Page History

Features

CUDA Graphs

• Enable CUDA Graphs for TRT (#19184)
• CUDA graphs support (#19142)
• Update cudnn version. (#19375)

CUDA 11 Support

• Update CUB and include it only for CUDA < 11 #18799' (#18975)
• Add new CI pipeline for building and testing with cuda 11.0. (#19149)
• Enable CUDA 11.0 on nightly development builds (#19314)

TensorRT

• TensorRT: add int8 with calibration (#19011)
• Add TRT verbose mode (#19100)
• Backporting TensorRT-Gluon Partition API (and TensorRT 7 support) (#18916)
• Backport TRT test update #19296 (#19298)

OneDNN

• Upgrade to oneDNN v1.6.3 (#19153) (#19161)
• Update oneDNN to official v1.6 release (#18867) (#18867)
• Upgrade to oneDNN v1.6 (#18822)
• bumped version to v1.6.5 (#19437)

IntGemm

• Backport of intgemm #17559 (#19099)
• Change intgemm to a submodule instead of fetch. (#19406)

Subgraph API

• Backport Fix for duplicate subgraph inputs/outputs (#16131) (#19112)

Extensions

• Backport #19103 (#19117)
• Backporting #19016 (#19069)
• Backport: Change Partition API's options_map to std::unordered_map #18929 (#18964)
• Backporting #18779 to v1.x (#18894)

ONNX

• Update onnx support to work with onnx 1.7.0 with most CV models (#19017)

Large Tensor

• Fix linalg_potri and linalg_potrf operators for large tensor. (#18752)
• Add forward, backward test for linalg.gemm2 (#18784)
• add large matrix tests for linalg ops: det, inverse, trsm, trmm (#18744)
• Add Large Tensor Test for linalg_syrk (#18782)
• Add Large Dim Checks for linalg Operators (#18816)
• Add forward & backward linalg.gemm test for large size (#18825)
• adding error message when attempting to use Large tensor with linalg_syevd (#18807)

Website Improvements

• v1.8 website patch (#19212)
• Automate website artifacts uploading (#19244)

Documentation

• Fix mxnet.test_utils.check_numeric_gradient documentation (#19060)
• Update windows_setup.md (#18874)

License

• Stop packaging GPL libquadmath.so (#19055)
• Remove mention of nightly in pypi (#18635) (#18884)
• Mkldnn header fix v1x for nightly binaries (#18797)
• Update LICENSE for all submodules. (#19440)
• LICENSE update (#19443)

CI Improvements

• Upgrade unix gpu toolchain (#18186) (#18785)
• Fix CI in v1.x branch (#18907)
• Remove extra --build-arg causing docker command to fail. (#19412)
• Fix CI builds failing due to invalid GPG keys. (#19377) (#19388)

Bug Fixes

Commits since 1.7.0 release:

51cc0af9f - GitHub, Fri Sep 25 18:03:55 2020 -0700 : added key for samskalicky (#19225)

...

• backport #19393 to v1.8.x (#19398)
• Fix SoftReLU fused operator numerical stability (#17849) (#19390)
• Temporary fix for RNN with oneDNN seg faults/core dumps (#19308)
• Fix MKLDNN BatchNorm with even number of channels (#19150) #19299 #19425 (#19428)
• Relaxing type requirements for broadcast_like (#17977) (#19448)
• Backporting: Fixed setting attributes in reviewSubgraph (#19278)

• Include oneDNN gemm fix (#19251)

• Fix for breaking change introduced in #17123 when batch_axis=0 (#19283)

• Backport PR #19272 to v1.8.x (#19273)

• Backport PRs in v1.7.x missing from v1.x to v1.8.x (#19262)

• delete executor before reallocating it memory (#19222)

...

• Nightly Large Tensor test cherrypicks (#19194) (#19215

...

• )

...

• Tweeking syntax to be closer to other tests (#19186) (#19206)

...

• ElementWiseSum fix for oneDNN (#18777) (#19200)

...

• Fix flaky intgemm test in v1.8.x too (#19204)

a35d568a0 - GitHub, Sat Sep 19 15:17:12 2020 -0700 : [1.x] Enable CUDA Graphs for TRT (#19184)

...

• Revert "Fix memory leaks in Gluon (#18328) (#18359)" (#19181

...

• )

...

• Improve environment variable handling in unittests (#18424) (#19173

...

• )

...

• Backport Unittest tolerance handling improvements (#18694). Also test seeding (#18762). (#19148)

620d058f7 - GitHub, Thu Sep 17 09:32:05 2020 -0700 : [v1.x] Add new CI pipeline for building and testing with cuda 11.0. (#19149)

...

• Fix the error of gradient of np.pad (#19044) (#19167)

...

• Backport Add cmake flag USE_FATBIN_COMPRESSION, ON by default (#19123) (#19158)

...

• SymbolBlock.imports ignore_extra & allow_missing (#19156

...

d2e645204 - GitHub, Wed Sep 16 09:41:35 2020 -0700 : [1.x] Backport of intgemm #17559 (#19099)

606933f2c - GitHub, Tue Sep 15 15:58:54 2020 -0700 : TensorRT: add int8 with calibration (#19011)

9dfac79d5 - GitHub, Mon Sep 14 22:57:05 2020 -0700 : [1.x] Backport Fix for duplicate subgraph inputs/outputs (#16131) (#19112)

e1bcb336f - GitHub, Mon Sep 14 18:12:38 2020 -0700 : [1.x] Backport 'Update CUB and include it only for CUDA < 11 #18799' (#18975)

...

• )

...

• Fix race condition in NaiveEngine::PushAsync (#19108) (#19122

...

• )

2d077db1c - GitHub, Thu Sep 10 13:12:44 2020 -0700 : Add TRT verbose mode (#19100)

...

• empty list cannot be cleared issue fixed. (#14882

...

• )

...

• Update base_module.py (#19096)

...

• fix block.export (#17970) (#19075)

b5e9c9984 - GitHub, Wed Sep 2 20:40:54 2020 -0700 : [1.x] Backporting #19016 (#19069)

...

• Support for fp16 in SpM x DnsM on GPU (#18930) (#19074

...

• )

...

• Backport of Fix LeakyRelu behaviour on empty input (#18934) (#19009)

...

• Get rid of monkey patching in LossScaler overflow handling (#18959) (#18973

...

9445a2dc0 - GitHub, Wed Aug 19 18:44:50 2020 -0700 : Backport: Change Partition API's options_map to std::unordered_map #18929 (#18964)

...

• )

...

...

• remove upper bound (#18857) (#18910

...

• )

9981e847f - GitHub, Tue Aug 18 08:32:35 2020 -0700 : [CI][1.x] Cherrypick: Upgrade unix gpu toolchain (#18186) (#18785)

...

• fix gelu to use erf based algorithm (#18827) (#18946)

...

• Cherry-pick #18635 to v1.7.x (#18935) (#18945)

...

• Backporting backward inference from 2.x #18348 and #18378 (#18895)

...

• backport Invoke mkldnn and cudnn BatchNorm when axis != 1 to v1.7.x (#18676) (#18890)

...

• bump version to 1.8.0 (#18899)

9fbf3d310 - GitHub, Wed Aug 12 10:12:06 2020 -0700 : Fix CI in v1.x branch (#18907)

171110371 - GitHub, Tue Aug 11 21:32:58 2020 +0800 : [1.x][submodule] Update oneDNN to official v1.6 release (#18867) (#18867)

0b5b44948 - GitHub, Mon Aug 10 12:19:04 2020 -0700 : Remove mention of nightly in pypi (#18635) (#18884)

cc287a0fc - GitHub, Fri Aug 7 22:09:24 2020 -0700 : Update windows_setup.md (#18874)

eae6171cb - GitHub, Tue Aug 4 12:26:46 2020 -0700 : [submodule] Upgrade to oneDNN v1.6 (#18822)

2aa270288 - GitHub, Mon Aug 3 12:33:39 2020 -0700 : [v1.x] Mkldnn header fix v1x for nightly binaries (#18797)

...

• Fixing ONNX spatial export for batchnorm (#17711) (#18846)

...

• Fix softmax, logsoftmax failed on empty ndarray (#18602) (#18708)

...

• Add unit tests for potri and potrf backward and check output shape in unit tests. (#18803

...

f4e62dfe9 - GitHub, Thu Jul 30 22:49:36 2020 -0700 : Add Large Dim Checks for linalg Operators (#18816)

84c9e0d46 - GitHub, Thu Jul 30 10:40:32 2020 -0700 : [v1.x][LT] Add forward & backward linalg.gemm test for large size (#18825)

...

• )

...

• Add syrk test shape check (#18812)

...

• Back port optimization to broadcast_axis to MXNet1.x (#18773)

d0093458e - GitHub, Mon Jul 27 20:28:43 2020 -0700 : [1.x][LT] Add forward, backward test for linalg.gemm2 (#18784)

566d9d348 - GitHub, Mon Jul 27 10:04:30 2020 -0700 : [v1.x] add large matrix tests for linalg ops: det, inverse, trsm, trmm (#18744)

85ff00d5c - GitHub, Fri Jul 24 15:55:05 2020 -0700 : Add Large Tensor Test for linalg_syrk (#18782)

...

• Fix

...

• crash when accessing already destructed static variables (#18768) (#18778)

...

• Cherrypick #18677 #18713 (#18742)