Commit Graph

  • 0c6f49fdbe Backport #17002, #17068 and #17114 to 1.6 branch (#17137) Przemyslaw Tredak 2019-12-20 14:07:21 -08:00
  • 8a3519934f Improve the speed of the pointwise fusion graph pass (#17114) Przemyslaw Tredak 2019-12-20 08:34:36 -08:00
  • 521c477ad3 [numpy] add op bitwise_not (#16947) Yiyan66 2019-12-19 14:42:09 +08:00
  • a7f33eb1e1 numpy bincount (#16965) Minghao Liu 2019-12-19 14:15:37 +08:00
  • a18250d57e Add silent option to quantization script (#17094) Zhennan Qin 2019-12-19 07:55:08 +08:00
  • ed09547170 [MXNET-978] Higher Order Gradient Support arcsin, arccos. (#15515) kshitij12345 2019-12-19 00:20:09 +05:30
  • faa283228d Add im2col and col2im operator (#16502) Wang Jiajun 2019-12-18 10:57:35 +08:00
  • a01ded2a04 FreeBSD comparibility patch (#17077) yuri@FreeBSD 2019-12-17 13:44:34 -08:00
  • 49fad87adf submodule update Iblis Lin 2019-12-17 13:01:31 +00:00
  • 4a2a9c9cce Merge branch 'ib/jl-cmake-build' of github.com:apache/incubator-mxnet into ib/jl-cmake-build Iblis Lin 2019-12-17 12:59:48 +00:00
  • 3f449aad15 backport Julia patches to v1.5.x (#16142) Iblis Lin 2019-12-17 20:57:15 +08:00
  • b28b847275 USE_JULIA_BLAS Iblis Lin 2019-09-14 13:59:29 +00:00
  • cede75b971 submodule update Iblis Lin 2019-09-14 21:32:02 +08:00
  • 46e7bd27db build with Julia's private libopenblas Iblis Lin 2019-09-14 13:16:11 +00:00
  • 27f996da89 julia: migrate build process to cmake Iblis Lin 2019-09-09 05:01:02 +00:00
  • f86a8d1007 [API] unified API for custom kvstores (#17010) Haibin Lin 2019-12-16 20:48:31 -08:00
  • 814be59534 Add #include <sys/wait.h> needed for waitpid (#17078) yuri@FreeBSD 2019-12-16 17:18:08 -08:00
  • 18c9a69e2f Install current OpenCV (#17090) Lukas Adamowicz 2019-12-16 19:09:58 -05:00
  • 4912465677 Update pypi_publish.py to disable nighlty build upload to Pypi (#17082) Leonard Lausen 2019-12-17 07:32:33 +09:00
  • db943e95e8 [BugFix] fix filter channel calculation in ModulatedDeformableConvV2 (#17070) Joshua Z. Zhang 2019-12-16 11:15:39 -08:00
  • 52c9a45abc [MKLDNN] enable MaxPooling with full pooling convention (#16860) Zhennan Qin 2019-12-16 16:05:32 +08:00
  • 897f4fae07 [MKLDNN] mkldnn RNN operator enhancement (#17075) Zixuan Wei 2019-12-16 16:03:12 +08:00
  • 843daf589b adding stacktrace in Jenkinsfile_utils.groovy to inspect Python2 failure cause in CI (#17065) Rohit Kumar Srivastava 2019-12-15 23:00:49 -08:00
  • 7dbb4b710c fix precision problem in linalg_solve, linalg_tensorinv, linalg_cholesky op test (#16981) dw_sjtu 2019-12-16 14:38:58 +08:00
  • 3d38dbde74 Adds min cuda version assertion decorator for unit tests (#17054) perdasilva 2019-12-14 22:47:42 +01:00
  • 831b548279 fix for number of inputs/outputs for backward custom ops (#17069) Sam Skalicky 2019-12-14 13:36:14 -08:00
  • bbdc1c3627 [reproducibility] multi_sum_sq review, AtomicAdd removal (#17002) MoisesHer 2019-12-14 08:32:50 -08:00
  • 696c54738f [BUGFIX] Fix trainer param order (#17068) Haibin Lin 2019-12-14 08:28:57 -08:00
  • 042682edca [DOC] Fix tutorial link, and better error msg (#17057) Haibin Lin 2019-12-14 08:28:25 -08:00
  • f045018f2c [MXNET-978] Higher Order Gradient Support logp1, expm1, square. (#15416) kshitij12345 2019-12-14 10:18:53 +05:30
  • 3b911cf5ed Website edits (#17050) Talia 2019-12-13 14:38:57 -08:00
  • c99a1108c0 Add my key to KEYS (#17060) Przemyslaw Tredak 2019-12-13 13:50:51 -08:00
  • ac6ed2c313 Add my key to KEYS (#17066) Przemyslaw Tredak 2019-12-13 11:55:11 -08:00
  • 61013a8bf9 use env var to control stack trace logging (#17038) Sheng Zha 2019-12-12 16:25:34 -08:00
  • f701f3f76a [MXNET-1431] Multiple channel support in Gluon PReLU (#16262) jonatanmil 2019-12-12 15:07:18 -08:00
  • 9092f17c90 array_split pr (#17032) Minghao Liu 2019-12-13 05:52:38 +08:00
  • c6755208f4 [BUGFIX] Fix race condition in kvstore.pushpull (#17007) (#17052) Przemyslaw Tredak 2019-12-11 21:11:57 -08:00
  • 5b9f79aa75 Omp fork numthreads fix 1.6 (#17000) Pedro Larroy 2019-12-11 18:34:19 -08:00
  • 634f95e243 swap wget to curl in Scala scripts (#17041) Lanking 2019-12-11 16:12:04 -08:00
  • 05af5c40bf [BUGFIX] Fix race condition in kvstore.pushpull (#17007) Haibin Lin 2019-12-11 15:18:09 -08:00
  • c7dab1704d Backport Gluon estimator changes to 1.6 (#17048) Przemyslaw Tredak 2019-12-11 13:57:33 -08:00
  • 04ebe45aed Prevent after-fork number of OMP threads being bigger than 1. (#16999) Pedro Larroy 2019-12-11 12:30:59 -08:00
  • c82af38211 Add support of plug and play fit_batch and evaluate_batch (#16982) liuzh91 2019-12-11 19:14:59 +08:00
  • 27389b1046 Fix Gluon Estimator nightly test (#17042) liuzh91 2019-12-11 13:30:10 +08:00
  • 9634786f96 mshadow: fix vector access (#17021) Tao Lv 2019-12-11 13:06:59 +08:00
  • c31ee9954f Fix omp assert issue (#17039) Chris Olivier 2019-12-10 20:55:31 -08:00
  • fe3439afca Workaround problem with fusion in CUDA 9 (#17028) (#17035) Przemyslaw Tredak 2019-12-10 12:54:33 -08:00
  • a576531836 Multi Precision Lamb Update operator (#16885) Rohit Kumar Srivastava 2019-12-09 16:52:02 -08:00
  • ff27b4b419 [OP] changing data type of 't' to int in lamb_update_phase1 (#16903) Rohit Kumar Srivastava 2019-12-06 16:17:28 -08:00
  • c7d484efd6 Lamb optimizer update (#16715) Rohit Kumar Srivastava 2019-11-23 22:19:07 -08:00
  • 7895f93e67 Replace mxnet_option macro with standard CMAKE_DEPENDENT_OPTION (#17018) Leonard Lausen 2019-12-11 02:52:20 +09:00
  • 0c17ddd619 Extend estimator.evaluate() to support event handlers (#16971) liuzh91 2019-12-11 01:50:22 +08:00
  • c973f01f95 Backport #16895, #16922, #16878, #16979 and #16900 to 1.6 (#17029) Przemyslaw Tredak 2019-12-10 09:26:26 -08:00
  • 60f77f5ab5 [CD] dynamic libmxet pipeline fix + small fixes (#16966) perdasilva 2019-12-10 16:44:58 +01:00
  • e18e4ce553 Include eval_net the validation model in the gluon estimator api (#16957) liuzh91 2019-12-10 14:12:15 +08:00
  • 9f5b8bca2e Workaround problem with fusion in CUDA 9 (#17028) Przemyslaw Tredak 2019-12-09 20:21:09 -08:00
  • bf3b4dcc3f Merge make/pip and make/maven configurations (#17027) Zach Kimberg 2019-12-09 20:23:47 -06:00
  • 8645b9a4a2 Fix CUDNN detection for CMake build (#17019) Leonard Lausen 2019-12-10 02:15:55 +00:00
  • 59535dbe7e Op_Diagonal [Numpy] (#16989) Minghao Liu 2019-12-10 09:52:39 +08:00
  • 248acfa655 Multi Precision Lamb Update operator (#16885) Rohit Kumar Srivastava 2019-12-09 16:52:02 -08:00
  • 986a90217b introduce gradient update handler to the base estimator (#16900) liuzh91 2019-12-10 07:13:24 +08:00
  • 1e1d774edf Revert downgrade cublas version as Nvidia-docker have fixed this (#17026) Jake Lee 2019-12-09 14:41:33 -08:00
  • 5f5b83f267 skip quantized conv flaky case (#16866) (#16774) alicia 2019-12-10 03:02:17 +08:00
  • 4b7262be31 set rtol = 1e-2 and atol = 1e-4 when dtype == np.float32 in test_numpy_op.py:test_np_linalg_solve (#17025) dw_sjtu 2019-12-10 03:01:11 +08:00
  • 21e320a985 [Bugfix] [Numpy] Add kAddTo and kNullOp to Transpose (#16979) Xingjian Shi 2019-12-09 10:41:28 -08:00
  • c573154e9d Link fixes5 (#16986) Talia 2019-12-09 10:20:03 -08:00
  • e48ff96ef4 Quantized Embedding (#16691) Xinyu Chen 2019-12-09 20:38:43 +08:00
  • 2d41f2a83d add inference benchmark script (#16978) Xinyu Chen 2019-12-09 20:37:29 +08:00
  • 538b18bb4e [MKL-DNN] Enable and Optimization for s8 eltwise_add (#16931) Xinyu Chen 2019-12-09 20:36:26 +08:00
  • ce97e22813 fix axis=-1 bug (#17016) Xi Wang 2019-12-09 16:53:41 +08:00
  • e40cceb271 Upgrade 3rdparty/openmp to release_90 version (#17012) Leonard Lausen 2019-12-09 08:19:28 +00:00
  • 44cd63e5a8 [Numpy] Implement numpy operator 'average' (#16720) Li Jun 2019-12-09 15:56:49 +08:00
  • 71b62725a0 use identity_with_cast (#16913) dw_sjtu 2019-12-09 15:54:21 +08:00
  • 7736bfdb78 [Numpy] add op full_like, c++ impl, fix zeros_like, ones_like type inference (#16804) alicia 2019-12-09 15:19:12 +08:00
  • 3b8fdacd16 skip quantized conv flaky case (#16866) Xinyu Chen 2019-12-09 13:57:52 +08:00
  • b009864774 large tensor faq doc fix (#16953) Rohit Kumar Srivastava 2019-12-08 21:30:06 -08:00
  • d58f6cb018 Add micro averaging strategy to pearsonr metric (#16878) zburning 2019-12-09 11:23:27 +08:00
  • 251e6f6cb7 Fix NDArrayIter cant pad when size is large (#17001) Jake Lee 2019-12-08 16:09:11 -08:00
  • 5cbf607cbb minor typo fix (#17008) Haibin Lin 2019-12-07 17:45:00 -08:00
  • 93cf48dfa4 Fix desired precision for test_ndarray.py:test_reduce (#16992) JackieWu 2019-12-08 07:59:28 +08:00
  • 083efb86e1 Fix crashing on Windows in ObjectPool ~ctor (#16941) dmitry-markeshov 2019-12-07 12:25:05 +03:00
  • e577c3d67d support np.argsort (#16949) Hao Jin 2019-12-06 22:14:55 -08:00
  • 82c34ee510 boolean_mask_assign with start_axis (#16886) Hao Jin 2019-12-06 22:13:46 -08:00
  • f06fdf6f79 [Gluon] Update contrib.Estimator LoggingHandler to support logging per batch interval (#16922) liuzh91 2019-12-07 13:05:53 +08:00
  • ca4939f26b [OP] changing data type of 't' to int in lamb_update_phase1 (#16903) Rohit Kumar Srivastava 2019-12-06 16:17:28 -08:00
  • fcc42de7e9 updating MXNet version to 1.6.0 in base.h for C APIs (#16905) Rohit Kumar Srivastava 2019-12-06 13:46:59 -08:00
  • 9c94fdbeb0 Fix NDArrayIter iteration bug when last_batch_handle='pad' (#16166) Jake Lee 2019-12-06 13:36:06 -08:00
  • 8f53e77fa8 mixed precision for power (#16859) Hao Jin 2019-12-06 12:57:51 -08:00
  • 61ef4a1657 Fix publishing scala gpu with cpu instance (#16987) Jake Lee 2019-12-06 11:46:33 -08:00
  • 5823ce64bf fix typo and doc (#16921) Huang, Guangtai 2019-12-06 17:12:18 +08:00
  • ae472c266d dynamic custom operator support (#15921) Sam Skalicky 2019-12-05 21:34:47 -08:00
  • 8dd70513af Downgrade the cublas version for nightly gpu (#16985) Jake Lee 2019-12-05 16:19:01 -08:00
  • 10903e8110 Fix ndarray indexing bug (#16895) reminisce 2019-12-05 14:02:46 -08:00
  • b5ee43e6fe add aligned roi introduced in Detectron2 (#16619) Jerry Zhang 2019-12-05 02:31:00 -08:00
  • 34e71163b1 [R package] Make R package compilation support opencv 4.0 (#16934) Hao.Chen 2019-12-05 18:27:52 +08:00
  • edb583b627 Fix scala publish & nvidia-docker cublas issue (#16968) Jake Lee 2019-12-04 22:22:01 -08:00
  • 02b4d2bddf fix for enable model parallelism for non-fp32 data (#16683) moin 2019-12-04 18:56:33 +01:00
  • 78a25236e4 [docs] Fix runtime feature detection documentation (#16746) Pedro Larroy 2019-12-02 10:53:52 -08:00
  • cc0c356500 squad multi-lamb benchmark Haibin Lin 2019-11-29 14:21:15 -08:00
  • 5fb29167a1 [MKLDNN] Use MKLDNNRun (#16772) Zhennan Qin 2019-11-29 13:24:06 +08:00