{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":633546100,"defaultBranch":"main","name":"onnxruntime","ownerLogin":"carzh","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2023-04-27T18:31:58.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/22922935?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1721671745.0","currentOid":""},"activityList":{"items":[{"before":"4af62918418a04bcd219471dfb8f8b27a1e8a8b5","after":"6cc06ad06967db4611138b4028df9ffacc6eb860","ref":"refs/heads/main","pushedAt":"2024-09-24T17:43:55.000Z","pushType":"push","commitsCount":162,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"GQA MLFloat16 cpu (#22102)\n\n### Description\r\n\r\n\r\n\r\n### Motivation and Context\r\n\r\n\r\n---------\r\n\r\nCo-authored-by: Your Name ","shortMessageHtmlLink":"GQA MLFloat16 cpu (microsoft#22102)"}},{"before":"fb9ce18e88b9fe3c7bb6f98e80587aad522f58da","after":"4af62918418a04bcd219471dfb8f8b27a1e8a8b5","ref":"refs/heads/main","pushedAt":"2024-08-23T22:20:51.000Z","pushType":"push","commitsCount":15,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"Refine `op_types_to_quantize` argument handling in matmul_4bits_quantizer.py (#21815)\n\n### Description\r\n\r\n\r\nRefine `op_types_to_quantize` argument handling in\r\nmatmul_4bits_quantizer.py\r\n\r\n### Motivation and Context\r\n\r\nThe default `op_types_to_quantize \"MatMul\"` will cause\r\n`tuple(args.op_types_to_quantize)` to become `('M', 'a', 't', 'M', 'u',\r\n'l')`, which is not expected.","shortMessageHtmlLink":"Refine op_types_to_quantize argument handling in matmul_4bits_quant…"}},{"before":"63e88499923cc89bb01b9b6844ffe7bb79a0149d","after":"fb9ce18e88b9fe3c7bb6f98e80587aad522f58da","ref":"refs/heads/main","pushedAt":"2024-08-21T18:16:41.000Z","pushType":"push","commitsCount":22,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"Add K=0 check to MatMul::Compute() specialization. (#21803)\n\nAdd K=0 check to `MatMul::Compute()` specialization.\r\nAdd unit test to cover both primary template and float specialization.","shortMessageHtmlLink":"Add K=0 check to MatMul<float>::Compute() specialization. (microsoft#…"}},{"before":"906ae77eeab8a42a64ab28e12d2fd8dd2b5a4a10","after":"63e88499923cc89bb01b9b6844ffe7bb79a0149d","ref":"refs/heads/main","pushedAt":"2024-08-16T22:26:27.000Z","pushType":"push","commitsCount":32,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"build_aar_package.py - Check that executable is present before trying to copy it. (#21730)\n\nCheck that executable is present before trying to copy it.\r\n\r\nAccommodate builds where we skip building the test executables.","shortMessageHtmlLink":"build_aar_package.py - Check that executable is present before trying…"}},{"before":"a6c5e2cd20dd890f416806e0afbb3b5968030f4d","after":"906ae77eeab8a42a64ab28e12d2fd8dd2b5a4a10","ref":"refs/heads/main","pushedAt":"2024-08-09T21:43:31.000Z","pushType":"push","commitsCount":109,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[TensorRT EP] Add null_ptr check to avoid crash when running session which was failed to generate trt_engine previously (#21621)\n\n### Description\r\n\r\nAdd null_ptr check to avoid crash when running session which was failed\r\nto generate trt_engine previously\r\n\r\n\r\n### Motivation and Context\r\n\r\n\r\nReported and verified by\r\nhttps://github.com/microsoft/onnxruntime/issues/21567","shortMessageHtmlLink":"[TensorRT EP] Add null_ptr check to avoid crash when running session …"}},{"before":null,"after":"16905ccdc59ab5f80a17914b3e5622717746c462","ref":"refs/heads/carzh/gh-pages-export","pushedAt":"2024-07-22T18:09:05.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"Update DirectML-ExecutionProvider.md (#21407)\n\nUpdate DML EP docs w/ correct DML + ONNX opset version\r\n\r\n### Description\r\n\r\n\r\n\r\n\r\n### Motivation and Context\r\n","shortMessageHtmlLink":"Update DirectML-ExecutionProvider.md (microsoft#21407)"}},{"before":"0f4c39ec47d7152520d5389aa3ac821db82fab0a","after":"a6c5e2cd20dd890f416806e0afbb3b5968030f4d","ref":"refs/heads/main","pushedAt":"2024-07-22T18:08:33.000Z","pushType":"push","commitsCount":14,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[CUDA] FusedMHARunnerFP16v2 thread-safe (#21420)\n\n### Description\r\n- [x] Rewrite FusedMHARunnerFP16v2 to make it thread-safe. \r\n- [x] Add multi-threading tests\r\n\r\nPreviously, the kernel parameters params is stored as a member of mha\r\nrunner, which means that different threads might change the params at\r\nthe same time and impacts the other threads.\r\n\r\nFor example, if batch_size and seq_len was changed by another thread to\r\nlarger values in setup(...), buffer overrun might happen in run(...)\r\nbecause a kernel could read/write memory out of range of allocated\r\nbuffers.\r\n\r\nIn new implementation, I change the api and remove mutable member\r\nvariables to make it thread safe. Below is summary of change:\r\n\r\nBefore:\r\n```\r\nclass FusedMHARunnerFP16v2::mhaImpl {\r\n void setup(int seq_len, int batch_size) {\r\n // change scalar params\r\n }\r\n\r\n void run(input, output) {\r\n // change params for input and output pointers\r\n // launch kernel using params\r\n }\r\n\r\n Fused_multihead_attention_params_v2 params; // mutable, not thread-safe\r\n}\r\n```\r\n\r\nAfter:\r\n```\r\nclass FusedMHARunnerFP16v2::FmhaImpl {\r\n void setup(int seq_len, int batch_size, Fused_multihead_attention_params_v2& params) {\r\n // change params\r\n }\r\n\r\n void run(params, input, output) {\r\n // change params with input and output pointers\r\n // launch kernel using params\r\n }\r\n}\r\n```\r\n\r\n### Motivation and Context\r\nhttps://github.com/microsoft/onnxruntime/issues/18854\r\nhttps://github.com/microsoft/onnxruntime/issues/21413","shortMessageHtmlLink":"[CUDA] FusedMHARunnerFP16v2 thread-safe (microsoft#21420)"}},{"before":null,"after":"0f4c39ec47d7152520d5389aa3ac821db82fab0a","ref":"refs/heads/carzh/export_working","pushedAt":"2024-07-17T17:12:24.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[ROCM] adjust test_flash_attn_rocm test tolerance (#21379)\n\nThe test_flash_attn_rocm.py from\r\nhttps://github.com/microsoft/onnxruntime/pull/21032 failed frequently.\r\nFor example, I saw two failed jobs today:\r\nE Max absolute difference: 0.002167\r\nE Max absolute difference: 0.002686\r\n\r\nAdjust the abs threshold from 0.002 to 0.005, and use default relative tolerance rtol=0.001.","shortMessageHtmlLink":"[ROCM] adjust test_flash_attn_rocm test tolerance (microsoft#21379)"}},{"before":"cc0de0d5262cd1a76532b79ddde4a7d799f840b4","after":"0f4c39ec47d7152520d5389aa3ac821db82fab0a","ref":"refs/heads/main","pushedAt":"2024-07-17T17:12:15.000Z","pushType":"push","commitsCount":34,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[ROCM] adjust test_flash_attn_rocm test tolerance (#21379)\n\nThe test_flash_attn_rocm.py from\r\nhttps://github.com/microsoft/onnxruntime/pull/21032 failed frequently.\r\nFor example, I saw two failed jobs today:\r\nE Max absolute difference: 0.002167\r\nE Max absolute difference: 0.002686\r\n\r\nAdjust the abs threshold from 0.002 to 0.005, and use default relative tolerance rtol=0.001.","shortMessageHtmlLink":"[ROCM] adjust test_flash_attn_rocm test tolerance (microsoft#21379)"}},{"before":"8c2689877fb48bfb4a6a133b020cdb5ec7c9b066","after":"cc0de0d5262cd1a76532b79ddde4a7d799f840b4","ref":"refs/heads/main","pushedAt":"2024-07-10T00:19:07.000Z","pushType":"push","commitsCount":23,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[Build] Propagate build option for CUDA minimal to TRT (#20695)\n\n### Description\r\n\r\nExtend cuda minimal option to TRT provider, as with TRT 10 no linking to\r\ncuDNN is required anymore\r\n.\r\nBesides that with the new engine dump feature it is also possible to\r\nembed an engine in to an ONNX and not ship a builder lib.\r\nIn addition to that this has roughly the same deserialization\r\ntime/session setup time that using TRT standalone has.\r\n\r\n### Motivation and Context\r\n\r\n```\r\nexe_builder_lib\\onnxruntime_perf_test.exe -I -e tensorrt -r 5 -i 'trt_engine_cache_enable|1 trt_timing_cache_enable|1 trt_dump_ep_context_model|1 trt_weightless_engine_enable|1' model.onnx\r\n\r\n\r\nexe_no_builder_lib\\onnxruntime_perf_test.exe -I -e tensorrt -r 5 -i 'trt_engine_cache_enable|1 trt_timing_cache_enable|1 trt_dump_ep_context_model|1 trt_weightless_engine_enable|1' model_ctx.onnx\r\n```","shortMessageHtmlLink":"[Build] Propagate build option for CUDA minimal to TRT (microsoft#20695)"}},{"before":"631a2c16be3ebb6043a0d276ad53d756604636e6","after":"8c2689877fb48bfb4a6a133b020cdb5ec7c9b066","ref":"refs/heads/main","pushedAt":"2024-07-01T17:55:33.000Z","pushType":"push","commitsCount":65,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"CoreML: Disable 1D ML Program matmul due to bug in coreml (#21186)\n\n### Description\r\nDisable using CoreML ML Program for a matmul where one of the inputs is\r\n1D as the CoreML implementation appears to be broken. See\r\nhttps://github.com/apple/coremltools/issues/2263\r\n\r\nAdd some debugging notes. \r\n\r\n### Motivation and Context\r\nFix failing test on macos-14.","shortMessageHtmlLink":"CoreML: Disable 1D ML Program matmul due to bug in coreml (microsoft#…"}},{"before":"c5a5cc1157515c7c937a09ad29371f4b1267c966","after":"8fcd74c2eac6ee8bd1f3ea1398d2f5c1aa7b2494","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-20T23:10:27.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"lintrunner sigh","shortMessageHtmlLink":"lintrunner sigh"}},{"before":"fff6723af0753514e21ab83f4fde955a974846d0","after":"c5a5cc1157515c7c937a09ad29371f4b1267c966","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-20T18:13:10.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"updated variable name + finds absolute path for external data file for deletion","shortMessageHtmlLink":"updated variable name + finds absolute path for external data file fo…"}},{"before":"0f939744fedb12e990407c278f9f391bff52cfaa","after":"fff6723af0753514e21ab83f4fde955a974846d0","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-20T07:29:28.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"switched to setting absolute path in constructor for temp onnx file path\n& relative path for external data","shortMessageHtmlLink":"switched to setting absolute path in constructor for temp onnx file path"}},{"before":"44acd08776bab8513b67f4bfd7df84853e9be758","after":"0f939744fedb12e990407c278f9f391bff52cfaa","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-19T18:33:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"comments","shortMessageHtmlLink":"comments"}},{"before":"c27820effc2bb62ae5f805aa2ed5b87f8ee57f2a","after":"44acd08776bab8513b67f4bfd7df84853e9be758","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-19T18:31:41.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"changed release method to __del__ & added comments to explain that generate_artifacts typically will throw an error if unsuccessful","shortMessageHtmlLink":"changed release method to __del__ & added comments to explain that ge…"}},{"before":"378afa8a09f903b39268ed0ed17629e522c01700","after":"c27820effc2bb62ae5f805aa2ed5b87f8ee57f2a","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-18T22:22:13.000Z","pushType":"push","commitsCount":8,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"Merge branch 'main' into carzh/generate-artifacts-large-models","shortMessageHtmlLink":"Merge branch 'main' into carzh/generate-artifacts-large-models"}},{"before":"a53f6928327631a02c834c4c1f6e9a5f2aa1565c","after":"631a2c16be3ebb6043a0d276ad53d756604636e6","ref":"refs/heads/main","pushedAt":"2024-06-18T20:31:39.000Z","pushType":"push","commitsCount":51,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"[js/web] skip default locateFile() when dynamic import is disabled (#21073)\n\n### Description\r\nskip default `locateFile()` when dynamic import is disabled. This allows\r\nthe file to work with bundlers to load WebAssembly file correctly if\r\n`env.wasm.wasmPaths` is not set.","shortMessageHtmlLink":"[js/web] skip default locateFile() when dynamic import is disabled (m…"}},{"before":"72a55c7a1a6d264aa5aa068dce26c9df95aa0264","after":"378afa8a09f903b39268ed0ed17629e522c01700","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-17T21:01:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"updated handling of temp files + added comment on crossentropyloss update","shortMessageHtmlLink":"updated handling of temp files + added comment on crossentropyloss up…"}},{"before":"a35bc8049f6ab429e687da213890972a88ab7a2e","after":"72a55c7a1a6d264aa5aa068dce26c9df95aa0264","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-13T07:03:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"removed some print statements + cleaned up some code","shortMessageHtmlLink":"removed some print statements + cleaned up some code"}},{"before":"29ad02a084a0fbd17f3d83549823c65f89e8cc3b","after":"a35bc8049f6ab429e687da213890972a88ab7a2e","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-12T23:04:13.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"lintrunner format","shortMessageHtmlLink":"lintrunner format"}},{"before":"8069700f7a4fcdfac468a7fa7f5c88ab6c400399","after":"29ad02a084a0fbd17f3d83549823c65f89e8cc3b","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-12T22:48:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"fine I'll use X|Y type annotations","shortMessageHtmlLink":"fine I'll use X|Y type annotations"}},{"before":"924cedae736df48e10a36dc46f333530a3985235","after":"8069700f7a4fcdfac468a7fa7f5c88ab6c400399","ref":"refs/heads/carzh/generate-artifacts-large-models","pushedAt":"2024-06-12T21:35:30.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"removed unnecessary helper functions + import from unit tests","shortMessageHtmlLink":"removed unnecessary helper functions + import from unit tests"}},{"before":"a05464f65ff126287fdc9c7db8230e15d2bd8153","after":"42e117b424cf47d2a9e4cccebe18379ec528a6a3","ref":"refs/heads/carzh/test-generate-artifacts-large-models","pushedAt":"2024-06-12T19:12:30.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"removed graph utis helper for value info","shortMessageHtmlLink":"removed graph utis helper for value info"}},{"before":"7b9e7e474bd8a2c6ef0cf9a8fa8f3242d2e99c5e","after":"a05464f65ff126287fdc9c7db8230e15d2bd8153","ref":"refs/heads/carzh/test-generate-artifacts-large-models","pushedAt":"2024-06-12T19:08:22.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"removed value info + added unit test","shortMessageHtmlLink":"removed value info + added unit test"}},{"before":"6d33b76a1da501e33ddb1b5cc1453e979bfc56f9","after":"7b9e7e474bd8a2c6ef0cf9a8fa8f3242d2e99c5e","ref":"refs/heads/carzh/test-generate-artifacts-large-models","pushedAt":"2024-06-11T21:28:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"wrote two unit tests + changed implementation to write to a temp file then load so that the shape inference is performed on the updated graph + tried adding value info for output nodes for softmaxcrossentropy loss","shortMessageHtmlLink":"wrote two unit tests + changed implementation to write to a temp file…"}},{"before":"924cedae736df48e10a36dc46f333530a3985235","after":"6d33b76a1da501e33ddb1b5cc1453e979bfc56f9","ref":"refs/heads/carzh/test-generate-artifacts-large-models","pushedAt":"2024-06-07T20:56:45.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"made model_path optional + added 2GB check in generate_artifacts","shortMessageHtmlLink":"made model_path optional + added 2GB check in generate_artifacts"}},{"before":null,"after":"924cedae736df48e10a36dc46f333530a3985235","ref":"refs/heads/carzh/test-generate-artifacts-large-models","pushedAt":"2024-06-07T20:55:44.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"switched to passing both the modelproto + modelpath to the model accessor, so that onnx.load is called twice instead of three times","shortMessageHtmlLink":"switched to passing both the modelproto + modelpath to the model acce…"}},{"before":"52874f628a14ce971470995fbe9c15512f40de5b","after":"a53f6928327631a02c834c4c1f6e9a5f2aa1565c","ref":"refs/heads/main","pushedAt":"2024-06-07T18:55:30.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"},"commit":{"message":"Update c-api-noopenmp-packaging-pipelines.yml: remove CUDA version parameter (#20955)\n\n### Description\r\nUpdate c-api-noopenmp-packaging-pipelines.yml: remove CUDA version\r\nparameter\r\nTo reduce confusion. This pipeline is for generating CUDA 11 packages.\r\nJust it. Not CUDA 12.\r\n\r\n### Motivation and Context\r\nIn the last release we accidentally published CUDA 12(instead of CUDA\r\n11) packages to nuget.org.\r\nWe also tried to publish CUDA 12 packages to\r\nhttps://aiinfra.visualstudio.com/PublicPackages/_artifacts/feed/ORT-Nightly.\r\nLuckily it didn't go through because a package with the same version\r\nnumber already existed there. Every time when someone runs this pipeline\r\nwith CUDA version set to 12, the built packages will be published to\r\nhttps://aiinfra.visualstudio.com/PublicPackages/_artifacts/feed/ORT-Nightly.\r\nAnd GenAI team's build pipelines are based on the nightly packages. So\r\nsometimes GenAI team builds their packages with CUDA 12 and sometimes\r\nwith CUDA 11, which is very random.\r\nTherefore, please limit the use of pipeline parameters. Most Azure\r\nDevOps yml files are template files. They should use parameters. But the\r\ntop level yml files should be more careful on that.","shortMessageHtmlLink":"Update c-api-noopenmp-packaging-pipelines.yml: remove CUDA version pa…"}},{"before":"b39c7ce422b9ffe1e0b928b7ef15ad202df0aa1a","after":null,"ref":"refs/heads/carzh/optional_shape_inference","pushedAt":"2024-06-06T23:02:43.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"carzh","name":"Caroline Zhu","path":"/carzh","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22922935?s=80&v=4"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"startCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0yNFQxNzo0Mzo1NS4wMDAwMDBazwAAAAS_ddWk","endCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNi0wNlQyMzowMjo0My4wMDAwMDBazwAAAARer4zp"}},"title":"Activity · carzh/onnxruntime"}