Commit Graph

  • 837c7e43c1 SymbolBlock.imports ignore_extra & allow_missing (#19156) Sam Skalicky 2020-09-16 17:30:06 -07:00
  • d2e6452042 [1.x] Backport of intgemm #17559 (#19099) kpuatamazon 2020-09-16 17:41:35 +01:00
  • ec62fcf0d9 [Submodule] Upgrade to oneDNN v1.6.3 (#19153) Tao Lv 2020-09-16 10:55:25 +08:00
  • 606933f2c2 TensorRT: add int8 with calibration (#19011) Serge Panev 2020-09-15 15:58:54 -07:00
  • db07b7e5a1 Update README.md (#19145) Sheng Zha 2020-09-15 12:13:08 -07:00
  • 11e1c0b2b9 Fix comment and cmake variables and search paths in FindMKL.cmake (#19152) Sylwester Fraczek 2020-09-15 19:05:05 +02:00
  • 9dfac79d5a [1.x] Backport Fix for duplicate subgraph inputs/outputs (#16131) (#19112) Sam Skalicky 2020-09-14 22:57:05 -07:00
  • 179262b1e3 Fix legacy codepath detection feature for decorated HybridBlocks (#19143) Leonard Lausen 2020-09-15 05:14:21 +00:00
  • e1bcb336fd [1.x] Backport 'Update CUB and include it only for CUDA < 11 #18799' (#18975) Serge Panev 2020-09-14 18:12:38 -07:00
  • 9383282996 [1.x] Backport #19103 (#19117) Sam Skalicky 2020-09-14 10:31:36 -07:00
  • 2697573bbf updated version (#19134) Sam Skalicky 2020-09-14 09:45:34 -07:00
  • 0e292804e2 add Jackie Wu as a new PPMC member (#19136) Sheng Zha 2020-09-13 23:23:19 -07:00
  • 6cbdfa5151 Elemwise binary op backward fix (#19106) Zhaoqi Zhu 2020-09-12 22:15:42 -07:00
  • 8b5687467a Fix race condition in NaiveEngine::PushAsync (#19108) (#19122) Leonard Lausen 2020-09-12 18:58:31 -07:00
  • f1acda73e0 Update setup.py for darwin builds (#19130) Sheng Zha 2020-09-12 08:39:25 -07:00
  • 5c1aadcd4b Add cmake flag USE_FATBIN_COMPRESSION, ON by default for CUDA >= 11 (#19123) Dick Carter 2020-09-11 23:37:31 -07:00
  • 856e2555b1 add Ciyong Chen as a new committer (#19115) Sheng Zha 2020-09-11 21:19:18 -07:00
  • 3240e1eb0c set v1.7 as default website version (#19120) Yang Shi 2020-09-11 19:37:16 -07:00
  • 1f158197aa Fix race condition in NaiveEngine::PushAsync (#19108) Leonard Lausen 2020-09-11 15:50:11 -07:00
  • d144f3b317 add v1.7 website entries to general version dropdown (#19116) Yang Shi 2020-09-11 11:24:59 -07:00
  • b888d3c71a [v1.x] Update onnx support to work with onnx 1.7.0 with most CV models (#19017) Joe Evans 2020-09-11 10:17:23 -07:00
  • e5a781430b Improve createOpState for custom ops (#19103) Sam Skalicky 2020-09-10 23:51:26 -07:00
  • 95e1814ba5 [PIP] add build target in cmake for osx compat (#19110) Sheng Zha 2020-09-10 21:38:38 -07:00
  • 131fa6931c Fix default context handling in Gluon 2 (#19113) Leonard Lausen 2020-09-10 19:47:31 -07:00
  • 2d077db1c9 Add TRT verbose mode (#19100) Serge Panev 2020-09-10 13:12:44 -07:00
  • 6b01dc24ad Fix for duplicate subgraph inputs/outputs (#16131) Sam Skalicky 2020-09-10 10:06:12 -07:00
  • 45952e21a3 Make C Runtime API a C API again (#19085) Leonard Lausen 2020-09-09 09:23:41 -07:00
  • a9bd1d2f18 empty list cannot be cleared issue fixed. (#14882) Hafiz Shehbaz Ali 2020-09-09 08:21:51 +04:00
  • 5051fd9a50 Fix mxnet.test_utils.check_numeric_gradient documentation (#19060) VanessaSingh 2020-09-08 13:49:08 -04:00
  • 5b7a6d979b Re-enable the test_gpu_memory_profiler_gluon test case (#18704) Bojian Zheng 2020-09-08 03:49:08 -04:00
  • 0c0457a8f7 Add a contrib operator for dynamic reshape (#15872) Vandana Kannan 2020-09-08 00:06:55 -07:00
  • 8dbed966e3 Update base_module.py (#19096) Sheng Zha 2020-09-07 13:35:45 -07:00
  • 3ba9ffd8a1 [COMMUNITY] Add mention of mirroring of RFC issues to dev@ (#19092) Marco de Abreu 2020-09-07 12:57:50 +02:00
  • 40a2a524a3 Use chain.from_iterable in 2 modules (#18606) Ram Rachum 2020-09-07 10:41:51 +03:00
  • 5afb4b1e52 enable dnnl primitive cache (#17891) Tao Lv 2020-09-07 15:24:55 +08:00
  • 04e394aa8f update runtime setting default values (#18987) Sheng Zha 2020-09-06 19:39:20 -07:00
  • c1eaadaf8f Update requirements (#19061) Chaitanya Prakash Bapat 2020-09-06 15:57:05 -07:00
  • 7e97627f7c Fix typo in ISSUE_TEMPLATE config.yml (#19093) Marco de Abreu 2020-09-07 00:25:28 +02:00
  • 5a99105963 [DOC] enable edit on github (#19040) Sheng Zha 2020-09-06 13:00:18 -07:00
  • 62b7f030b7 Assure NDArray.reshape does not change the array size (#19078) r3stl355 2020-09-06 13:30:57 +01:00
  • 92a9f87608 Update license_header.py marcoabreu-patch-3 Marco de Abreu 2020-09-06 14:22:50 +02:00
  • fab0dd27ad Add issue template for website and documentation Marco de Abreu 2020-09-06 13:53:43 +02:00
  • 23b3665290 [COMMUNITY] display github discussions link in issues (#19090) Sheng Zha 2020-09-06 01:05:45 -07:00
  • af2b4bcecf [API] clean up gluon contrib (#19046) Sheng Zha 2020-09-05 11:11:38 -07:00
  • 748eebd747 fix block.export (#17970) (#19075) James Mracek 2020-09-03 09:21:50 -07:00
  • 363caacf6f Add build config for darwin native distribution (#19083) Sheng Zha 2020-09-03 09:20:27 -07:00
  • bb7e0cb29d Fix binary broadcast shape large tensor (#19070) Zhaoqi Zhu 2020-09-03 01:04:26 -07:00
  • b5e9c99849 [1.x] Backporting #19016 (#19069) Sam Skalicky 2020-09-02 20:40:54 -07:00
  • 1a398e5806 Support for fp16 in SpM x DnsM on GPU (#18930) (#19074) Rohit Kumar Srivastava 2020-09-02 18:47:01 -07:00
  • 45bfb83097 Update .htaccess (#19079) Leonard Lausen 2020-09-02 18:45:30 -07:00
  • f732530c8e [FEATURE]Enable large tensor support for batch flatten (#19063) Rohit Kumar Srivastava 2020-09-02 18:44:05 -07:00
  • ddfd9e486f Resolve minor docs issues (#19062) Aditya Mandke 2020-09-03 04:00:33 +05:30
  • 0e86a918cc [DOC][v1.7.x] Fix pip cu102link (#19054) Chaitanya Prakash Bapat 2020-09-02 10:47:25 -07:00
  • f2e90a22c7 Update NEWS, README and website for 1.7.0 (#19047) ciyong 2020-09-03 01:15:46 +08:00
  • 4cfb9e1164 Fix MNIST classification example (#19072) r3stl355 2020-09-02 18:13:08 +01:00
  • 5122d322c8 fix block.export (#17970) chinakook 2020-09-02 04:58:09 +08:00
  • 8379740012 MXNet Extensions enhancements2 (#19016) Sam Skalicky 2020-09-01 13:26:09 -07:00
  • d144edd46c 1.x: Stop packaging GPL libquadmath.so (#19055) Leonard Lausen 2020-09-01 11:03:17 -07:00
  • 9268f89080 Numpy pick large tensor fix (#19025) Zhaoqi Zhu 2020-09-01 09:01:08 -07:00
  • af467d2d70 Fix the error of gradient of np.pad (#19044) Wentao Xu 2020-09-01 10:09:58 +08:00
  • 9b22c8c2e9 Stop packaging GPL libquadmath.so (#19053) Leonard Lausen 2020-08-31 14:45:36 -07:00
  • de34763b94 set_lower_bound(1) so that stride is not zero (#18997) Aditya Mandke 2020-09-01 02:28:16 +05:30
  • ddb798a2a5 Numpy RNN operator large dim checks (#19033) Zhaoqi Zhu 2020-08-31 12:52:13 -07:00
  • 89e48d4a73 Numpy nonzero large tensor fix (#19015) Zhaoqi Zhu 2020-08-31 11:12:57 -07:00
  • 309b39a0d9 _slice_assign and _slice_assign_scalar large tensor fix (#19032) Zhaoqi Zhu 2020-08-31 10:20:50 -07:00
  • 13936020d4 [MXNET-1446] Quantization: intgemm matrix multiply wrappers (#17559) kpuatamazon 2020-08-31 18:10:47 +01:00
  • e2aacce02b Fix fusion of clip if a_min or a_max are not given (#19035) Przemyslaw Tredak 2020-08-30 22:08:18 -07:00
  • 74edcde1b0 add .asf.yaml for enabling wiki (#19042) Sheng Zha 2020-08-30 18:37:15 -07:00
  • ea3235ee80 switch pip doc to markdown (#19037) Sheng Zha 2020-08-30 13:17:37 -07:00
  • a3115d5529 [DOC] update contributors and readme (#19041) Sheng Zha 2020-08-30 13:17:21 -07:00
  • eeb6f34898 [CI] use fixture to reduce download of cifar10 (#19039) Sheng Zha 2020-08-30 12:43:57 -07:00
  • b0c39f7ea9 Update basic_layers.py (#19038) Sheng Zha 2020-08-29 19:10:46 -07:00
  • 48c21988cd Add size check for numpy ctc loss (#19028) Zhaoqi Zhu 2020-08-28 18:30:36 -07:00
  • 29c90ee204 Support for fp16 in SpM x DnsM on GPU (#18930) Rohit Kumar Srivastava 2020-08-28 14:53:11 -07:00
  • a9ca660038 [COMMUNITY] remove mentions of JIRA, update community docs (#18992) Sheng Zha 2020-08-27 18:10:00 -07:00
  • 6b596050e7 Numpy Pooling and ROI Pooling Large Dimension Checks (#19013) Zhaoqi Zhu 2020-08-27 08:05:30 -07:00
  • 998c6cea73 Numpy Gather ND Large Tensor fix (#18981) Zhaoqi Zhu 2020-08-27 07:21:20 -07:00
  • 8245fe2be7 Fix numpy equal broadcast (#19008) Zhaoqi Zhu 2020-08-25 18:19:05 -07:00
  • bce4cc69db [1.x] Backport of Fix LeakyRelu behaviour on empty input (#18934) (#19009) bgawrych 2020-08-26 03:18:31 +02:00
  • 49cc229be5 fix doc for SigmoidBinaryCrossEntropyLoss (#19010) Ziyue Huang 2020-08-26 09:17:55 +08:00
  • 566bbf74c9 Numpy Random Ops Fix & Testcases (#19004) Zhaoqi Zhu 2020-08-24 23:51:33 -07:00
  • b2c736fb12 [numpy] fix logical ops gpu build (#18983) Yiyan66 2020-08-25 13:42:07 +08:00
  • 8be953f80a Fix isnan usage in RTC (#18984) Przemyslaw Tredak 2020-08-24 19:55:20 -07:00
  • 3c4ac19daa Fix LeakyRelu behaviour on empty input (#18934) (#18996) bgawrych 2020-08-24 19:12:12 +02:00
  • dfefe87999 Get rid of monkey patching in LossScaler overflow handling (#18959) (#18973) mk-61 2020-08-24 10:00:52 -07:00
  • 0de7484884 clean up C API (#18969) Sheng Zha 2020-08-23 15:24:40 -07:00
  • 72eff9b66e Reducing the number of calls of the NDArray constructors (#18570) Andrei Ivanov 2020-08-22 15:51:13 -07:00
  • 46c0fb3483 Numpy Random Ops Large Tensor Fix (#18976) Zhaoqi Zhu 2020-08-21 22:18:10 -07:00
  • a847df78f9 Fixing tvmgpu issue & not restoring tvmop checks (#18818) Jinbo Ci 2020-08-22 08:46:04 +08:00
  • 4a7282f104 fix #18933 (#18974) Sheng Zha 2020-08-21 13:58:54 -07:00
  • 9bdd4d6347 fix #18927 (#18972) Sheng Zha 2020-08-20 17:01:10 -07:00
  • 20957ffcf1 Numpy Ops Large Tensor Tests Batch 2 (#18968) Zhaoqi Zhu 2020-08-20 16:04:36 -07:00
  • 29d6f273f3 Use RTC for elementwise and broadcast ops (#18622) Przemyslaw Tredak 2020-08-20 14:14:03 -07:00
  • bbc39fa6ef add signal handler for fpe, bus error (#18956) Sheng Zha 2020-08-20 07:42:08 -07:00
  • adcd9970d7 CI: Remove nccl installation after upstream fix (#18965) Leonard Lausen 2020-08-20 06:14:33 +00:00
  • 5e408bce1b add ffi (#18868) Yijun Chen 2020-08-20 13:42:33 +08:00
  • 9445a2dc03 Backport: Change Partition API's options_map to std::unordered_map #18929 (#18964) Serge Panev 2020-08-19 18:44:50 -07:00
  • 69ec3389f3 Numpy Ops Large Tensor Tests (#18932) Zhaoqi Zhu 2020-08-19 14:10:55 -07:00
  • 50312af58b turn on MXNET_SAFE_ACCUMULATION by default (#18961) Sheng Zha 2020-08-19 10:20:11 -07:00
  • 22d69ef32c Get rid of monkey patching in LossScaler overflow handling (#18959) mk-61 2020-08-19 10:07:28 -07:00