b5e8 Proper bulking of ops not using FCompute (#15272) Przemyslaw Tredak 2019-06-20 16:50:54 -07:00
  • 2de0db0911 Showing proper error when csr array is not 2D in shape. (#15242) Piyush Ghai 2019-06-20 13:03:48 -07:00
  • 12c4226974 update committer info (#15289) Lin Yuan 2019-06-20 02:27:45 -07:00
  • 2b7fbc5f61 [bug] fix higher grad log (#15120) kshitij12345 2019-06-20 11:34:19 +05:30
  • 4d9667121a fixing var-seq-len rnn backward() operator (#15278) stephenrawls 2019-06-19 20:55:33 -07:00
  • 145f82d820 Updating SymbolBlock.imports to support different dtypes (#15230) Thomas Delteil 2019-06-19 18:14:39 -07:00
  • eb48370711 Added transform tutorial (#15114) Thom Lane 2019-06-19 15:38:05 -07:00
  • 6f60b9b372 fix span issue on tutorial index (#15279) Thom Lane 2019-06-19 15:31:45 -07:00
  • ccbbf6b4b7 Fix java install docs (#15250) Zach Kimberg 2019-06-18 17:44:52 -07:00
  • d7e2139393 [MXNET-1417][Performance] Caching Dynamic Shape Checking Result (#15262) Junru Shao 2019-06-18 11:06:41 -07:00
  • cab1dfad37 Upgrade archive utility and add back FC improvement (#15171) Anirudh Subramanian 2019-06-17 14:50:20 -07:00
  • 85aaa3a488 Fix horovod build failure when mxnet is built from source (#15213) Lin Yuan 2019-06-16 16:52:09 -07:00
  • c4ea674ed6 [Dependency Update] Bump up cuDNN & NCCL version (#15142) Jake Lee 2019-06-16 13:01:44 -07:00
  • 5b1603f5b3 Fix nightly build warning (#15248) Sergey Sokolov 2019-06-15 12:12:46 -07:00
  • 45844b2268 Sort benchmark output per op name, add some validation to CLI (#15249) Pedro Larroy 2019-06-14 21:10:01 -07:00
  • 13cf5dbe06 Profiler API Enhancements (#15132) Zhaoqi Zhu 2019-06-14 21:09:30 -07:00
  • 3b663ef6cc [MXNET-1415]Add MXEnginePushAsyncND and MXEnginePushSyncND C APIs (#15177) JackieWu 2019-06-15 02:31:11 +08:00
  • 41d35c4566 [TUTORIAL] Add multiple GPUs training tutorial (#15158) Sergey Sokolov 2019-06-14 09:55:41 -07:00
  • b8b352d55b Cleaned up profiling tutorial (#15228) Thom Lane 2019-06-13 13:47:58 -07:00
  • b58cf0ad55 Updated Image Augmentation tutorial to use Gluon Transforms. (#15197) Thom Lane 2019-06-13 13:46:31 -07:00
  • 09202f7f26 Improve static cached_op optimization (#15187) Xinyu Chen 2019-06-13 11:08:15 +08:00
  • a862270beb Make MXNDArrayFromDLPack backward compatible (#15227) Anirudh Subramanian 2019-06-12 20:06:13 -07:00
  • e5902ec626 Update float16 tutorial (#15107) Sergey Sokolov 2019-06-12 16:50:19 -07:00
  • 0be6d7ecd5 Add an utility for operator benchmarks (#14977) Sandeep Krishnamurthy 2019-06-12 11:47:31 -07:00
  • b2e6dace96 fix imresize interp docs (#15066) Aaron Markham 2019-06-12 10:45:59 -07:00
  • 027b547aa8 Bugfix: accept NN as an inter_method for augmentation in ImageRecordI… (#15221) Carlos Sánchez 2019-06-12 19:43:47 +02:00
  • 579c5abb9d Docs fixes (NDabs, Proposal, MultiProposal) (#15185) Aaron Markham 2019-06-12 10:32:48 -07:00
  • 2e2009457a Fixed a bug in Gluon DataLoader. (#15195) Chandana Satya Prakash 2019-06-12 02:40:54 -04:00
  • 769b882732 Add missing file to Maven clean (#15216) Zach Kimberg 2019-06-11 23:39:51 -07:00
  • 29fea487fd Readd Java HelloWorld example for intellij tutorial (#15217) Zach Kimberg 2019-06-11 21:39:14 -07:00
  • d32a3da7a7 [MKLDNN] Fix quantized act and concat (#15209) Xinyu Chen 2019-06-12 09:31:47 +08:00
  • 62a85f365b Broken link fixes for the website (#15205) Aaron Markham 2019-06-11 16:18:05 -07:00
  • 1c3e964fb6 fix the cmake cmd with wrong path (#15111) Neo Zhang Jianyu 2019-06-11 15:54:42 +08:00
  • 8b2ae57196 Fix MKLDNNDataReorderAsync (#15163) Zhennan Qin 2019-06-11 00:53:17 -05:00
  • 3c82ce2405 [MXNET-978] Second order gradient support for some unary operators (#14613) Lin Yuan 2019-06-10 11:13:22 -07:00
  • b64e00a26d Fix wrong description of output range of ToTensor (#14794) Alexander Grund 2019-06-09 07:12:05 +02:00
  • 8d8c5d5233 fix graident of boolean mask (#15175) JackieWu 2019-06-09 13:08:26 +08:00
  • 99f8fc9f91 don't check for nullptr before deleting; closes #14580 (#14901) bernie gray 2019-06-08 22:19:36 -04:00
  • 21d7ac00cc Fix installation dependencies (#14987) Aaron Markham 2019-06-08 17:22:17 -07:00