)]}'
{
  "log": [
    {
      "commit": "b84609d3fc73d20929c114eab95faaa56e6c5ede",
      "tree": "4093e85bafe43a0f658d7910f87f85b19d3affcb",
      "parents": [
        "9f013ddad0877923f34d2d152c283fbdf7a2dc06"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Thu Jan 26 13:28:45 2023 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jan 26 13:28:45 2023 -0800"
      },
      "message": "Bump tzinfo from 1.2.6 to 1.2.10 in /docs/static_site/src (#21139)\n\nBumps [tzinfo](https://github.com/tzinfo/tzinfo) from 1.2.6 to 1.2.10.\r\n- [Release notes](https://github.com/tzinfo/tzinfo/releases)\r\n- [Changelog](https://github.com/tzinfo/tzinfo/blob/master/CHANGES.md)\r\n- [Commits](https://github.com/tzinfo/tzinfo/compare/v1.2.6...v1.2.10)\r\n\r\n---\r\nupdated-dependencies:\r\n- dependency-name: tzinfo\r\n  dependency-type: direct:production\r\n...\r\n\r\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\r\n\r\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\r\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "9f013ddad0877923f34d2d152c283fbdf7a2dc06",
      "tree": "10323a2af519ffa4c7455dd1d0df91bc9bdd4ead",
      "parents": [
        "48d7f4af70015a559b20bc58aea037548c7e2658"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Thu Jan 12 07:31:22 2023 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jan 12 07:31:22 2023 -0800"
      },
      "message": "Upgrade boto3 to latest version. (#21167)\n\n* Upgrade boto3 to latest version.\r\n\r\n* Change link_check workflow to use ubuntu 20.04 instead of latest, until we can upgrade toolchain."
    },
    {
      "commit": "48d7f4af70015a559b20bc58aea037548c7e2658",
      "tree": "8dfbca2f386e3e9ff14044311f5c32a4fb17b5c4",
      "parents": [
        "7d602e3b2382eb501fdeb94c4d97e652a723af11"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Wed Jan 04 04:09:23 2023 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jan 04 07:09:23 2023 -0500"
      },
      "message": "Port top-level-project updates from v1.x branch (#21162)\n\n"
    },
    {
      "commit": "7d602e3b2382eb501fdeb94c4d97e652a723af11",
      "tree": "cb5c1461a7da2242164a505f2337a3c12c369882",
      "parents": [
        "c8922fedff48cf0c4f15bf0e5fe805be8be34512"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Mon Sep 26 14:15:06 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Sep 26 14:15:06 2022 +0200"
      },
      "message": "[DOC] Fix the table in Improving accuracy with INC (#21140)\n\nFix formating of  the table in https://mxnet.apache.org/versions/master/api/python/docs/tutorials/performance/backend/dnnl/dnnl_quantization_inc.html"
    },
    {
      "commit": "c8922fedff48cf0c4f15bf0e5fe805be8be34512",
      "tree": "0debe21e7284628feb83c09d4f2131562c3ac8d5",
      "parents": [
        "bd6405b787c5153856232b49b552b8e98127390e"
      ],
      "author": {
        "name": "hankaj",
        "email": "hanna.jarlaczynska@intel.com",
        "time": "Fri Sep 16 13:01:27 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Sep 16 13:01:27 2022 +0200"
      },
      "message": "Python string formatting (#21136)\n\n* Improve string formatting\r\n\r\n* Another formatting fix\r\n\r\n* Minor fixes\r\n\r\n* Review suggestion\r\n\r\n* Review suggestions\r\n\r\n* Review suggestions once again\r\n\r\n* Review additional suggestions\r\n\r\n* A few more fixes\r\n\r\n* Remove f-strings from logging functions\r\n\r\n* Fix lack of \u0027f\u0027\r\n\r\n* Minor fixes\r\n\r\n* Fix lint errors"
    },
    {
      "commit": "bd6405b787c5153856232b49b552b8e98127390e",
      "tree": "086ffde5c14d889559160b4bd922ef37d8d3ce47",
      "parents": [
        "f803641b53f5ad9b14fc498da64f62161256df87"
      ],
      "author": {
        "name": "hankaj",
        "email": "hanna.jarlaczynska@intel.com",
        "time": "Thu Sep 15 18:13:20 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Sep 15 18:13:20 2022 +0200"
      },
      "message": "Add quantized batch norm operator fused with ReLU (#21137)\n\n* Delete fuse_norm_relu flag\r\n\r\n* Refactor BN operator\r\n\r\n* Add quantized bn relu\r\n\r\n* Fix error\r\n\r\n* Small fix\r\n\r\n* Fix issues after rebase\r\n\r\n* Change output size"
    },
    {
      "commit": "f803641b53f5ad9b14fc498da64f62161256df87",
      "tree": "051969b6f6949190e5c12b2b6dd6ae55fe55e005",
      "parents": [
        "3a19f0e50d75fedb05eb558a9c835726b57df4cf"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Thu Sep 08 10:04:27 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Sep 08 10:04:27 2022 +0200"
      },
      "message": "[DOC] Add custom strategy script to quantization with INC example (#21134)\n\n* [DOC] Add custom strategy script to quantization with INC example\r\n\r\nUpdate results to measured on latest MxNet build\r\n(2.0.0b20220902 (commit 3a19f0e50d75fedb05eb558a9c835726b57df4cf))\r\n\r\n* Fix script name"
    },
    {
      "commit": "3a19f0e50d75fedb05eb558a9c835726b57df4cf",
      "tree": "d7a6e6768fe15cba6ee7a8121de4028cfd39a4d4",
      "parents": [
        "8d933fdcdbcc3b8b0ba04a0e584778e62e988f08"
      ],
      "author": {
        "name": "Kacper Pietkun",
        "email": "kacper.pietkun@intel.com",
        "time": "Wed Aug 31 09:01:31 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Aug 31 09:01:31 2022 +0200"
      },
      "message": "[FEATURE] Dnnl sum primitive path (#21132)\n\n* Added dnnl_sum primitive path to mxnet binary_add when shapes are the same\r\n\r\n* added test coverage\r\n\r\n* added operation check\r\n\r\n* Random number for tests\r\n\r\n* delete unnecessary variables\r\n\r\n* review changes"
    },
    {
      "commit": "8d933fdcdbcc3b8b0ba04a0e584778e62e988f08",
      "tree": "c0c603fd6135483d4161e3e5bda260c57eec2d88",
      "parents": [
        "2d72ce465a42d47d28ee9a090cc9cae8b54b0301"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Mon Aug 29 11:32:07 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 29 11:32:07 2022 +0200"
      },
      "message": "Add proper link to scripts in quantization with INC example (#21133)\n\n"
    },
    {
      "commit": "2d72ce465a42d47d28ee9a090cc9cae8b54b0301",
      "tree": "3e7fb6cac0d0882d83a5d7f9fc5b138f2788748e",
      "parents": [
        "1a418e4e1cf89a87838eb02abf74ccfa4bfe6c37"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Fri Aug 26 13:10:50 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 26 13:10:50 2022 +0200"
      },
      "message": "[DOC] Add tutotrial about improving accuracy of quantization with oneDNN (#21127)\n\n* Add tutotrial about improving accuracy of quantization with oneDNN\r\n\r\n* Fix link to mse strategy description\r\n\r\n* Temporary remove links to files from the example\r\n\r\n* Apply suggestions from code review\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e"
    },
    {
      "commit": "1a418e4e1cf89a87838eb02abf74ccfa4bfe6c37",
      "tree": "aebd37f3e26f7b73dee0ab963ac5aed39a678381",
      "parents": [
        "daac02c7854ffa71bc11fd950c2d6c9ea356b394"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Tue Aug 23 15:33:02 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 23 15:33:02 2022 +0200"
      },
      "message": "[FEATURE] Add query_keys transformer version without split (#21115)\n\n* Add query_keys transformer version without split\r\n\r\n* Add op to amp, implement different shaped inputs cases\r\n\r\n* Fix sanity\r\n\r\n* Remove semicolon\r\n\r\n* Fix windows build\r\n\r\n* Fix windows build sanity\r\n\r\n* Change to bool template, fix comments, shorten test case\r\n\r\n* Change comment message\r\n\r\n* Fix SelectInput(), input names and Forward()"
    },
    {
      "commit": "daac02c7854ffa71bc11fd950c2d6c9ea356b394",
      "tree": "9563a03b04fefe1322de6ffc050046f3c55f2d38",
      "parents": [
        "7748ae7edf4d48908b70a98c0f3ea24e4853fd61"
      ],
      "author": {
        "name": "hankaj",
        "email": "hanna.jarlaczynska@intel.com",
        "time": "Mon Aug 22 13:53:44 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 22 13:53:44 2022 +0200"
      },
      "message": "Fix fused resnet low accuracy (#21122)\n\n* Change flag to postop\r\n\r\n* Add attributes to batch norm relu node\r\n\r\n* Refactor code with batch norm relu op\r\n\r\n* Delete fuse_norm_relu flag\r\n\r\n* Delete fuse_norm_relu flag\r\n\r\n* Refactor BN operator\r\n\r\n* Review suggestions\r\n\r\n* Review suggestions once again\r\n\r\n* Fix formatting\r\n\r\n* Fix lint errors"
    },
    {
      "commit": "7748ae7edf4d48908b70a98c0f3ea24e4853fd61",
      "tree": "7673886bad2015bc04d164261d6931a2b809ec23",
      "parents": [
        "1058369f8a3eea3147a727ac98fadfcb2ffe5d5a"
      ],
      "author": {
        "name": "Tim Gates",
        "email": "tim.gates@iress.com",
        "time": "Wed Aug 17 04:43:51 2022 +1000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 16 11:43:51 2022 -0700"
      },
      "message": "docs: Fix a few typos (#21094)\n\nThere are small typos in:\r\n- docs/python_docs/python/tutorials/getting-started/crash-course/7-use-gpus.md\r\n- docs/static_site/src/pages/api/cpp/docs/tutorials/multi_threaded_inference.md\r\n- example/README.md\r\n\r\nFixes:\r\n- Should read `specifying` rather than `specifieing`.\r\n- Should read `multi threaded` rather than `iultithreaded`.\r\n- Should read `provisioning` rather than `provisionning`."
    },
    {
      "commit": "1058369f8a3eea3147a727ac98fadfcb2ffe5d5a",
      "tree": "a5083177575b967bce64ce8103f06ebb5879842f",
      "parents": [
        "736313f4e7e35563418df2285ff521242324ed1a"
      ],
      "author": {
        "name": "Kacper Pietkun",
        "email": "kacper.pietkun@intel.com",
        "time": "Fri Aug 12 09:31:54 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 12 09:31:54 2022 +0200"
      },
      "message": "[BUGFIX] _npi_repeats with swap (#21112)\n\n* fixed _npi_repeats_swap bug\r\n\r\n* review changes\r\n\r\n* Throw error with wrong repeats parameter (mx.np.repeat)\r\n\r\n* Shape check moved to FInferShape\r\n\r\n* Simplify type size check\r\n\r\n* code refactoring\r\n\r\n* Revert \"code refactoring\"\r\n\r\nThis reverts commit 18a42e501b837397110fa7ed8a543ba7aa1c7829.\r\n\r\n* fix for gpu version\r\n\r\n* review_changes_2\r\n\r\n* Revert \"review_changes_2\"\r\n\r\nThis reverts commit 789ad1a626980660876608884b8cc084e53a890a.\r\n\r\n* review_3\r\n\r\n* clang formatting"
    },
    {
      "commit": "736313f4e7e35563418df2285ff521242324ed1a",
      "tree": "1d9bfc80f5ac0735e9f37b1281d4fa72afbdd837",
      "parents": [
        "6d1fbe35d220a7789f5f79eac8bd55716d7d2492"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Fri Aug 12 09:25:31 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 12 09:25:31 2022 +0200"
      },
      "message": "Add support for bool data type for condition in where operator (#21103)\n\n* Add support for bool data type for condition in where operator\r\n\r\n* Update src/operator/nn/dnnl/dnnl_where.cc\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n\r\n* Update src/operator/nn/dnnl/dnnl_where.cc\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n\r\n* apply review comment\r\n"
    },
    {
      "commit": "6d1fbe35d220a7789f5f79eac8bd55716d7d2492",
      "tree": "bd7cba8cfe061ac795ba4ec605eb97f40a8fe506",
      "parents": [
        "9975ab41a6ecbc8338213315405c780c6444cb99"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Thu Aug 04 17:08:23 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Aug 04 17:08:23 2022 +0200"
      },
      "message": "Add size threshold for few oneDNN operators (#21106)\n\n* Add size threshold for few oneDNN operators\r\n\r\n* add const\r\n\r\n* Add PR number\r\n\r\n* Apply suggestions from code review\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n\r\n* Refactor binary op\r\n\r\n* Fix true_divide\r\n\r\nCo-authored-by: Bartlomiej Gawrych \u003cbarlomiej.gawrych@intel.com\u003e\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e"
    },
    {
      "commit": "9975ab41a6ecbc8338213315405c780c6444cb99",
      "tree": "72f54fee0fbd6b2c44b89e55f78bcdf8d0522728",
      "parents": [
        "97e25cfc7a723a42c007eae90736ac1faa6c4c98"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Thu Aug 04 01:41:47 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Aug 04 01:41:47 2022 -0700"
      },
      "message": "[BUGFIX] Reenable fwd conv engine 5 on test_group_conv2d_16c (#21104)\n\n* Reenable fwd conv engine 5 on test_group_conv2d_16c\r\n\r\n* Test conv plan fix devel\r\n\r\n* Fix clang format\r\n\r\n* More clang format fixes\r\n\r\n* switch to using std::vector::data()"
    },
    {
      "commit": "97e25cfc7a723a42c007eae90736ac1faa6c4c98",
      "tree": "ceb54ecfd502bf470bfc5d7fe443f06345a7903b",
      "parents": [
        "0b4ecdbc4a35e50e097c6e171fcf0d78d46df007"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Aug 04 08:46:28 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Aug 04 08:46:28 2022 +0200"
      },
      "message": "[submodule] Upgrade oneDNN to v2.6.1 (#21108)\n\n"
    },
    {
      "commit": "0b4ecdbc4a35e50e097c6e171fcf0d78d46df007",
      "tree": "d4a900a9dc9d9afb32255ac7391efbd7d72eef85",
      "parents": [
        "dedb8c97af040c46bda3890e55472c70f16d237e"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Tue Aug 02 02:59:57 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 02 02:59:57 2022 -0700"
      },
      "message": "[BUGFIX] Fix threadsafety and shutdown issues with threaded_engine_perdevice (#21110)\n\n* Fix threadsafety and shutdown issues with threaded_engine_perdevice\r\n\r\n* Fix lint\r\n\r\n* Add MXNET_USE_CUDA compile guards\r\n\r\n* Remove unneeded include"
    },
    {
      "commit": "dedb8c97af040c46bda3890e55472c70f16d237e",
      "tree": "4fa3636640ed4234144e3f7914293ffb5dd9a8e2",
      "parents": [
        "db39bb1126dee62e44ca0bd1121695a77e2873d2"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Mon Aug 01 10:21:45 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 01 13:21:45 2022 -0400"
      },
      "message": "[WIP] [BUGFIX] Fix flakey TemporaryDirectory() cleanup on Windows (#21107)\n\n* Trigger CI to show windows-gpu job failures\r\n\r\n* Fix TemporaryDirectory cleanup issues on Windows\r\n\r\n* Fix PermissionError handling\r\n\r\n* Remove temporary comments\r\n\r\n* Fix TemporaryDirectory() to yield dir.name\r\n\r\n* Remove temp dirs even with exceptions"
    },
    {
      "commit": "db39bb1126dee62e44ca0bd1121695a77e2873d2",
      "tree": "7de356fa9741573a308e438137ff1a85727a85b9",
      "parents": [
        "5e5e0e3fc153acd53b4eb28284d4d372340d41ca"
      ],
      "author": {
        "name": "hankaj",
        "email": "hanna.jarlaczynska@intel.com",
        "time": "Fri Jul 29 13:06:00 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 29 13:06:00 2022 +0200"
      },
      "message": "Fix multivariate normal bug (#21105)\n\n"
    },
    {
      "commit": "5e5e0e3fc153acd53b4eb28284d4d372340d41ca",
      "tree": "5e31419e7b6e7b71c1c3735a00bde39d382a0727",
      "parents": [
        "ecb50261163c5d4afe307d6e1224530151d96c22"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Fri Jul 22 14:43:37 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 22 14:43:37 2022 +0200"
      },
      "message": "[BUGFIX] Fix SupportDNNL for multiple inputs (#21102)\n\n* Fix SupportDNNL for multiple inputs\r\n\r\n* Fix SupportDNNL condition for multiple inputs"
    },
    {
      "commit": "ecb50261163c5d4afe307d6e1224530151d96c22",
      "tree": "ce6d97339f875b18ee676026b8a94b647b1fa523",
      "parents": [
        "183e012f012fbf8c3799094a3365686bd19f50e3"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Thu Jul 21 09:43:23 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 21 09:43:23 2022 +0200"
      },
      "message": "Comments formatting fix (#21101)\n\n* Fix comments formating\r\n\r\n* Review suggestion\r\n\r\n* Fix file name"
    },
    {
      "commit": "183e012f012fbf8c3799094a3365686bd19f50e3",
      "tree": "dd88bd8cb16ce2d109e43a367ee49b484f733e2c",
      "parents": [
        "7b1daf9bc30ae92c0581b26449e00d2ae5882230"
      ],
      "author": {
        "name": "hankaj",
        "email": "hanna.jarlaczynska@gmail.com",
        "time": "Tue Jul 19 17:00:19 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 19 17:00:19 2022 +0200"
      },
      "message": "Get rid of warnings (#21099)\n\n* Get rid of warnings\r\n\r\n* Fix clang formatting\r\n\r\n* Fix problem with redefinition\r\n\r\n* Fix \u0027uninitialized\u0027 warning\r\n\r\n* Update CONTRIBUTORS.md"
    },
    {
      "commit": "7b1daf9bc30ae92c0581b26449e00d2ae5882230",
      "tree": "d703f6b04c6ca56d3c8a07936bd987af92675f8d",
      "parents": [
        "cca8f4e8c6f4d1a6fa824fbb3e7b0415de468de0"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Tue Jul 19 16:44:52 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 19 16:44:52 2022 +0200"
      },
      "message": "Requantize scale fix (#21100)\n\n* Requantize scale fix and refactor\r\n\r\n* Comment fix\r\n\r\n* Review suggestions\r\n\r\n* Formatting fix"
    },
    {
      "commit": "cca8f4e8c6f4d1a6fa824fbb3e7b0415de468de0",
      "tree": "0bdb308732789ede93e30a2e54a0055bb8dccbc2",
      "parents": [
        "ef0415d6452160370cddfe654bb0abeaf53af894"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Tue Jul 19 09:51:42 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 19 09:51:42 2022 +0200"
      },
      "message": "Reduce overhead in sg_onednn_fully_connected for floats (#21092)\n\n* dispatch float fc to simpler function\r\n\r\n* apply review comments\r\n\r\n* remove unnecesarry variables\r\n\r\n* fix sanity\r\n"
    },
    {
      "commit": "ef0415d6452160370cddfe654bb0abeaf53af894",
      "tree": "c0fcb4b5f3dff954b832c7d7abaa1bbf119f796e",
      "parents": [
        "cf15e0a478ea45166b2f397ff440e30601ea2ec4"
      ],
      "author": {
        "name": "Kacper Pietkun",
        "email": "kacper.pietkun00@gmail.com",
        "time": "Mon Jul 18 15:50:14 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jul 18 15:50:14 2022 +0200"
      },
      "message": "[BUGFIX] Fix floor divide (#21096)\n\n* Fixed floor_divide bug on float16\r\n\r\n* Made separate version for half_t with division made on float\r\n\r\n* Update Contributors.md"
    },
    {
      "commit": "cf15e0a478ea45166b2f397ff440e30601ea2ec4",
      "tree": "c647eca67f57674c54d8285e4ea453aa481ad813",
      "parents": [
        "f6d1ed1872b745c3a9dd944b7ad5c2ce988b014b"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Fri Jul 15 13:47:18 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 15 13:47:18 2022 +0200"
      },
      "message": "[BUGFIX] Fix remove Cast fuse (#21086)\n\n* Fix remove Cast fuse\r\n\r\n* Fix Reset()\r\n\r\n* Fix review\r\n\r\n* Fix sanity"
    },
    {
      "commit": "f6d1ed1872b745c3a9dd944b7ad5c2ce988b014b",
      "tree": "d90469c092ebf0ba07cd7a0a45df5ee95c991fa2",
      "parents": [
        "ded609612623fedcfb5fcc0366c5ee524b4a989b"
      ],
      "author": {
        "name": "Paweł Głomski",
        "email": "pawel.glomski@intel.com",
        "time": "Fri Jul 15 10:56:38 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 15 10:56:38 2022 +0200"
      },
      "message": "Improve bf16 support (#21002)\n\n* AMP improvements + enable bf16 input for quantize_v2\r\n\r\n* Fix sanity\r\n\r\n* Improve tests, AMP conversion interface, fix forwad hooks\r\n\r\n* Fix tests\r\n\r\n* Fix imports in tests\r\n\r\n* Use different lp16_fp32 op in test\r\n\r\n* Add amp.disable_amp() context, fix tests\r\n\r\n* Add tests, generalize optimization disabling\r\n\r\n* Fix sanity\r\n\r\n* Review fixes\r\n\r\n* Use is_integral\u003c\u003e::value\r\n\r\n* Review fixes\r\nChange flag type to unsigned int\r\nAdd a warning for an incorrect flag attribute value\r\n\r\n* Extend bf16 support\r\n\r\n* Combine enable_float_output and amp_out_dtype parameters\r\n\r\n* Add bf16 support to _dnnl_batch_dot\r\n\r\n* Fix sanity\r\n\r\n* Add bf16 support to all dnnl ops, add tests\r\n\r\n* Add license\r\n\r\n* Fix conv activation fuse, disable masked_softmax bf16 support\r\n\r\n* Fix sanity, add softmax test cases\r\n\r\n* Compare bf16 outputs with fp32 reference\r\n\r\nCo-authored-by: Bartlomiej Gawrych \u003cbarlomiej.gawrych@intel.com\u003e"
    },
    {
      "commit": "ded609612623fedcfb5fcc0366c5ee524b4a989b",
      "tree": "149e127a24a297533e53b42eb42efa18597c4f06",
      "parents": [
        "e522bea5135f61761a82426d01d9778054d8ae77"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Fri Jul 15 09:57:57 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 15 09:57:57 2022 +0200"
      },
      "message": "[FEATURE] Add pytest with benchmarking operator (#21088)\n\n* Add benchmarking operator pytest using benchmark_utils\r\n\r\n* Unify quote mark use in benchmark"
    },
    {
      "commit": "e522bea5135f61761a82426d01d9778054d8ae77",
      "tree": "4517af17c283c7ded54361516784d4db8c076717",
      "parents": [
        "e36c9f075aa7ee14e1a8ccf245df5d1e1648515b"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Fri Jul 15 09:50:44 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jul 15 09:50:44 2022 +0200"
      },
      "message": "[BUGFIX] Fix Gluon2.0 guide (#21090)\n\n* Fix Gluon2.0 guide\r\n\r\n* Update docs/python_docs/python/tutorials/getting-started/gluon_migration_guide.md\r\n\r\nCo-authored-by: bgawrych \u003cbartlomiej.gawrych@intel.com\u003e\r\n\r\n* Fix NumPy mentions\r\n"
    },
    {
      "commit": "e36c9f075aa7ee14e1a8ccf245df5d1e1648515b",
      "tree": "71fbd261bf46578c0ba34795aa42f0c7ee3e857f",
      "parents": [
        "26243eea86ffa54dab9463e8babd375fa2fcb463"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Jul 07 16:35:29 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 07 16:35:29 2022 +0200"
      },
      "message": "Refactor fc_sum_fuse (#21077)\n\n* Refactor fc_sum_fuse\r\n\r\n* Fix sanity\r\n\r\n* Simplify Selector\r\n\r\n* Restore ConnectSubgraphOutputs in fc_sum_fuse_property\r\n\r\n* Fix node name"
    },
    {
      "commit": "26243eea86ffa54dab9463e8babd375fa2fcb463",
      "tree": "c13b3055bca18d45a69766ec1b3db7cc5b44b00e",
      "parents": [
        "84b1626b66f84a35e87a95d26226c3a3171ea78c"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Jul 07 14:49:46 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 07 14:49:46 2022 +0200"
      },
      "message": "Fix broadcast ops descriptions (#21087)\n\n"
    },
    {
      "commit": "84b1626b66f84a35e87a95d26226c3a3171ea78c",
      "tree": "b012508a8f20ef9d10175a976b697501c64c97ea",
      "parents": [
        "b713dc5aa3da76e2244cd4df9715a4fb2b2545ac"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Wed Jul 06 12:02:58 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jul 06 12:02:58 2022 +0200"
      },
      "message": "oneDNN FullyConnected weight caching \u0026 refactor (#21047)\n\n* FC weight and bias caching\r\n\r\n* prepare output for sum\r\n\r\n* check initialization conditions\r\n\r\n* create output mem desc\r\n\r\n* PrepareQuantization\r\n\r\n* remove unused variables\r\n\r\n* cleanup\r\n\r\n* Enable BRGEMM\r\n\r\n* Reorder functions\r\n\r\n* make minmax enum anonymous\r\n\r\n* node identificator \u0026 env flag\r\n\r\n* fix sanity\r\n\r\n* fix sanity\r\n\r\n* apply review\r\n\r\n* rename variable\r\n"
    },
    {
      "commit": "b713dc5aa3da76e2244cd4df9715a4fb2b2545ac",
      "tree": "fe5adb99b50a4a701c19eacfd445658feb568f2c",
      "parents": [
        "5abdc77f3c191bc771a5e97a9cecf30fd832de96"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Tue Jul 05 09:42:02 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 05 09:42:02 2022 +0200"
      },
      "message": "[BUGFIX] Fix DNNL requantize operator overflow error (#21079)\n\n* Fix DNNL requantize operator overflow error\r\n\r\n* Update src/operator/quantization/dnnl/dnnl_requantize-inl.h\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n"
    },
    {
      "commit": "5abdc77f3c191bc771a5e97a9cecf30fd832de96",
      "tree": "14a3685fbea658f7a0d6a4833e575d90633c74dc",
      "parents": [
        "9745d36ff48102aae0b010840d7f6c16e8982349"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Jul 05 09:39:50 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 05 09:39:50 2022 +0200"
      },
      "message": "[FEATURE] Add _npi_power_scalar and _npi_multiply_scalar fuse (#20976)\n\n* [FEATURE] Add _npi_power_scalar and _npi_multiply_scalar fuse\r\n\r\n* Merge _npi_power_scalar implementation with implementation of this fuse\r\n\r\n* Fix clang\r\n\r\n* Fix CI\r\n\r\n* Fix review and simplify the implementation\r\n\r\n* Add checks for the amount of inputs and outputs\r\n\r\n* Fix CI\r\n\r\n* Add Reset() function\r\n\r\n* Fix DNNLPowMulScalarShape and Type functions\r\n\r\n* Fix DNNLPowMulScalarType\r\n\r\n* Fix DNNLPowMulScalarType\r\n\r\n* Add generic implementation for sq_pow_mul_scalar operator\r\n\r\n* Fix sanity\r\n\r\n* Fix req\r\n\r\n* Add Filter method to property\r\n\r\n* Add new line\r\n\r\n* Fix gpu CI\r\n\r\n* Add \u0027_sg_pow_mul_scalar\u0027 to symbol_fp16.py\r\n\r\n* Fix CI on MacOS\r\n\r\n* Fix SupportDNNL*\r\n\r\n* Make PowMulScalarCompute more readable\r\n\r\n* Fix PowMulScalarCompute\r\n\r\n* Fix memory usage\r\n\r\n* Fix build"
    },
    {
      "commit": "9745d36ff48102aae0b010840d7f6c16e8982349",
      "tree": "fdfe7a33e4c3fa69b08abaecf95e194ecbcfa5f8",
      "parents": [
        "d8872c876b3adeefc11cdf2f38a51ea153b71e24"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Wed Jun 29 10:34:31 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 29 10:34:31 2022 +0200"
      },
      "message": "Improve masked_softmax performance with temperature parameter (#21082)\n\n"
    },
    {
      "commit": "d8872c876b3adeefc11cdf2f38a51ea153b71e24",
      "tree": "28949439f8e57a9b2f930ab5179fcd1b94d759e0",
      "parents": [
        "cdffaf09945f9cf3afd85ff91c78d66673247b15"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Tue Jun 28 08:53:03 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 28 08:53:03 2022 -0700"
      },
      "message": "Fix building of master website due to removed blog page. (#21083)\n\n"
    },
    {
      "commit": "cdffaf09945f9cf3afd85ff91c78d66673247b15",
      "tree": "1c7761af2b4d2744dc0ed05b6b268e1e055b247b",
      "parents": [
        "c486a0e304b61ab5f90b1ab0fc3afd1481bb7d66"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Tue Jun 28 14:53:16 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 28 14:53:16 2022 +0200"
      },
      "message": "[FEATURE] Add tanh approximation for GeLU activation (#21034)\n\n* Add gelu tanh activation\r\n\r\n* fix sanity\r\n\r\n* change tolerance\r\n\r\n* apply review\r\n\r\n* change backend function name to gelu_erf\r\n\r\n* define const\r\n"
    },
    {
      "commit": "c486a0e304b61ab5f90b1ab0fc3afd1481bb7d66",
      "tree": "4fade1e2dc78bb9785bb178e88df36695b7e3c61",
      "parents": [
        "afbef154ed09810e8addccd37169ffed8b4f7cda"
      ],
      "author": {
        "name": "PiotrWolinski - Intel",
        "email": "piotr.wolinski@intel.com",
        "time": "Thu Jun 23 16:58:47 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 23 16:58:47 2022 +0200"
      },
      "message": "[master] Node elimination graph pass (#21046)\n\n* Added node elimination graph pass\r\n\r\n* Fixed typo\r\n\r\n* Fix typo\r\n\r\n* Fix test_conv_subgraph.py\r\n\r\n* Fix test_amp_concat\r\n\r\n* Fix review\r\n\r\nCo-authored-by: Bartosz Kuncer \u003cbartosz.kuncer@intel.com\u003e"
    },
    {
      "commit": "afbef154ed09810e8addccd37169ffed8b4f7cda",
      "tree": "984f6d7bd21c7be0f76de1123d7fe9ccc52f2b19",
      "parents": [
        "b4aca83e31455fa0bec306f3753846c9e9c16141"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Thu Jun 23 15:29:23 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 23 15:29:23 2022 +0200"
      },
      "message": "Type fix for FullyConnected with sum (#21043)\n\n"
    },
    {
      "commit": "b4aca83e31455fa0bec306f3753846c9e9c16141",
      "tree": "9180c57ba92010cb9b2b3520e8ef5d13bc04df9c",
      "parents": [
        "ef2be5126592bb73bf3e42c29722b93b89b250dc"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Jun 23 13:37:47 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 23 13:37:47 2022 +0200"
      },
      "message": "Use requested mem in dot op to reduce memory usage (#21067)\n\n* Use requested mem in dot op to reduce memory usage\r\n\r\n* Fix oneDNN still allocating the memory\r\n\r\n* Assign pointer to a variable for simplification"
    },
    {
      "commit": "ef2be5126592bb73bf3e42c29722b93b89b250dc",
      "tree": "88b76adf72be6a959a86052b1f7d1ff6c99d71d8",
      "parents": [
        "63aea9e0310ed672b8816d332ae9ec25f9d8a893"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Thu Jun 23 09:41:21 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 23 09:41:21 2022 +0200"
      },
      "message": "Refactor SupportDNNL functions (#21032)\n\n* Refactor SupportDNNL functions\r\n\r\n* FixSupportDNNL for quantized reshape\r\n\r\n* Update src/operator/nn/dnnl/dnnl_softmax_output.cc\r\n\r\nCo-authored-by: bgawrych \u003cbartlomiej.gawrych@intel.com\u003e\r\n\r\n* Update src/operator/nn/dnnl/dnnl_reduce.cc\r\n\r\nCo-authored-by: bgawrych \u003cbartlomiej.gawrych@intel.com\u003e"
    },
    {
      "commit": "63aea9e0310ed672b8816d332ae9ec25f9d8a893",
      "tree": "22f8118772108f8e5c402e944f31bd9457434754",
      "parents": [
        "1ad198d6390d66d2f54dc8820cda6abe754fa0d2"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Wed Jun 22 08:27:26 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 22 08:27:26 2022 +0200"
      },
      "message": "[FEATURE] Add quantization for npi_add with oneDNN (#21041)\n\n* [FEATURE] Add quantization for npi_add with oneDNN\r\n\r\n+ tests for it,\r\n+ benchmark for FC + npi_add with broadcasted input\r\n\r\n* Fix tests in test_fc_subgraph.py\r\n\r\nThe initialisation of model weights was wrong\r\nas negative numbers couldn\u0027t be represented with uint8\r\n\r\n* Fix merge\r\n\r\n* Fix callibration and scales\r\n\r\n+ some rearrangment to make code more readable\r\n\r\n* Add tests for full path of quantization of add operators\r\n\r\n* Fix test_fc_subgraph.py::test_fc_eltwise[exp-True-True\r\n\r\nAfter previous fix initializing FC weights to only positive values for\r\nuint8, input for exp function was to high."
    },
    {
      "commit": "1ad198d6390d66d2f54dc8820cda6abe754fa0d2",
      "tree": "b264c4f610e46b96849c702e2a1561a55a803973",
      "parents": [
        "1dba76998de3584bfa98c05989845ac08db2079a"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Tue Jun 21 17:46:47 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 21 17:46:47 2022 +0200"
      },
      "message": "[FEATURE] Refactor SwapAxis operator. (#21024)\n\n* Refactor SwapAxis operator\r\n\r\n* Rename kData to kOut\r\n\r\n* Fix description of SwapAxis"
    },
    {
      "commit": "1dba76998de3584bfa98c05989845ac08db2079a",
      "tree": "585d0c45bba716eebbc4395510ff810a18155370",
      "parents": [
        "08f578b9462208be89f58ed25d653deb099b21cf"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Jun 21 10:23:18 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 21 10:23:18 2022 +0200"
      },
      "message": "Diversify default RNG seed (#21058)\n\n* Differentiate default RNG seed\r\n\r\n* Fix clang"
    },
    {
      "commit": "08f578b9462208be89f58ed25d653deb099b21cf",
      "tree": "04e15aa0cf2a4b2c154fcfa211be468bdacb36fe",
      "parents": [
        "b322bee0e7151366fee206f0bd8ce1cd8d75d4bf"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Tue Jun 21 08:53:09 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 21 08:53:09 2022 +0200"
      },
      "message": "Fix test_bf16_binary_broadcast_elemwise_mixed_input (#20986)\n\n"
    },
    {
      "commit": "b322bee0e7151366fee206f0bd8ce1cd8d75d4bf",
      "tree": "a99d1ccc8b32a1ba7318aba7800432a29a248475",
      "parents": [
        "36f4f58887b5952d19a3fc6becb3a4811cb09a38"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Mon Jun 20 13:06:55 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jun 20 13:06:55 2022 +0200"
      },
      "message": "[FEATURE] Add property removing duplicate Cast operations (#21020)\n\n* [FEATURE] Add property removing duplicate Cast operations to \u0027ONEDNN\u0027 subgraph backend\r\n\r\n* Fix CI\r\n\r\n* Fix sanity\r\n\r\n* Add checks for the amount of inputs and outputs\r\n\r\n* Fix review\r\n\r\n* Fix review"
    },
    {
      "commit": "36f4f58887b5952d19a3fc6becb3a4811cb09a38",
      "tree": "6a951ace54dc728d855c2d4e4fab80d2eaa4eb93",
      "parents": [
        "6b2d4d41f5eca839c309bcc2c76d5435f831c435"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Mon Jun 20 08:41:48 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jun 20 08:41:48 2022 +0200"
      },
      "message": "Fix pip package description (#21064)\n\n"
    },
    {
      "commit": "6b2d4d41f5eca839c309bcc2c76d5435f831c435",
      "tree": "e076716d7936df6afe0cc752accebc7db871cfef",
      "parents": [
        "9491442d147e51e883708ebdf3a2b4cbf5b5247a"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Fri Jun 17 11:02:20 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jun 17 11:02:20 2022 -0700"
      },
      "message": "Merge #21055 from v1.9.x - website changes. (#21068)\n\n* Merge #21055 from v1.9.x - website changes.\r\n\r\n* Disable clang format for javascript files."
    },
    {
      "commit": "9491442d147e51e883708ebdf3a2b4cbf5b5247a",
      "tree": "366f6c681389f8d8d0d51ebea11153d8dc441237",
      "parents": [
        "3263aeeebb3118c41cc795ec535eed4f8c4009b9"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Tue Jun 14 15:27:26 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 14 15:27:26 2022 +0200"
      },
      "message": "Update oneDNN quantization tutorial (#21060)\n\n"
    },
    {
      "commit": "3263aeeebb3118c41cc795ec535eed4f8c4009b9",
      "tree": "a1a1955d59a33939deb2584b3179cb6cd9b3b576",
      "parents": [
        "4e161d02d4f8a0b67f2fe710931f8c1525b98390"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Tue Jun 14 15:25:47 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 14 15:25:47 2022 +0200"
      },
      "message": "fix linkcheck (#21057)\n\n"
    },
    {
      "commit": "4e161d02d4f8a0b67f2fe710931f8c1525b98390",
      "tree": "63b83eb7b0e538d4c4d77c7fe51f183d9dbfba39",
      "parents": [
        "15cacdbdbf4b8609252827382bbf0e83fbe40063"
      ],
      "author": {
        "name": "waytrue17",
        "email": "52505574+waytrue17@users.noreply.github.com",
        "time": "Fri Jun 10 14:51:54 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jun 10 14:51:54 2022 -0700"
      },
      "message": "update readme for mx1.9.1 (#21050)\n\nCo-authored-by: Wei Chu \u003cweichu@amazon.com\u003e"
    },
    {
      "commit": "15cacdbdbf4b8609252827382bbf0e83fbe40063",
      "tree": "24e10733be6307bcc80c3e8ad498882fa438cb0f",
      "parents": [
        "102388a0557c530741ed8e9b31296416a1c23925"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Tue Jun 07 09:15:21 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 07 09:15:21 2022 +0200"
      },
      "message": "broadcast_like CPU optimization (#21004)\n\n* working with tmp\r\n\r\n* working without tmp\r\n\r\n* refactor\r\n\r\n* condition\r\n\r\n* refactor\r\n\r\n* remove onednn broadcast\r\n\r\n* remove temporary memory resource\r\n\r\n* fix sanity\r\n\r\n* Fix tests\r\n\r\n* apply review comments\r\n\r\n* sanity\r\n\r\n* update comment\r\n\r\nCo-authored-by: Bartlomiej Gawrych \u003cbarlomiej.gawrych@intel.com\u003e"
    },
    {
      "commit": "102388a0557c530741ed8e9b31296416a1c23925",
      "tree": "ef45c83db02d25a0ee3ea9e8c8025d9583fdbd83",
      "parents": [
        "9152b9078fa862099dc479d35ae8c26120b2c3c9"
      ],
      "author": {
        "name": "PiotrWolinski - Intel",
        "email": "piotr.wolinski@intel.com",
        "time": "Thu Jun 02 08:40:48 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 02 08:40:48 2022 +0200"
      },
      "message": "[master] Remove dnnl_ops-inl.h file (#20997)\n\n* Removed dnnl_ops-inl.h\r\n\r\n* Removed dnnl_ops-inl.h file from tracking\r\n\r\n* Formatted new files\r\n\r\n* Added missing import\r\n\r\n* Applied changes from review\r\n\r\n* Linting\r\n\r\n* Refactored DNNLSumFwd class\r\n\r\n* Linting\r\n\r\n* Fixed issue with DNNLSumFwd\r\n\r\n* Removed unused variables\r\n\r\n* Fixed oneDNN usage in refactored files\r\n\r\n* Fix sanity\r\n\r\n* Changed GetConcatForward to GetCache dand placed it in the DNNLConcatFwd classD\r\n\r\n* Removed missed comment\r\n\r\n* Moved DNNLSum function from dnnl_sum files to dnnl_base and renamed it to DNNLMemorySum"
    },
    {
      "commit": "9152b9078fa862099dc479d35ae8c26120b2c3c9",
      "tree": "4e606c118b9cf1da145f046803c838c108bb5e2d",
      "parents": [
        "ab30c56bce902988a58c1578298060835ca5c1f9"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Tue May 31 22:48:30 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue May 31 13:48:30 2022 -0700"
      },
      "message": "Fix linkcheck (#21045)\n\n"
    },
    {
      "commit": "ab30c56bce902988a58c1578298060835ca5c1f9",
      "tree": "b756087d58eb14c8eda9b677a51bbbce72e2179f",
      "parents": [
        "1687b442d9eedf4d9acc36bb84ed2f26da6bb969"
      ],
      "author": {
        "name": "PiotrWolinski - Intel",
        "email": "piotr.wolinski@intel.com",
        "time": "Mon May 30 13:36:45 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon May 30 13:36:45 2022 +0200"
      },
      "message": "[master] Fix issue with fc_eltwise fusing (#20958)\n\n* Fixed issue with fc_eltwise\r\n\r\n* Disabled exp quantization and improved functions readability\r\n\r\n* Update src/operator/nn/dnnl/dnnl_act.cc\r\n\r\nSuggested change\r\n\r\nCo-authored-by: Andrzej Kotłowski \u003cAndrzej.Kotlowski@intel.com\u003e\r\n\r\n* Fixed linting\r\n\r\n* Enabled exp for test_fc_eltwise\r\n\r\nCo-authored-by: Andrzej Kotłowski \u003cAndrzej.Kotlowski@intel.com\u003e"
    },
    {
      "commit": "1687b442d9eedf4d9acc36bb84ed2f26da6bb969",
      "tree": "050a5e9bd3970f802f3a10adffd7eb4d965511a2",
      "parents": [
        "9ca3b27635a289ef197a17eeb627464ce1c20b76"
      ],
      "author": {
        "name": "waytrue17",
        "email": "52505574+waytrue17@users.noreply.github.com",
        "time": "Wed May 25 11:29:08 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed May 25 11:29:08 2022 -0700"
      },
      "message": "[website] Update website for mxnet 1.9.1 (#21038)\n\n* [website][v1.9.x] Update website for 1.9.1 release (#21033)\r\n\r\n* update website for 1.9.1 release\r\n\r\n* set default to 1.9.1\r\n"
    },
    {
      "commit": "9ca3b27635a289ef197a17eeb627464ce1c20b76",
      "tree": "b24d26d6fd76fafa1b816d3d44640661b2f53915",
      "parents": [
        "0497efd4e4fcd18ed1cc459218daf6e67b38d536"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Mon May 23 07:54:38 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon May 23 07:54:38 2022 +0200"
      },
      "message": "[master] Enabled tests using the whole batch for calibration (#21008)\n\n* Added possibility to rerun test with all batches used for calibration\r\n\r\n* Review suggestions\r\n\r\n* Calibration dataset fix"
    },
    {
      "commit": "0497efd4e4fcd18ed1cc459218daf6e67b38d536",
      "tree": "31b5d224d216cf42dfb3e90ac6ee17bca019060b",
      "parents": [
        "a9574d2b398bde186d2a5d1405c03babdf2ebaf6"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Thu May 19 08:59:07 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 19 08:59:07 2022 +0200"
      },
      "message": "fix transformer optimization for gpt-2 (#21007)\n\n"
    },
    {
      "commit": "a9574d2b398bde186d2a5d1405c03babdf2ebaf6",
      "tree": "8c11b24b19d480612d0658ddf58fe2643d9d84dc",
      "parents": [
        "3098af4a808f905e285a4100c6b96b8e126c2c5a"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Wed May 18 07:35:38 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed May 18 07:35:38 2022 +0200"
      },
      "message": "Fix quantized_elemwise_add (#21031)\n\nWhen inputs were different types results was wrong. Now unsigned input\r\ntensor is converted to signed and later common path is used.\r\n+ Modify test to check results for different input types + some cleanup."
    },
    {
      "commit": "3098af4a808f905e285a4100c6b96b8e126c2c5a",
      "tree": "5639bb1f76789e8d3476f997ed0825501abc7766",
      "parents": [
        "0208063a89b8523c71390ef93773c6f093f74826"
      ],
      "author": {
        "name": "Paweł Głomski",
        "email": "pawel.glomski@intel.com",
        "time": "Tue May 17 14:34:09 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue May 17 14:34:09 2022 +0200"
      },
      "message": "AMP improvements + enable bf16 input for quantize_v2 (#20983)\n\n* AMP improvements + enable bf16 input for quantize_v2\r\n\r\n* Fix sanity\r\n\r\n* Improve tests, AMP conversion interface, fix forwad hooks\r\n\r\n* Fix tests\r\n\r\n* Fix imports in tests\r\n\r\n* Use different lp16_fp32 op in test\r\n\r\n* Add amp.disable_amp() context, fix tests\r\n\r\n* Add tests, generalize optimization disabling\r\n\r\n* Fix sanity\r\n\r\n* Review fixes\r\n\r\n* Use is_integral\u003c\u003e::value\r\n\r\n* Review fixes # rerun CI\r\nChange flag type to unsigned int\r\nAdd a warning for an incorrect flag attribute value\r\n\r\n* Add message to static_assert\r\n\r\n* Cast flag attribute value to int before using\r\n\r\n* Test amp node excluding, change names of tests\r\n\r\n* Fix static_cast type\r\n\r\n* Fix sanity"
    },
    {
      "commit": "0208063a89b8523c71390ef93773c6f093f74826",
      "tree": "0caba3dc0de1153886d1f69e43d2d04082764b70",
      "parents": [
        "eb8982deb2cb7bc724146b774f5a8d299feab4d4"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Mon May 16 19:00:16 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon May 16 19:00:16 2022 -0700"
      },
      "message": "Port website fixes (#20993, #20995, #21009) from v1.9.x branch (#21015)\n\n* [v1.9.x] Update website footer to include required Apache links (#20993)\r\n\r\n* Update footer to include required Apache links, according to https://www.apache.org/foundation/marks/pmcs#navigation\r\n\r\n* Better layout with col-3 class.\r\n\r\n* Fix \u0027module \u0027jinja2\u0027 has no attribute \u0027contextfilter\u0027\u0027\r\n\r\n* Update Makefile\r\n\r\n* Update Makefile\r\n\r\nCo-authored-by: Bartosz Kuncer \u003cbartosz.kuncer@intel.com\u003e\r\n\r\n* [v1.9.x] Fix css for Apache links, add to Python docs. (#20995)\r\n\r\n* Add Apache links to python docs footer, update hostname in CI script.\r\n\r\n* Don\u0027t use relative_url function.\r\n\r\n* Add link to ASF.\r\n\r\n* Fix medium blog integration. (#21009)\r\n\r\n* Update broken links to fix linkcheck.\r\n\r\nCo-authored-by: Bartosz Kuncer \u003cbartosz.kuncer@intel.com\u003e\r\nCo-authored-by: matherit \u003c97054789+matherit@users.noreply.github.com\u003e"
    },
    {
      "commit": "eb8982deb2cb7bc724146b774f5a8d299feab4d4",
      "tree": "ff5c5b459535e01419901af614f57607a2204b42",
      "parents": [
        "024e79180d296458b6558e17cb4d40f153a205ef"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Mon May 16 08:25:38 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon May 16 08:25:38 2022 -0700"
      },
      "message": "Fix CI in master (#21026)\n\n* Update minor versions of nvidia cuda containers to use that have the latest keys pre-installed.\r\n\r\n* Update the TensorRT pipeline to Cuda 11.2.\r\n\r\n* Update TensorRT pipeline to use Cuda 11.4 and update libnvinfer to 8.2.4.\r\n\r\n* Allow setting TRT version as argument in docker-compose.yml and update to TRT 8.2.4 for cuda 11.4.\r\n\r\n* Use python3 executable when building tensorrt (so we can update to ubuntu 20.04 base) and enable int64 build.\r\n\r\n* Remove unneeded line.\r\n\r\n* Support TRT 8+.\r\n\r\n* Update onnx-tensorrt to 22.02 release.\r\n\r\n* Add support for trt \u003e\u003d 8.\r\n\r\n* Fix lint\r\n\r\n* Remove debug line.\r\n\r\n* Don\u0027t upgrade libcudnn, use what is in the latest container from nvidia.\r\n\r\n* Set CUDNN_VERSION inside nvidia containers when NV_CUDNN_VERSION is set instead.\r\n\r\n* Go back to updating libcudnn8."
    },
    {
      "commit": "024e79180d296458b6558e17cb4d40f153a205ef",
      "tree": "84b1f5cf47162a772337b3890e38e246f2060384",
      "parents": [
        "bb919783b5d1f7023284b5513a2615a085b76575"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Thu Apr 28 13:32:58 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Apr 28 13:32:58 2022 +0200"
      },
      "message": "Fix bitwise_right_shift with uint input (#20974)\n\n"
    },
    {
      "commit": "bb919783b5d1f7023284b5513a2615a085b76575",
      "tree": "c0e20d4e3361099f6ef604987230d2cf53bfd3d1",
      "parents": [
        "e43c1e065e950d53093ea84ae43897f7c2129c8a"
      ],
      "author": {
        "name": "PiotrWolinski - Intel",
        "email": "piotr.wolinski@intel.com",
        "time": "Wed Apr 27 13:14:17 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Apr 27 13:14:17 2022 +0200"
      },
      "message": "[master] Unified oneDNN implementation calls for quantized operators (#20987)\n\n* Done unifying for requantize, quantize and quantized_fully_connected\r\n\r\n* Unified oneDNN implementation calls for FComputeEx\r\n\r\n* Done unifying for requantize, quantize and quantized_fully_connected\r\n\r\n* Unified oneDNN implementation calls for FComputeEx\r\n\r\n* Changed dnnl_quantize operator to avoid duplicated reorder\r\n\r\n* Added linting\r\n\r\n* Added SupportDNNLQuantize function"
    },
    {
      "commit": "e43c1e065e950d53093ea84ae43897f7c2129c8a",
      "tree": "84a29cf24afbff050c3ea915184b1af6be419a61",
      "parents": [
        "c70ea500142576439e1eb1e8725be6bed063457b"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Tue Apr 26 11:09:41 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Apr 26 11:09:41 2022 +0200"
      },
      "message": "[BUGFIX] Fix and cleanup _contrib_quantized_elemwise_add (#20999)\n\n* Fix and cleanup _contrib_quantized_elemwise_add\r\n\r\nThe key for caching oneDNN kernel was made from different elements than\r\n used for building the kernel.\r\n+ some additional cleanup\r\n\r\n* Align names according to convention"
    },
    {
      "commit": "c70ea500142576439e1eb1e8725be6bed063457b",
      "tree": "3a83ada6530adffc3d1f201a4c16ef9f71542f27",
      "parents": [
        "90d9054b7b3f8d38fd693278d440524e5815f59d"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Tue Apr 26 10:31:34 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Apr 26 10:31:34 2022 +0200"
      },
      "message": "[FEATURE] Integrate oneDNN support for scalar power() operator  (#20904)\n\n* Integrate oneDNN support for power() operator from MXNet NumPy and NDArray modules\r\n\r\n* Move DNNLPowerForward function to dnnl file\r\n\r\n* Fix include files, declaration, SupportDNNL function and OPCHECK macros\r\n\r\n* Support only fp32"
    },
    {
      "commit": "90d9054b7b3f8d38fd693278d440524e5815f59d",
      "tree": "1a88fc8a09fdc8b73aa2bcaf8cfd81374fbba9bb",
      "parents": [
        "dadb5b7c1729418f98926438b69eedae5bc8dbbc"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Apr 19 19:23:42 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Apr 19 19:23:42 2022 +0200"
      },
      "message": "[submodule] Upgrade oneDNN to v2.6 (#20982)\n\n"
    },
    {
      "commit": "dadb5b7c1729418f98926438b69eedae5bc8dbbc",
      "tree": "2dbd3b32f12c3f5dff2d4cc94bf17439c326d273",
      "parents": [
        "ea6c91abd5cb3795223b797813e2c1ea9a132e7e"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Fri Apr 15 07:01:52 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Apr 14 22:01:52 2022 -0700"
      },
      "message": "Fix CI on master branch (#21001)\n\n* Dissable test_conv2d_16c for gpu CI\r\n\r\n* Add space\r\n\r\n* Fix linkcheck"
    },
    {
      "commit": "ea6c91abd5cb3795223b797813e2c1ea9a132e7e",
      "tree": "e0d9c2388fd012f423131d951920b3aac0a86536",
      "parents": [
        "08737b289ddc109c3e0665bc746505b1a52dfdb1"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Wed Apr 06 09:59:07 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Apr 06 09:59:07 2022 +0200"
      },
      "message": "Restore quantized RNN to master (#20952)\n\n* Restore quantized RNN\r\n\r\nsanity\r\n\r\n* Add tests \u0026 disable LSTMP from quantization\r\n\r\n* apply review comments\r\n\r\n* change link\r\n\r\n* Add new lines at the EOF\r\n\r\n* Add ops to amp lists\r\n\r\n* Remove unused features\r\n\r\n* Fix DataDesc handling in quantization\r\n\r\n* fix website\r\n\r\n* fix sanity\r\n\r\n* remove magic number\r\n\r\nCo-authored-by: Bartlomiej Gawrych \u003cbarlomiej.gawrych@intel.com\u003e"
    },
    {
      "commit": "08737b289ddc109c3e0665bc746505b1a52dfdb1",
      "tree": "afba20fa299acd59d2ca0f2790277aaac6b7b329",
      "parents": [
        "f16c3c77cb5ceddb5b8a8327e84a7d60c237a313"
      ],
      "author": {
        "name": "mozga",
        "email": "mateusz.ozga@intel.com",
        "time": "Fri Apr 01 19:12:23 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Apr 01 19:12:23 2022 +0200"
      },
      "message": "[Master] Decouple OneDNN data structures in MXNet C++ API (#20624)\n\n* NDArry file has been modified, there are a few chnages:\r\n\r\n    1. OneDNN header was moved into *cc file\r\n    2. OneDNN object are created on-the-fly: static_cast is needed\r\n\r\n* Removed static_cast\u003c*\u003e\r\n\r\n* Roll back ndarray file\r\n\r\n* Roll Back - fwd namespace declaration\r\n\r\n* Clang-format: auto-format\r\n\r\n* Sanity fix\r\n\r\n* Conv saity-fix\r\n\r\n* Fix typo\r\n\r\n* Fix tenary\r\n\r\n* Review\r\n\r\n* Review: test file\r\n\r\n* Static_cast: removed redundant cast"
    },
    {
      "commit": "f16c3c77cb5ceddb5b8a8327e84a7d60c237a313",
      "tree": "82aec8a9ef211ee0c1b3c23851c4158c6d41cd76",
      "parents": [
        "e8ff13ce6bb84a0f22fe7a4a6edca1312bbe8ef0"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Thu Mar 31 10:55:47 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 31 10:55:47 2022 +0200"
      },
      "message": "Post quantize property improvement (#20929)\n\n* Post quantize property improvement\r\n\r\n* Review suggestion\r\n\r\n* Sanity fix\r\n\r\n* Review change"
    },
    {
      "commit": "e8ff13ce6bb84a0f22fe7a4a6edca1312bbe8ef0",
      "tree": "8d81968264041bf0739b31b745c3f6e6b4032e84",
      "parents": [
        "4672ca186f911abdcee5bf178c73f4d13f6fb0ce"
      ],
      "author": {
        "name": "Paweł Głomski",
        "email": "pawel.glomski@intel.com",
        "time": "Thu Mar 31 09:42:36 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 31 09:42:36 2022 +0200"
      },
      "message": "Improve AMP, bf16 support. Support oneDNN ops in AMP (#20753)\n\n* Add amp_cast fuse\r\n\r\n* Do not fuse with quantized nodes\r\n\r\n* Port to 2.0, add convolution fuse support\r\n\r\n* Fix offline casting of parameters\r\n\r\n* Fix symbol_bf16\r\n\r\n* for review\r\n\r\n* Fix amp fuse with branched output, add bf16 support in necessary places\r\n\r\n* Map dtype to name in a switch\r\n\r\n* Handle mutable inputs for widest_dtype nodes\r\n\r\n* Correct handling amp_cast nodes from the original graph\r\n\r\n* Add license\r\n\r\n* Fix sanity\r\n\r\n* Fix InferType functions\r\n\r\n* Fix importing of bf16 parameters, minor fixes\r\n\r\n* Refactor\r\n\r\n* Fix tests\r\n\r\n* Fix amp_out_dtype description\r\n\r\n* Simplify AMP python code, unify dtype operations\r\n\r\n* WIP compiled\r\n\r\n* Fix issues\r\n\r\n* Add convolution to amp_cast fuse op list\r\n\r\n* Fix tests\r\n\r\n* Undo infertype changes in unrelated ops\r\n\r\n* Fix sanity\r\n\r\n* Fix dtype comparison, standardize dtype operations\r\n\r\n* Mark AMP tests to use np semantics\r\n\r\n* Remove redundant const\r\n\r\n* Disable new bf16 tests on macos\r\n\r\n* Enable bf16 concat test\r\n\r\n* Remove unused pass\r\n\r\n* Use actual node entries for mapping\r\n\r\n* Remove extra ops from bf16 amp lists\r\n\r\n* Review fixes + comments and cleanup\r\n\r\n* Review fixes\r\n\r\n* Review fixes + additional checks during conversion\r\n\r\n* Hybridize only when necessary\r\n\r\n* Fix warning message\r\n\r\n* Review fixes"
    },
    {
      "commit": "4672ca186f911abdcee5bf178c73f4d13f6fb0ce",
      "tree": "0ed0bd2bd01a2f92b6bc07e9229c0fec48c83d6b",
      "parents": [
        "52d8d48c16e20b93853706e2151eec359a6756f1"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Mar 29 17:59:27 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 29 17:59:27 2022 +0200"
      },
      "message": "[FEATURE] Add oneDNN support for npi: exp, square and sqrt (#20981)\n\n* [FEATURE] Add oneDNN support for npi: log, exp, square and sqrt\r\n\r\n* Fix warnings\r\n\r\n* Fix CI"
    },
    {
      "commit": "52d8d48c16e20b93853706e2151eec359a6756f1",
      "tree": "5ee198bf57c3c5ce9cf4614c7d7924df33b3efe8",
      "parents": [
        "6b9213cf98c39e3541c185f0a7b8fd9704d86d38"
      ],
      "author": {
        "name": "RafLit",
        "email": "rafal.litka@intel.com",
        "time": "Tue Mar 29 17:50:12 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 29 17:50:12 2022 +0200"
      },
      "message": "fix onednn install header location (#20893)\n\n* fix onednn install header location\r\n\r\n* move GNUInstallDirs"
    },
    {
      "commit": "6b9213cf98c39e3541c185f0a7b8fd9704d86d38",
      "tree": "ca71ce1206a703084cf01a6a9bc1a53aced89ba6",
      "parents": [
        "db730cc495da896f6882a6f2680b5918e48cd817"
      ],
      "author": {
        "name": "AdamGrabowski",
        "email": "adam.grabowski@intel.com",
        "time": "Mon Mar 28 14:07:23 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 28 14:07:23 2022 +0200"
      },
      "message": "Support bfloat and float mixed input in elemwise broadcast ops (#20932)\n\n"
    },
    {
      "commit": "db730cc495da896f6882a6f2680b5918e48cd817",
      "tree": "d9d5789eb0198aa8b4417c59ba34041ef8f6d8eb",
      "parents": [
        "e165b35ce0c96fd27e293e5eaac74be7ad9d5b0f"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Mon Mar 28 09:51:01 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 28 09:51:01 2022 +0200"
      },
      "message": "[submodule] Upgrade oneDNN to v2.5.4 (#20980)\n\n"
    },
    {
      "commit": "e165b35ce0c96fd27e293e5eaac74be7ad9d5b0f",
      "tree": "702db2b5ac46434d369d77f9faf3c7b46075cce2",
      "parents": [
        "b03d717118621f112f1645da4aa95c6e0704a3f0"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Mon Mar 28 09:49:58 2022 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 28 09:49:58 2022 +0200"
      },
      "message": "Enable possibility of disabling particular graph pass (#20961)\n\nThere was no possibility to disable particular fusing pass via\r\nenvironment variable when use optimize_for function."
    },
    {
      "commit": "b03d717118621f112f1645da4aa95c6e0704a3f0",
      "tree": "3c3afa37950008d7bd72902ea210cf6e3dbd7054",
      "parents": [
        "ef378e46505f6fcb31ccb03faa62b4f18345cc78"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Fri Mar 25 14:40:04 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 25 14:40:04 2022 +0100"
      },
      "message": "[BUGFIX] Fix tests/python/dnnl/subgraphs/test_conv_subgraph.py (#20970)\n\n* [BUGFIX] Fix tests/python/dnnl/subgraphs/test_conv_subgraph.py\r\n\r\n* Fix sanity"
    },
    {
      "commit": "ef378e46505f6fcb31ccb03faa62b4f18345cc78",
      "tree": "89da58ce3f13e1c42364e16f0088ed5e58633232",
      "parents": [
        "54d153caacf5205f12aa017ee5c6489f1f7ae8ca"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Mar 24 15:59:48 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 24 15:59:48 2022 +0100"
      },
      "message": "[operator] Integrate oneDNN matmul primitive to mxnet dot operator (#20911)\n\n* [operator] Integrate oneDNN matmul primitive to mxnet dot operator\r\n\r\n* Fix numpy dot test\r\n\r\n* Fix reviews"
    },
    {
      "commit": "54d153caacf5205f12aa017ee5c6489f1f7ae8ca",
      "tree": "3dc47d346f713ceb614b89fa311a3e0d40ecd86e",
      "parents": [
        "11423bc74c732a859adf466aee05b856eeb118f3"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Thu Mar 24 14:41:25 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 24 14:41:25 2022 +0100"
      },
      "message": "[operator] Integrate oneDNN eltwise_tanh into _npi_tanh operator (#20972)\n\n* [operator] Integrate oneDNN eltwise_tanh into _npi_tanh operator\r\n\r\n* Fix SupportDNNLTanh function"
    },
    {
      "commit": "11423bc74c732a859adf466aee05b856eeb118f3",
      "tree": "1fff7c2f9f4f4b5eb08b9d0f81d135da051b5b3d",
      "parents": [
        "83c875d2077b87ef0f61a9122f1649cd896aac21"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Thu Mar 24 09:38:23 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 24 09:38:23 2022 +0100"
      },
      "message": "Fix subgraph tests (#20969)\n\nPreviously single output was wrongly compared with tolerance calculate\r\nseparately for each row instead of whole tensor.\r\nIt fixes tests test_fc_subgraph.py -k \"test_fc_transpose and shape2\"\r\nrun with MXNET_TEST_SEED\u003d1655704507 form the issue #20967."
    },
    {
      "commit": "83c875d2077b87ef0f61a9122f1649cd896aac21",
      "tree": "377b2ef0228ac0c21eddd880aa775e836dbc114e",
      "parents": [
        "51e5204cab23b59182860aa7dbedef67dedc1e06"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Mar 22 14:15:27 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 22 14:15:27 2022 +0100"
      },
      "message": "Rename SetMKLMem func to SetDNNLMem to avoid confusion (#20948)\n\n"
    },
    {
      "commit": "51e5204cab23b59182860aa7dbedef67dedc1e06",
      "tree": "dccadb565f5d2173e884782e51c3885a1d30da97",
      "parents": [
        "5ab7f64c2e99890502aa456798b74523e76f3c17"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Mar 22 14:12:25 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 22 14:12:25 2022 +0100"
      },
      "message": "Add support for up 12 dims for oneDNN tensors in MXNet (#20913)\n\n"
    },
    {
      "commit": "5ab7f64c2e99890502aa456798b74523e76f3c17",
      "tree": "b69cc76501b2155ada2970cc15e2564de9799e53",
      "parents": [
        "95a6a54adf8b9dfbabe8d558109896e0b88fb9e4"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Fri Mar 18 16:55:07 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 18 16:55:07 2022 -0700"
      },
      "message": "[2.0] [BACKPORT] of [1.x][FEATURE] CUDA graphs support (#19142) (#20324)\n\n* [1.x][FEATURE] CUDA graphs support (#19142)\r\n\r\n* Initial cherry-pick\r\n\r\n* Store NodeAttrs in OpExecutor\r\n\r\n* Do not allow stateful operations in CUDA graphs and provide mechanism\r\nfor marking ops as safe\r\n\r\n* Guard against using ops with synchronization\r\n\r\n* Cleaning\r\n\r\n* Properly guard graphs\r\n\r\n* Limit graphs to CUDA 10.2+\r\n\r\n* Fix the compilation when graphs are not available\r\n\r\n* Guarding the libcuda.so usage behind RTC compilation flag\r\n\r\n* Document the env variables\r\n\r\n* Add test\r\n\r\n* Fix the test\r\n\r\n* Use with_environment\r\n\r\n* Fix compile and test_cuda_graphs\r\n\r\n* Fix lint\r\n\r\n* Mark more ops as not CUDA Graphs compatible\r\n\r\n* Mark some linalg ops as not CUDA Graphs compatible\r\n\r\n* Marked 2 ops CUDA Graphs incompatible due to cpu-\u003egpu copy\r\n\r\n* Mark cuDNN Dropout as fully CUDA Graphs compatible.  Reenable tests.\r\n\r\n* clang-tidy fixes\r\n\r\n* More clang-tidy fixes\r\n\r\n* Avoid CUDA_CALL(e): improper macro expansion\r\n\r\n* Add compile guard to Dropout\u0027s FIsCUDAGraphsCompatible def\r\n\r\n* Temporarily add \u0027-s\u0027 to pytest serial tests\r\n\r\n* Fix DropoutOp.dropout_passthrough_ handling for CUDA Graphs\r\n\r\n* Adapt test_gluon_gpu.py::test_cuda_graphs for gluon2.0\r\n\r\n* Create CUDA Graph \u0027dot\u0027 files if MXNET_CUDA_GRAPHS_DBG_FILE\u003d\u003cfile_prefix\u003e\r\n\r\n* Fix clang-tidy\r\n\r\n* Fix more clang-tidy\r\n\r\n* Skip test_np_standard_binary_funcs test of 0-dim array broadcast\r\n\r\n* Improve test_rnn_layers_fp{16,32} invocation\r\n\r\n* Run test_rnn_layers_fp32 only when cuDNN is present\r\n\r\n* Fix potential out-of-bounds write in count_sketch.cu\r\n\r\n* Add temp output to debug centos crash\r\n\r\n* Mark InstanceNorm and LeakyRELU as not CUDA Graphs compatible\r\n\r\n* Ops calling FStatefulCompute* are not CUDA Graphs compatible by default\r\n\r\n* Fix clang-tidy\r\n\r\n* Revert \"Add temp output to debug centos crash\"\r\n\r\nThis reverts commit e013a85ea599fa761cb98762f11feab6e7d74049.\r\n\r\n* Quiet \u0027unused variable\u0027 compilation warning\r\n\r\n* Trigger CI\r\n\r\n* Check of FCreateOpState removed given new check for FStatefulCompute*\r\n\r\n* Revert \"Temporarily add \u0027-s\u0027 to pytest serial tests\"\r\n\r\nThis reverts commit 5a2f847558a7f55790f1ad1fb5ee930b4ad1a3a9.\r\n\r\nCo-authored-by: Przemyslaw Tredak \u003cptredak@nvidia.com\u003e"
    },
    {
      "commit": "95a6a54adf8b9dfbabe8d558109896e0b88fb9e4",
      "tree": "fab02a19e4cb570944cd8f63a74856edc4bf62e6",
      "parents": [
        "e97255629c122b03383f4771479c90d912e73773"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Fri Mar 18 16:41:44 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 18 16:41:44 2022 -0700"
      },
      "message": "[v2.0] [BUGFIX] Port #20940 (Upgrade numpy to \u003c1.20.0 ...) from v1.9.x (#20957)\n\n* Stop skipping tests mentioned in issue 18600\r\n\r\n* Reenable test_np_random_chisquare also\r\n\r\n* [v1.9.x] [BUGFIX] Upgrade numpy to \u003c1.20.0 to avoid security vulnerabilities affecting numpy\u003c1.19.1 (#20940)\r\n\r\n* Pin numpy\u003d\u003d1.19.1 to demonstrate issues\r\n\r\n* Relax min numpy version\r\n\r\n* Make test_np_array_function_protocol delete() testing work with numpy\u003e\u003d1.19\r\n\r\n* Fix test_np_delete to also work with numpy\u003e\u003d1.19\r\n\r\n* Pip install python module \u0027packaging\u0027\r\n\r\n* More pip-install \u0027packaging\u0027\r\n\r\n* Update windows requirements.txt\r\n\r\n* Allow numpy as advanced as 1.19.5\r\n\r\n* Fix test_np_random_{beta,f,chisquare}"
    },
    {
      "commit": "e97255629c122b03383f4771479c90d912e73773",
      "tree": "62b66aed40264dd88a92764c1ce8d87a126824bd",
      "parents": [
        "8f199c29eb0420a4f3306a413f6dfc7469fca526"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Thu Mar 17 16:15:53 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 17 16:15:53 2022 -0700"
      },
      "message": "[BUGFIX] Fix mxnet.numpy.eye() handling of extreme k parameter values (#20965)\n\n* Fix mxnet.numpy.eye() handling of extreme k parameter values\r\n\r\n* Fix broken web link"
    },
    {
      "commit": "8f199c29eb0420a4f3306a413f6dfc7469fca526",
      "tree": "ed49239e69dd3392580a1f74781098b24c8cc837",
      "parents": [
        "ded04008fd3cd95793e0eaaa72f5904deb77f838"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Thu Mar 17 07:40:21 2022 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 17 15:40:21 2022 +0100"
      },
      "message": "Fix broken web link (#20966)\n\nMerging to unblock CI - my concerns can be discussed in issue if necessary"
    },
    {
      "commit": "ded04008fd3cd95793e0eaaa72f5904deb77f838",
      "tree": "d04e9981466afd4da5df4c2e798161ed383a0220",
      "parents": [
        "a03dc1eff506f2f3a4978ba5536df42d0c022246"
      ],
      "author": {
        "name": "bartekkuncer",
        "email": "bartosz.kuncer@intel.com",
        "time": "Tue Mar 15 18:18:52 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 15 18:18:52 2022 +0100"
      },
      "message": "[submodule] Upgrade oneDNN to v2.5.3 (#20946)\n\n"
    },
    {
      "commit": "a03dc1eff506f2f3a4978ba5536df42d0c022246",
      "tree": "001b9c22d9f83314e8928dafe57a4df2f0f3a444",
      "parents": [
        "927d68bbe531c6112342806014b2fcad856584cc"
      ],
      "author": {
        "name": "RafLit",
        "email": "rafal.litka@intel.com",
        "time": "Tue Mar 15 17:17:12 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 15 09:17:12 2022 -0700"
      },
      "message": "port #20926 to master (#20949)\n\n"
    },
    {
      "commit": "927d68bbe531c6112342806014b2fcad856584cc",
      "tree": "ee078e335d51825c21663d624bac4f3e37d2ac24",
      "parents": [
        "fc54fabe61573d0405b6ddead8c1501aedad3a11"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Tue Mar 15 14:57:29 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 15 14:57:29 2022 +0100"
      },
      "message": "[PERFORMANCE] Improve shuffle implementation (#20928)\n\n* Improve shuffle implementation\r\n\r\n* Get rid of compilation warning treated as an error\r\n\r\n* Fix compilation on Windows"
    },
    {
      "commit": "fc54fabe61573d0405b6ddead8c1501aedad3a11",
      "tree": "f36094c15fc04af07a68f9bb0ed26f2d5ecef468",
      "parents": [
        "13b869002877a80f0f50a0f6cd2b0d8b55ea905f"
      ],
      "author": {
        "name": "Dick Carter",
        "email": "dcarter@nvidia.com",
        "time": "Tue Mar 08 15:00:20 2022 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 08 15:00:20 2022 -0800"
      },
      "message": "[FEATURE] Add g5 instance to CI (#20876)\n\n* Add g5 instance to jenkinsfiles where both p3 and g4 are mentioned\r\n\r\n* Remove reference to non-existent restricted-mxnetlinux-gpu-g5\r\n\r\n* Enable unittest job on g5\r\n\r\n* Fix Jenkinsfile_unix_gpu syntax\r\n\r\n* Include A10G arch 86 in build for g5\r\n\r\n* Update is_TF32_enabled() for SM arch \u003e 80\r\n\r\n* Remove gpu arch 86 from centos builds on cuda 10\r\n\r\n* Fix test_convolution_{grouping,dilated_impulse_response}, test_np_linalg_qr\r\n\r\n* Fix test_convolution_grouping on A100\r\n\r\n* Fix test_rnn_unroll_variant_length\r\n\r\n* Fix test_convolution_dilated_impulse_response\r\n\r\n* Skip test_np_standard_binary_funcs test of 0-dim array broadcast\r\n\r\n* Temporarily add \u0027-s\u0027 to pytest cpu tests\r\n\r\n* Revert \"Temporarily add \u0027-s\u0027 to pytest cpu tests\"\r\n\r\nThis reverts commit 4a9056a26f8c210497e3b5ed2318e30c8c2dbc5e.\r\n\r\n* Improve test_rnn_layers_fp{16,32} invocation\r\n\r\n* Pin MarkupSafe\u003d\u003d2.0.1 to avoid soft_unicode import failure\r\n\r\n* Run test_rnn_layers_fp32 only when cuDNN is present\r\n\r\n* Fix potential out-of-bounds write in count_sketch.cu\r\n\r\n* Revert \"Pin MarkupSafe\u003d\u003d2.0.1 to avoid soft_unicode import failure\"\r\n\r\nThis reverts commit ae17b1f2af787427740c66a05ee1fb733ea56dd3."
    },
    {
      "commit": "13b869002877a80f0f50a0f6cd2b0d8b55ea905f",
      "tree": "89fc80f59e3a6b395559b29fa8c4726dc444b20b",
      "parents": [
        "7a49008de42e97c3f5d1634955e0db4954739907"
      ],
      "author": {
        "name": "Joe Evans",
        "email": "joseph.evans@gmail.com",
        "time": "Tue Mar 08 09:46:34 2022 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 08 09:46:34 2022 -0800"
      },
      "message": "Avoid modifying loaded library map while iterating in lib_close() (#20941)\n\n* Update close libs to not modify map while iterating over opened libraries, rename loaded_libs to loaded_libs_ to signify it is a private member.\r\n\r\n* Clean up and simplify library close code.\r\n\r\n* Fix clang-format."
    },
    {
      "commit": "7a49008de42e97c3f5d1634955e0db4954739907",
      "tree": "3a94fd87fea31f9034c1742634769f74481da5a3",
      "parents": [
        "588f541a9b6d512f885bb2d759530c1eebb75e67"
      ],
      "author": {
        "name": "DominikaJedynak",
        "email": "dominika.jedynak@intel.com",
        "time": "Tue Mar 08 10:06:59 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Mar 08 10:06:59 2022 +0100"
      },
      "message": "[BUGFIX] Type fix for large tensors (#20922)\n\n* Dimension type fix\r\n\r\n* Review suggestion"
    },
    {
      "commit": "588f541a9b6d512f885bb2d759530c1eebb75e67",
      "tree": "821abbf4dc750ea8e798ab91cfa77d6f59500cb8",
      "parents": [
        "54e122c97b659602e96d63284fae777965c6c046"
      ],
      "author": {
        "name": "mozga",
        "email": "mateusz.ozga@intel.com",
        "time": "Mon Mar 07 12:00:38 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 07 12:00:38 2022 +0100"
      },
      "message": "[master] 2022.00 MKL\u0027 version, update (#20865)\n\n* MKL\u0027 version update\r\n\r\n* MKL\u0027 version update 2022.1.2\r\n\r\n* MKL\u0027 version update 2022.0.2"
    },
    {
      "commit": "54e122c97b659602e96d63284fae777965c6c046",
      "tree": "d415435b0ee5944e7b21ff5056ff6f30bd2a3abd",
      "parents": [
        "6c2b3dc41ae0be1ae583a931838058388a37ec1c"
      ],
      "author": {
        "name": "RafLit",
        "email": "rafal.litka@intel.com",
        "time": "Mon Mar 07 08:54:45 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 07 08:54:45 2022 +0100"
      },
      "message": "quantized transpose operator (#20817)\n\n* add quantized transpose\r\n\r\n* add quantized transpose test\r\n\r\n* add license to new files\r\n\r\n* check support, fix warnings\r\n\r\n* remove inplace quantized transpose\r\n\r\n* fix inplace quantized transpose\r\n\r\n* add quantized transpose test\r\n\r\n* separate transpose ops\r\n\r\n* review fixes\r\n\r\n* add fallbacks for NP and ND\r\n\r\n* fix formatting\r\n\r\n* fix formatting\r\n\r\n* update np transpose test\r\n\r\n* add operators to amp lists\r\n\r\n* add macro operator creation\r\n\r\n* add tests\r\n\r\n* review fix\r\n\r\n* remove relative includes"
    },
    {
      "commit": "6c2b3dc41ae0be1ae583a931838058388a37ec1c",
      "tree": "deb8ea22ab1a5fb305ebe3b69a2b60009e2f15a0",
      "parents": [
        "6e25c884bab56727ef0e9b81b444875d492d0587"
      ],
      "author": {
        "name": "Andrzej Kotłowski",
        "email": "Andrzej.Kotlowski@intel.com",
        "time": "Fri Mar 04 10:13:50 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 04 10:13:50 2022 +0100"
      },
      "message": "Remove first_quantization_pass FC property (#20908)\n\nremove first_quantization_pass attribute from subgraph FC as it is not needed any longer"
    },
    {
      "commit": "6e25c884bab56727ef0e9b81b444875d492d0587",
      "tree": "8e58a62b0fd69425c4a9d1d9a02cb62824d9752e",
      "parents": [
        "93aa9e33fcb2f216179c691ed9461bc96e37ae70"
      ],
      "author": {
        "name": "bgawrych",
        "email": "bartlomiej.gawrych@intel.com",
        "time": "Fri Mar 04 09:26:20 2022 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 04 09:26:20 2022 +0100"
      },
      "message": "Add oneDNN support for \"where\" operator (#20862)\n\n* Where operator enabled in oneDNN\r\n\r\n* Fix bug \u0026 refactor\r\n\r\n* fix sanity\r\n\r\n* apply review\r\n\r\n* Fix get_broadcastable_shape function\r\n\r\n* Apply review\r\n\r\n* Remove unused variable\r\n\r\n* Apply suggestions from code review\r\n\r\nCo-authored-by: bartekkuncer \u003cbartosz.kuncer@intel.com\u003e\r\n"
    },
    {
      "commit": "93aa9e33fcb2f216179c691ed9461bc96e37ae70",
      "tree": "e17935b2eef8a2b0d5ece8354819b2977709774a",
      "parents": [
        "8069a18cc3c71094458095655b2f81a7f34ac274"
      ],
      "author": {
        "name": "Sheng Zha",
        "email": "szha@users.noreply.github.com",
        "time": "Wed Mar 02 09:17:08 2022 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Mar 02 09:17:08 2022 -0500"
      },
      "message": "add Bartłomiej as committer (#20896)\n\nSigned-off-by: Sheng Zha \u003czhasheng@amazon.com\u003e"
    }
  ],
  "next": "8069a18cc3c71094458095655b2f81a7f34ac274"
}