Commit Graph

  • e01c386038 min fix of 'storage_type' index mismatch (#14560) Songshan Gong 2019-06-08 14:04:12 +08:00
  • c5874ddf01 [BUGFIX] Fix a bug in Auto Function. (#15184) Da Zheng 2019-06-07 22:12:21 -07:00
  • 745a41ca1a [MXNET-1416] Fix inception inference example for potential index out of range error. (#15179) Amol Lele 2019-06-07 15:04:37 -07:00
  • bcff49888f reenable two unit tests (#15116) Jake Lee 2019-06-07 11:18:34 -07:00
  • 03374516c8 Python test failures under windows now report the exit code (#15147) david-seiler 2019-06-07 17:53:34 +02:00
  • eaacbafdd1 Fix Scala release (#15155) Zach Kimberg 2019-06-06 23:55:30 -07:00
  • 77da1e27a4 add cast_dtype option to load parameters (#15168) Haibin Lin 2019-06-06 22:20:24 -07:00
  • 6c00a5adaf update LICENSE (#15128) Lai Wei 2019-06-06 19:55:43 -07:00
  • 3f4f3d578c fix kvstore nightly failure (#15156) Lai Wei 2019-06-06 06:49:47 -07:00
  • b241c41d66 Backporting the fix to v1.4.x branch to prevent the crash in naive engine. (#15154) Amol Lele 2019-06-05 10:11:17 -07:00
  • c474f5f88a Update env_var.md (#15153) Haibin Lin 2019-06-05 04:56:21 -07:00
  • 910583ed7a fix misspell (#15149) Zhaoqi Zhu 2019-06-04 15:49:30 -07:00
  • 134a3e8cd3 fix nightly (#15141) Lai Wei 2019-06-04 07:43:49 -07:00
  • 28c528e16b [clojure] clojurify function names in image.clj namespace (#15121) Arthur Caillau 2019-06-04 15:37:04 +02:00
  • 75feadef0c [Flaky Test] Bump up bilinear_resize (#15129) Jake Lee 2019-06-03 23:03:56 -07:00
  • a37cd7abd5 [MXNET-1405] tests for large tensor support for Softmax operator (#15042) Rohit Kumar Srivastava 2019-06-03 21:32:24 -07:00
  • 3ebd873cd1 rat check make target (#15127) Sheng Zha 2019-06-03 11:55:42 -07:00
  • 9125f6ae8c remove warning in tutorial: (#15135) Lai Wei 2019-06-03 08:02:17 -07:00
  • 99e69e6825 [MXNET-1410]Adding Large Tensor Support for tensor transpose (#15059) Rohit Kumar Srivastava 2019-06-02 15:06:33 -07:00
  • 360f8d003d [BUGFIX] fix unknown parameter shapes when np_shape is turned on. (#15097) Da Zheng 2019-06-02 10:59:58 -07:00
  • 52f77c971e [clojure] fix: image test does not rely on s3 to run (#15122) Arthur Caillau 2019-06-02 18:20:49 +02:00
  • 6118dcc129 Add gluonCV to fix AMP Tutorial (#15039) Chance Bair 2019-06-02 08:04:15 +02:00
  • cdb7b7272f Fix mkldnn backend when using naive engine (#15089) Zhennan Qin 2019-06-01 18:40:10 -05:00
  • 75a90d0036 [MXNET-1408] Adding test to verify Large Tensor Support for ravel and unravel (#15048) Rohit Kumar Srivastava 2019-06-01 13:13:59 -07:00
  • e8a20fbb15 Enable serializing/deserializing ndarrays in np_shape semantics (#15090) reminisce 2019-06-01 11:50:46 -07:00
  • 866ec108ea Fix R installation instructions (windows / mac) (#15087) Aaron Markham 2019-05-31 13:54:00 -07:00
  • 6f01a6ff65 add epsilon for tolerance level (#15098) Jake Lee 2019-05-31 11:23:51 -07:00
  • 294a34a1da move amp test and change op support to warning (#15085) Sheng Zha 2019-05-30 21:30:09 -07:00
  • 31438583d7 fix gluon rnn cell single step unroll (#15081) Sheng Zha 2019-05-30 15:34:06 -07:00
  • 50495d7b94 NAG Optimizer with multi-precision support (#14568) Anirudh 2019-05-30 15:13:47 -07:00
  • 6b8e107f19 fix the if condition for LayerNorm (#15094) Tao Lv 2019-05-31 05:00:14 +08:00
  • e9a5a0a8f9 Change mx.test_utils.list_gpus to mx.context.num_gpus where possible (#14946) Pedro Larroy 2019-05-30 11:56:57 -07:00
  • 9c5b88f2c0 [MKLDNN]Fix mkldnn deconvolution forward with bias (#15088) Xinyu Chen 2019-05-30 07:51:47 -07:00
  • 65434886f6 Revert "Improve FC perf when no_bias=False (#15033)" (#15099) Lai Wei 2019-05-30 00:02:54 -07:00
  • 5fc4fc53df [MXNET] Updated http://data.dmlc.ml/ links to http://data.mxnet.io/ (#15065) IvyBazan 2019-05-28 18:20:06 -07:00
  • 8a9dd721bb [MXNET-978] Support higher order gradient for log. (#14992) kshitij12345 2019-05-28 23:12:21 +05:30
  • bbab5276be NER example: fix divisions by zero (#15068) William Tambellini 2019-05-27 23:09:44 -07:00
  • 6cf964af0d Improve FC perf when no_bias=False (#15033) Anirudh Subramanian 2019-05-26 18:07:42 -07:00
  • 01cf29d0aa Fix typo in amalgamation README.md (TREAD->THREAD) (#15058) Disi A 2019-05-26 10:08:03 -04:00
  • 7edd69eb93 Rename np_compat to np_shape (#15063) reminisce 2019-05-25 22:21:04 -07:00
  • f689042686 [MXNET-1382] Add the index_array operator (#14638) Nick Guletskii 2019-05-25 23:54:30 +03:00
  • 49ae15bf39 Restore save/load ndarray to 1.4.1 (#15073) reminisce 2019-05-25 12:56:37 -07:00
  • 136a5df3cd GPU RNN to use TempSpace resource for workspace. (#15056) Dick Carter 2019-05-24 22:13:39 -07:00
  • 9250a73d78 Revert the change broadcast_to param shape (#14998) reminisce 2019-05-24 21:06:25 -07:00
  • db2295b0f3 Allow clearing gpu cache (#14252) vlado 2019-05-24 19:38:40 -06:00
  • 653cbb4f14 MKLDNN RNN Inference Integration(fp32 LSTM and vRNN with tanh and relu) (#14713) Hao Li 2019-05-25 06:50:20 +08:00
  • d0ff3cdca6 [MXNET-545] Fix broken cython build (#10951) Deokjae Lee 2019-05-25 07:36:30 +09:00
  • eb0b8af430 Add cpu implementation for Deformable PSROIPooling (#14886) Wang Jiajun 2019-05-24 11:21:15 -05:00
  • 93fdcad2df [C++] fix type inconsistent issue when loading quantized parameters (#15038) Wuxun Zhang 2019-05-24 16:44:13 +08:00
  • 5763ba9706 Decreases test sensitivity (#15014) perdasilva 2019-05-24 05:20:57 +02:00
  • eade26f651 Add cpu implementation for Deformable Convolution (#14879) Wang Jiajun 2019-05-23 21:00:33 -05:00
  • 66aa983e4d Fix crash in random.shuffle operator (#15041) Lin Yuan 2019-05-23 13:02:15 -07:00
  • 503c7509db Fix warning / static function in header. (#14900) Pedro Larroy 2019-05-23 10:31:46 -07:00
  • d2e397a013 Add warning for fp16 inputs with MXNET_SAFE_ACCUMULATION=0 (#15046) Haibin Lin 2019-05-22 21:52:21 -07:00
  • 038b9fbffa Simplify creation of NodeEntry instances and use emplace_back (#14095) Pedro Larroy 2019-05-22 20:08:26 -07:00
  • d4e458eddc [Quantization] Support zero-size tensor input for quantization flow (#15031) ciyong 2019-05-23 09:38:57 +08:00
  • b0be6c50ae Integrating the MKL VML functions to MXNET to speed-up the (element-wised) mathematic computation (#14893) Shufan 2019-05-23 07:17:38 +08:00
  • 37f5315a26 Disables flaky test_random_size_crop (#15019) perdasilva 2019-05-23 01:13:58 +02:00
  • b3c91bfbbf Safe LayerNorm (#15002) Xingjian Shi 2019-05-23 02:56:23 +08:00
  • cca9c9badf use MXNET_SAFE_ACCUMULATION for softmax accumulator (#15037) Sheng Zha 2019-05-22 11:52:17 -07:00
  • 81b4ce2a5e fix link (#15036) Tao Lv 2019-05-22 20:38:14 +08:00
  • e5316b15b4 [MXNET-1401] adding more operators to test support for Large Tensor (#14944) Rohit Kumar Srivastava 2019-05-22 00:04:39 -07:00
  • 90b66b8f65 [MXNET-1406] [BUG] Fix DLManagedTensor deleter (#15016) Junru Shao 2019-05-21 18:14:12 -07:00
  • 5854b98c0a Fix infer shape partial after unknown shape changed to -1 (#14869) Lai Wei 2019-05-21 11:40:46 -07:00
  • aa55e3d6a8 Fixes lint issue in AMP (#15015) perdasilva 2019-05-21 11:31:14 +02:00
  • f680255fbf bump up atol for test_bilinear_resize_op (#15011) Hao Jin 2019-05-20 22:31:58 -07:00
  • 5bc08cec23 MXNet AMP (automatic mixed precision) (#14173) Przemyslaw Tredak 2019-05-20 20:21:26 -07:00
  • 789a6be4eb Add STL checks via -D_GLIBCXX_ASSERTIONS in debug mode (#14896) Pedro Larroy 2019-05-20 19:31:36 -07:00
  • 2cc822a52a [MKLDNN] Remove repeat parts in MKLDNN.md (#14995) PengZhao 2019-05-21 10:04:39 +08:00
  • 6dd2dda043 Disables flaky test_droupout (#15003) perdasilva 2019-05-21 03:37:03 +02:00
  • 91e81150bf Disables flaky test_l2_normalization (#15006) perdasilva 2019-05-21 03:36:49 +02:00
  • 7b48c24f57 Merge pull request #14887 from stu1130/publish_cuda10_1 Jake Lee 2019-05-20 18:35:15 -07:00
  • 2e827ab64d pylint fix stu1130 2019-05-19 14:46:49 -07:00
  • fbd9df1890 add upper bound stu1130 2019-05-18 12:57:26 -07:00
  • d9c6ed71c2 delete the unused python2 pylint stu1130 2019-05-08 14:04:46 -07:00
  • 657e699d7a update the dmlc-core commit to have lint.py use python3 stu1130 2019-05-08 10:22:44 -07:00
  • 95ffa2b4e0 add python2 pylint stu1130 2019-05-03 15:47:02 -07:00
  • 35d9c70ddf missing pylint upgrade stu1130 2019-05-02 10:40:43 -07:00
  • bbbb64a02f fix the typo stu1130 2019-05-01 17:14:44 -07:00
  • 2f50c33063 add missing numpy version update stu1130 2019-04-02 13:14:11 -07:00
  • 1e6a1ab964 update the numpy version stu1130 2019-04-01 16:47:14 -07:00
  • 8ae24bf77c LayerNorm acceleration on GPU (#14935) Xingjian Shi 2019-05-21 08:25:37 +08:00
  • 96dd059c6c Fix test randint (#14990) Anirudh Subramanian 2019-05-20 14:46:34 -07:00
  • 0ac48b6808 [Fit-API] Adress PR comments (#14885) Lai Wei 2019-05-14 15:24:38 -07:00
  • 780ad04393 [Fit API] update estimator (#14849) Lai Wei 2019-05-02 13:36:04 -07:00
  • 40ee94f156 [MXNET-1396][Fit-API] Update default handler logic (#14765) Lai Wei 2019-04-23 20:33:03 -07:00
  • 47b65d06cb [Fit API] improve event handlers (#14685) Lai Wei 2019-04-18 22:27:52 -07:00
  • 7ae7cb1fd7 move to gluon contrib (#14635) Lai Wei 2019-04-05 13:58:38 -07:00
  • ce2181471f move estimator to contrib (#14633) Lai Wei 2019-04-05 13:10:58 -07:00
  • d4f774479a [MXNET-1344, 1346][FIT API] Retrieve Batch size and Logging verbose support for Gluon fit() API (#14587) Karan Jariwala 2019-04-05 11:04:04 -07:00
  • ab7039edad [MXNet-1343][Fit API]Add CNN integration test for fit() API (#14405) Abhinav Sharma 2019-04-03 15:12:56 -07:00
  • c6afb73017 [MXNet-1375][Fit API]Added RNN integration test for fit() API (#14547) Karan Jariwala 2019-04-03 14:28:08 -07:00
  • 767e34d0de [MXNet-1340][Fit API]Update train stats (#14494) Lai Wei 2019-04-03 14:18:13 -07:00
  • faec0a9a3a [MXNet-1349][Fit API]Add validation support and unit tests for fit() API (#14442) Abhinav Sharma 2019-03-25 11:17:31 -07:00
  • e58a001ed1 Fixed issue where the estimator was printing beyond the dataset size … (#14464) Piyush Ghai 2019-03-20 18:44:00 -07:00
  • 03e059fd31 [MXNet-1334][Fit API]base class for estimator and eventhandler (#14346) Lai Wei 2019-03-15 22:16:42 -07:00
  • aac3cdb83d Revert "[MXNET-1333] Estimator and Fit API (#14629)" (#15008) Lai Wei 2019-05-20 11:31:09 -07:00
  • ace478f3c1 bump up cudnn to 7.5.1 & nccl 2.4.2 (#14988) Jake Lee 2019-05-20 10:54:53 -07:00
  • 807e40ae67 [MXNET-1402] MXNet docs change for 1.4.1 release (#14949) Junru Shao 2019-05-20 10:39:53 -07:00
  • 3cbfe48c27 Add matrix inversion operator in linalg (#14963) Wang Jiajun 2019-05-20 08:40:07 -05:00