Cherry pick PR for rel-0.2.0 rc5 #380
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Enable GQA on CPU. It depends on ORT PR: microsoft/onnxruntime#20299. DO NOT MERGE this PR before that one.
…meration instead of dxcore (#377)
yufenglee
approved these changes
May 2, 2024
|
Do you want to update the version in this PR or a separate PR? |
yufenglee
approved these changes
May 2, 2024
Port over a fix from the onnxruntime-extensions tokenizer library to fix an invalid UTF-8 issue.
yufenglee
approved these changes
May 3, 2024
### Description This PR sets `pad_token_id` in `genai_config.json` to a single value when a model does not specify a pad token id but it specifies a list of EOS token ids. ### Motivation and Context When the pad token id is not specified, `pad_token_id` in `genai_config.json` stores the same value that `eos_token_id` in `genai_config.json` contains. When `eos_token_id` has a list of EOS token ids, then `pad_token_id` also has a list of pad token ids. This causes a parsing issue in ONNX Runtime GenAI because it expects only one pad token id. This PR also fixes [this issue](#384).
This pull request includes a change to the `nuget.config` file. The `nuget.org` package source has been removed and replaced with `ORT-Nightly` as the primary package source.
baijumeswani
force-pushed
the
baijumeswani/cherrypick-rc5
branch
from
356d000 to
050fd43
Compare
This pull request includes a change to the `nuget.config` file. The `nuget.org` package source has been removed and replaced with `ORT-Nightly` as the primary package source.
This pull request includes changes to multiple pipeline and job files, primarily in the `.pipelines` directory. The changes mainly involve removing the `publish_to_ado_feed` parameter and related steps, modifying the `ort_filename` value, and adding new parameters and steps for packaging and downloading operations. Changes to pipeline parameters: * [`.pipelines/nuget-publishing.yml`](diffhunk://#diff-e6ec5ea53c5d408941757974cb3e698481ef8f5bf36870475df104ef6e7b3277L30-L34): Removed `ort_dml_version` and `publish_to_ado_feed` parameters, and added a new `pipelines` section under `resources`. [[1]](diffhunk://#diff-e6ec5ea53c5d408941757974cb3e698481ef8f5bf36870475df104ef6e7b3277L30-L34) [[2]](diffhunk://#diff-e6ec5ea53c5d408941757974cb3e698481ef8f5bf36870475df104ef6e7b3277L45-R55) * [`.pipelines/pypl-publishing.yml`](diffhunk://#diff-e7c846e17ab5319529d839ddce86308159f05bc27ed172b501c1b996d00d838cL30-L34): Similar changes as in `nuget-publishing.yml`. [[1]](diffhunk://#diff-e7c846e17ab5319529d839ddce86308159f05bc27ed172b501c1b996d00d838cL30-L34) [[2]](diffhunk://#diff-e7c846e17ab5319529d839ddce86308159f05bc27ed172b501c1b996d00d838cL45-R55) Changes to job parameters: * [`.pipelines/stages/jobs/nuget-packaging-job.yml`](diffhunk://#diff-e86543a0294e755b065e28bb918be89e0e76957f3bb45725f351fd69bee7663fL16-L17): Removed `publish_to_ado_feed` parameter, added `os` parameter, and modified `ort_filename` value. [[1]](diffhunk://#diff-e86543a0294e755b065e28bb918be89e0e76957f3bb45725f351fd69bee7663fL16-L17) [[2]](diffhunk://#diff-e86543a0294e755b065e28bb918be89e0e76957f3bb45725f351fd69bee7663fR45-R59) * [`.pipelines/stages/jobs/py-packaging-job.yml`](diffhunk://#diff-f7337e936426594c6f395ba828d8edbfe1935e7bda0ef22ca42a5393f12d70b8L16-R16): Similar changes as in `nuget-packaging-job.yml`. [[1]](diffhunk://#diff-f7337e936426594c6f395ba828d8edbfe1935e7bda0ef22ca42a5393f12d70b8L16-R16) [[2]](diffhunk://#diff-f7337e936426594c6f395ba828d8edbfe1935e7bda0ef22ca42a5393f12d70b8R69-R83) Changes to steps: * [`.pipelines/stages/jobs/steps/capi-linux-step.yml`](diffhunk://#diff-b80db2c36162ca0823444533174a11beb8ce98d4ed7977c92c6c398473c02589R4-R9): Added `ep` and `arch` parameters, and modified steps for packaging and publishing artifacts. [[1]](diffhunk://#diff-b80db2c36162ca0823444533174a11beb8ce98d4ed7977c92c6c398473c02589R4-R9) [[2]](diffhunk://#diff-b80db2c36162ca0823444533174a11beb8ce98d4ed7977c92c6c398473c02589R35-R41) [[3]](diffhunk://#diff-b80db2c36162ca0823444533174a11beb8ce98d4ed7977c92c6c398473c02589R74) [[4]](diffhunk://#diff-b80db2c36162ca0823444533174a11beb8ce98d4ed7977c92c6c398473c02589R83-R98) * [`.pipelines/stages/jobs/steps/capi-win-step.yml`](diffhunk://#diff-27eab732db1fb7f28df66c1afc24b25200371487299e3b52c546e19fc158c016R8-R10): Similar changes as in `capi-linux-step.yml`. [[1]](diffhunk://#diff-27eab732db1fb7f28df66c1afc24b25200371487299e3b52c546e19fc158c016R8-R10) [[2]](diffhunk://#diff-27eab732db1fb7f28df66c1afc24b25200371487299e3b52c546e19fc158c016R37-R43) [[3]](diffhunk://#diff-27eab732db1fb7f28df66c1afc24b25200371487299e3b52c546e19fc158c016L86-R105) * [`.pipelines/stages/jobs/steps/utils/download-ort.yml`](diffhunk://#diff-b33e2529c1e7627822171d08975dbb8c52cdde1bcce3a5931a90ddf24c5019b1L8-R46): Modified parameters and steps for downloading and extracting files. Removed steps: * [`.pipelines/stages/jobs/steps/nuget-ado-feed-releasing-step.yml`](diffhunk://#diff-8ba1e9c7b359a5686f00bb59d92a476485b82a83d7dc717944e2a617f6eae1d4L1-L10): All steps removed. * [`.pipelines/stages/jobs/steps/py-ado-feed-releasing-step.yml`](diffhunk://#diff-c662251b0a6856bac6f84b6d21f507239e0787822e9a007c1c87da99c2df532bR1-R73): All steps removed. Other changes: * [`.pipelines/stages/jobs/steps/utils/capi-archive.yml`](diffhunk://#diff-50994eaa3e8cfdd2e3882713401cd849b4f1e383e36a23f05575a5a944c68f3cL23-R23): Modified file content in `steps`. * [`.pipelines/stages/jobs/steps/utils/download-ort-build.yml`](diffhunk://#diff-c662251b0a6856bac6f84b6d21f507239e0787822e9a007c1c87da99c2df532bR1-R73): New file added with parameters and steps for downloading and packaging operations. ---- Manually Input---- This PR will Allow ADO packing pipelines to use the specific ORT commit Ths PR also allow utilize the Cpack to do the zip packaging
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.