Commit Graph

  • c09bbc1432 Fix Horovod build error due to missing exported symbols (#18958) Leonard Lausen 2020-08-19 16:16:50 +00:00
  • 8dcc653048 [1.x] Backporting TensorRT-Gluon Partition API (and TensorRT 7 support) (#18916) Serge Panev 2020-08-19 00:22:02 -07:00
  • b4da2dd59f remove upper bound (#18857) (#18910) JackieWu 2020-08-19 12:45:22 +08:00
  • 8794a0adf4 Numpy Dot Large Tensor Fix (#18925) Zhaoqi Zhu 2020-08-18 17:36:42 -07:00
  • d1ac7c849d [1.x] Backporting #18779 to v1.x (#18894) Sam Skalicky 2020-08-18 15:50:30 -07:00
  • 32994bbad1 Fix setting cudnn bias stride (#18905) Przemyslaw Tredak 2020-08-18 15:15:50 -07:00
  • c789d02044 Fix Python docs (#18924) Leonard Lausen 2020-08-18 16:57:35 +00:00
  • 9981e847ff [CI][1.x] Cherrypick: Upgrade unix gpu toolchain (#18186) (#18785) Chaitanya Prakash Bapat 2020-08-18 08:32:35 -07:00
  • 0afeb971a3 Fix instancenorm math equation (#18955) nihui 2020-08-18 22:03:57 +08:00
  • 6ae469a17e fix gelu to use erf based algorithm (#18827) (#18946) ciyong 2020-08-18 15:10:57 +08:00
  • e06ee4e1a7 Faster GPU frozen BatchNorm (#17368) Przemyslaw Tredak 2020-08-17 23:07:37 -07:00
  • 5e0db7a7c2 Cherry-pick #18635 to v1.7.x (#18935) (#18945) Sam Skalicky 2020-08-17 13:55:37 -07:00
  • 7a84fe11dd Cherry-pick #18635 to v1.7.x (#18935) Carin Meier 2020-08-16 12:53:47 -04:00
  • 2610c10701 Change Partition API's options_map to std::unordered_map (#18929) Serge Panev 2020-08-15 19:30:50 -07:00
  • be12c8d7af [Website] adjust website structure (#18839) Sheng Zha 2020-08-14 17:35:34 -07:00
  • daf8b438b0 Support extra inputs for subgraph ops (#18779) Sam Skalicky 2020-08-14 14:36:30 -07:00
  • 86e96dc7de Fix backward of arctan2 and rarctan2 scalar on GPU (#18440) Przemyslaw Tredak 2020-08-13 22:27:10 -07:00
  • 6b568fd2c1 Backporting backward inference from 2.x #18348 and #18378 (#18895) Serge Panev 2020-08-13 22:22:10 -07:00
  • ee80b77d42 Fix default CPU allocator memory alignment (#18885) bgawrych 2020-08-14 07:19:52 +02:00
  • d32ba4f24d [v1.7.x] backport Invoke mkldnn and cudnn BatchNorm when axis != 1 to v1.7.x (#18676) (#18890) Jake Lee 2020-08-13 22:19:03 -07:00
  • 344587f295 Safe accumulation for computing gradient in Embedding & Take (#18385) MoisesHer 2020-08-13 22:18:26 -07:00
  • d2d6408aaf bump version to 1.8.0 (#18899) Sheng Zha 2020-08-13 16:44:29 -07:00
  • a2b400c72e fix center element not being copied (#18917) Joshua Z. Zhang 2020-08-12 22:47:47 -07:00
  • 8c416d76ec Fix CI in v1.x branch (#18907) (#18912) Joe Evans 2020-08-12 17:13:20 -07:00
  • 9fbf3d3104 Fix CI in v1.x branch (#18907) Joe Evans 2020-08-12 10:12:06 -07:00
  • e2cbf6605e Revert "drop list support for gluon trainer (#18877)" (#18892) Leonard Lausen 2020-08-12 16:22:17 +00:00
  • 83d2af58c2 Gamma reparameterization gradient (#18852) Xi Wang 2020-08-12 15:25:17 +08:00
  • f2a8b977b0 Remove manually created symbolic link to ninja-build (#18906) Leonard Lausen 2020-08-12 01:51:11 +00:00
  • 016c166af4 remove upper bound (#18857) JackieWu 2020-08-12 03:54:46 +08:00
  • df0c31fbf1 [1.7.x][submodule] Upgrade to oneDNN v1.6 (#18875) bartekkuncer 2020-08-11 15:34:09 +02:00
  • 1711103717 [1.x][submodule] Update oneDNN to official v1.6 release (#18867) (#18867) bartekkuncer 2020-08-11 15:32:58 +02:00
  • e101d68d8b [Gluon] Add VAE demo (#18758) Xi Wang 2020-08-11 12:50:26 +08:00
  • 0b5b449483 Remove mention of nightly in pypi (#18635) (#18884) Sheng Zha 2020-08-10 12:19:04 -07:00
  • 651e24b281 [v1.7.x] backport Invoke mkldnn and cudnn BatchNorm when axis != 1 to v1.7.x (#18676) Jake Lee 2020-08-10 11:37:35 -07:00
  • d0e17e561f [Numpy] FFI: sort, argsort, vstack etc (#17857) Ke Han 2020-08-10 12:48:56 +08:00
  • 5c50475271 fix pooling_convention warning when convert model to onnx (#18529) Liu, Hao 2020-08-10 08:15:22 +08:00
  • d52d9c6f80 Revert "Add SOVERSION when build shared libmxnet.so library (#17815)" (#18882) Sheng Zha 2020-08-09 13:33:03 -07:00
  • 706c369fd0 fix trainer when the model involves share_parameters (#18880) Ziyue Huang 2020-08-09 07:55:16 +08:00
  • cc287a0fc2 Update windows_setup.md (#18874) Tong He 2020-08-08 13:09:24 +08:00
  • cf908fd483 [Numpy][Bugfix] Add hybridization test to loss layers (#18876) Xingjian Shi 2020-08-07 19:55:36 -07:00
  • d5fdcbf3ba drop list support for gluon trainer (#18877) Haibin Lin 2020-08-07 18:11:16 -07:00
  • dde635f8c7 Re-enable the linker version scripts for binary distribution (#18872) Leonard Lausen 2020-08-07 21:16:24 +00:00
  • 1694d2f037 [CI] remove data.mxnet.io usage for CI stability (#18871) Sheng Zha 2020-08-07 11:21:22 -07:00
  • 708a9002d9 Fix a bug in MXNet-TensorRT (#18870) Serge Panev 2020-08-07 10:46:22 -07:00
  • d101c3c2d8 Add SOVERSION when build shared libmxnet.so library (#17815) Gustavo Alvarez 2020-08-07 04:34:51 +02:00
  • a3eabf0d29 Fix MXLibInfoCompiledWithCXX11ABI (#18864) Leonard Lausen 2020-08-06 15:52:52 +00:00
  • 84f898409a ElementWiseSum fix for oneDNN (#18859) bgawrych 2020-08-06 04:32:39 +02:00
  • a78f137eb1 improve python api website ux - make toc sticky (#18863) Yang Shi 2020-08-05 14:24:46 -07:00
  • 0f65ef614e nb fix (#18858) Xi Wang 2020-08-05 10:48:50 +08:00
  • 3143aabb60 [v1.7.x] ElementWiseSum fix for oneDNN (#18777) bgawrych 2020-08-05 04:38:26 +02:00
  • 7b7cef5e0b MXNet-TRT: Add PrePartition param caching - move init_tensorrt_params logic (#18490) Serge Panev 2020-08-04 18:23:48 -07:00
  • 59e200a336 fix nn.dense doc (#18830) Haibin Lin 2020-08-04 17:01:23 -07:00
  • 2e97226585 Fix edge case when casting gluon Block before export (#18853) Leonard Lausen 2020-08-04 21:11:32 +00:00
  • b8eccc8675 fix set default website version rewrite rule for cdn (#18856) Yang Shi 2020-08-04 14:08:09 -07:00
  • eae6171cb5 [submodule] Upgrade to oneDNN v1.6 (#18822) bartekkuncer 2020-08-04 21:26:46 +02:00
  • 7a402194be Remove check for subgraph with cycles (#18555) Serge Panev 2020-08-04 10:34:21 -07:00
  • 95fa63f765 Update the onnx-tensorrt submodule - CI to TRT7 (#18574) Serge Panev 2020-08-03 17:15:02 -07:00
  • 7f2e314294 update setup.py (#18850) Haibin Lin 2020-08-03 16:09:48 -07:00
  • f872b43407 Protobuf_USE_STATIC_LIBS must be set on Apple too (#18851) Leonard Lausen 2020-08-03 20:11:06 +00:00
  • 2aa2702880 [v1.x] Mkldnn header fix v1x for nightly binaries (#18797) Chaitanya Prakash Bapat 2020-08-03 12:33:39 -07:00
  • 4bb82245ee Fixed python website double scroller and improve UX (#18845) Yang Shi 2020-08-03 12:30:13 -07:00
  • 7a5a48835b Fix broken link in docs/README.md (#18847) Iblis Lin 2020-08-04 03:28:08 +08:00
  • 5db8dee9be [ONNX export] Fixing spatial export for batchnorm (#17711) (#18846) Chaitanya Prakash Bapat 2020-08-03 12:05:20 -07:00
  • 534cdbca4c Create greetings.yml (#18842) Sheng Zha 2020-08-03 11:58:33 -07:00
  • 9fd2cce3c9 Update tests/README.md Docker instructions to match ci/README.md (#18848) kpuatamazon 2020-08-03 17:40:44 +01:00
  • 54b9e9c8d0 remove unnecessary usage of pretrained models, and prefer smaller size (#18844) Sheng Zha 2020-08-03 08:59:33 -07:00
  • 73d3a7bc9c [v1.x Backport] Fix softmax, logsoftmax failed on empty ndarray (#18602) (#18708) bgawrych 2020-08-03 16:20:49 +02:00
  • 51340d8f40 Add compiled_with_cxx11_abi API (#18836) Haibin Lin 2020-08-01 16:23:03 -07:00
  • 5a22193b55 [NumPy] allow mixed array types (#18562) Sheng Zha 2020-07-31 17:06:17 -07:00
  • 1a31cea22b Add unit tests for potri and potrf backward and check output shape in unit tests. (#18803) Joe Evans 2020-07-31 13:55:43 -07:00
  • 08a5ee3331 fix gelu to use erf based algorithm (#18827) Tao Lv 2020-08-01 03:38:20 +08:00
  • f4e62dfe96 Add Large Dim Checks for linalg Operators (#18816) Zhaoqi Zhu 2020-07-30 22:49:36 -07:00
  • ac3608932f Fixup move gluon.metric api docs (#18748) Leonard Lausen 2020-07-31 04:54:10 +00:00
  • 7a240061bb Enable DIST_KVSTORE by default in staticbuild (#18796) Leonard Lausen 2020-07-31 02:58:55 +00:00
  • aa53291855 add adaptive left margin for python site document body (#18828) Yang Shi 2020-07-30 19:53:27 -07:00
  • 045efb2784 [NumPy] DLPack refactor and npx.from_numpy (#18656) Sheng Zha 2020-07-30 19:19:33 -07:00
  • 608afef6fb Fix dirichlet flaky tests (#18817) Xi Wang 2020-07-31 02:30:25 +08:00
  • 84c9e0d46e [v1.x][LT] Add forward & backward linalg.gemm test for large size (#18825) Chaitanya Prakash Bapat 2020-07-30 10:40:32 -07:00
  • ca6bcf3480 adding error message when attempting to use Large tensor with linalg_syevd (#18807) Rohit Kumar Srivastava 2020-07-29 15:16:10 -07:00
  • 6bbd53107a Update clang-tidy integration (#18815) Leonard Lausen 2020-07-29 20:31:19 +00:00
  • b685fada9c use regex that is supported by all browsers (#18811) Yang Shi 2020-07-29 12:22:12 -07:00
  • 9308acaa6b remove other language bindings section from website api page (#18783) Yang Shi 2020-07-29 12:21:42 -07:00
  • 915f6b43de Remove deepnumpy reference and move Numpy tutorials to top level (#18798) Yang Shi 2020-07-29 11:28:37 -07:00
  • 85eb528c1f Add syrk test shape check (#18812) Zhaoqi Zhu 2020-07-28 23:08:53 -07:00
  • e9829e71a7 Cherry-pick large tensor support from #18752. (#18804) Joe Evans 2020-07-28 18:53:29 -07:00
  • 7bef9cb23b Back port optimization to broadcast_axis to MXNet1.x (#18773) Rohit Kumar Srivastava 2020-07-28 16:58:07 -07:00
  • 126636cc44 Fix naming in runtime_functions.sh (#18795) Leonard Lausen 2020-07-28 22:11:20 +00:00
  • f83dbac14e remove executor manager from API doc (#18802) Haibin Lin 2020-07-28 11:48:05 -07:00
  • 7908d7eb56 [numpy] fix flaky mixed precision binary error (#18660) Yiyan66 2020-07-28 15:11:19 +08:00
  • a807f6de32 [NumPy] loss for np array (#17196) Sheng Zha 2020-07-27 22:06:50 -07:00
  • d0093458e3 [1.x][LT] Add forward, backward test for linalg.gemm2 (#18784) Chaitanya Prakash Bapat 2020-07-27 20:28:43 -07:00
  • 74430a92f5 remove NLL in metric (#18794) phile 2020-07-28 06:44:54 +08:00
  • 9e77e8134e Update CUB and include it only for CUDA < 11 (#18799) Przemyslaw Tredak 2020-07-27 14:27:52 -07:00
  • 566d9d348a [v1.x] add large matrix tests for linalg ops: det, inverse, trsm, trmm (#18744) Manu Seth 2020-07-27 10:04:30 -07:00
  • 98b3f73bd0 add support for np.ndarray in autograd.function (#18790) Sheng Zha 2020-07-25 16:19:36 -07:00
  • c1db2d5636 Remove caffe plugin (#18787) Leonard Lausen 2020-07-25 16:58:45 +00:00
  • 2fbd1827ad Split up CI sanity test functions to enable fine-grained trigger (#18786) Leonard Lausen 2020-07-25 02:48:30 +00:00
  • 85ff00d5c1 Add Large Tensor Test for linalg_syrk (#18782) Zhaoqi Zhu 2020-07-24 15:55:05 -07:00
  • 06b5d227bb ONNX import: use Conv pad attribute for symmetrical padding (#18675) Serge Panev 2020-07-24 14:22:42 -07:00
  • e6de5ae475 Fix linalg_potri and linalg_potrf operators for large tensor. (#18752) Joe Evans 2020-07-24 09:50:46 -07:00