From ae8d9c100e77d1dcdbee0cfd8763d1f3d19514a9 Mon Sep 17 00:00:00 2001 From: Akos Hadnagy Date: Tue, 26 Aug 2025 20:40:09 +0000 Subject: [PATCH 1/5] Add collated reports job to Nvidia CI --- .github/workflows/self-nightly-caller.yml | 1 + .github/workflows/self-scheduled.yml | 15 +++++++++++++++ 2 files changed, 16 insertions(+) diff --git a/.github/workflows/self-nightly-caller.yml b/.github/workflows/self-nightly-caller.yml index 48edaf8187fe..4950eb43276e 100644 --- a/.github/workflows/self-nightly-caller.yml +++ b/.github/workflows/self-nightly-caller.yml @@ -49,6 +49,7 @@ jobs: slack_report_channel: "#transformers-ci-past-future" docker: huggingface/transformers-all-latest-torch-nightly-gpu ci_event: Nightly CI + runner_type: "a10" report_repo_id: hf-internal-testing/transformers_daily_ci_with_torch_nightly commit_sha: ${{ github.event.workflow_run.head_sha || github.sha }} secrets: inherit diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml index b5f6685d30e7..bd39a2b95da7 100644 --- a/.github/workflows/self-scheduled.yml +++ b/.github/workflows/self-scheduled.yml @@ -31,6 +31,9 @@ on: commit_sha: required: false type: string + runner_type: + required: false + type: string env: @@ -542,3 +545,15 @@ jobs: report_repo_id: ${{ inputs.report_repo_id }} secrets: inherit + + collated_reports: + name: Collated Reports + if: ${{ always() }} + needs: run_models_gpu + uses: huggingface/transformers/.github/workflows/collated-reports.yml@main + with: + job: run_models_gpu + report_repo_id: ${{ inputs.report_repo_id }} + gpu_name: ${{ inputs.runner_type }} + machine_type: ${{ inputs.machine_type == '1gpu' && 'single-gpu' || 'multi-gpu' }} + secrets: inherit From 244e17bbf94137a09b84284aee8a74a862740e27 Mon Sep 17 00:00:00 2001 From: Akos Hadnagy Date: Tue, 26 Aug 2025 20:46:17 +0000 Subject: [PATCH 2/5] machine_type --- .github/workflows/self-scheduled.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml index bd39a2b95da7..fa38ed15a5ed 100644 --- a/.github/workflows/self-scheduled.yml +++ b/.github/workflows/self-scheduled.yml @@ -555,5 +555,5 @@ jobs: job: run_models_gpu report_repo_id: ${{ inputs.report_repo_id }} gpu_name: ${{ inputs.runner_type }} - machine_type: ${{ inputs.machine_type == '1gpu' && 'single-gpu' || 'multi-gpu' }} + machine_type: ${{ matrix.machine_type }} secrets: inherit From 4be041d3a49ea4773fd29b267553f7c46a167bb9 Mon Sep 17 00:00:00 2001 From: Akos Hadnagy Date: Thu, 28 Aug 2025 07:34:32 +0000 Subject: [PATCH 3/5] Move collated reports job to model_jobs --- .github/workflows/model_jobs.yml | 15 +++++++++++++++ .github/workflows/self-scheduled.yml | 13 +------------ 2 files changed, 16 insertions(+), 12 deletions(-) diff --git a/.github/workflows/model_jobs.yml b/.github/workflows/model_jobs.yml index 6ca019e7c93f..3d22bff3823c 100644 --- a/.github/workflows/model_jobs.yml +++ b/.github/workflows/model_jobs.yml @@ -25,6 +25,9 @@ on: required: false default: run_models_gpu type: string + runner_type: + required: false + type: string env: HF_HOME: /mnt/cache @@ -143,3 +146,15 @@ jobs: with: name: ${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ env.matrix_folders }}_test_reports path: /transformers/reports/${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports + + collated_reports: + name: Collated Reports + if: ${{ always() }} + needs: run_models_gpu + uses: huggingface/transformers/.github/workflows/collated-reports.yml@main + with: + job: run_models_gpu + report_repo_id: ${{ inputs.report_repo_id }} + gpu_name: ${{ inputs.runner_type }} + machine_type: ${{ inputs.machine_type }} + secrets: inherit diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml index 5d095cba4eb9..7c3be3616ea5 100644 --- a/.github/workflows/self-scheduled.yml +++ b/.github/workflows/self-scheduled.yml @@ -129,6 +129,7 @@ jobs: runner_map: ${{ needs.setup.outputs.runner_map }} docker: ${{ inputs.docker }} commit_sha: ${{ inputs.commit_sha || github.sha }} + runner_type: ${{ inputs.runner_type }} secrets: inherit run_trainer_and_fsdp_gpu: @@ -548,15 +549,3 @@ jobs: report_repo_id: ${{ inputs.report_repo_id }} secrets: inherit - - collated_reports: - name: Collated Reports - if: ${{ always() }} - needs: run_models_gpu - uses: huggingface/transformers/.github/workflows/collated-reports.yml@main - with: - job: run_models_gpu - report_repo_id: ${{ inputs.report_repo_id }} - gpu_name: ${{ inputs.runner_type }} - machine_type: ${{ matrix.machine_type }} - secrets: inherit From 5fe4258f092442c8ecf86db921141d5907b55832 Mon Sep 17 00:00:00 2001 From: Akos Hadnagy Date: Thu, 28 Aug 2025 10:07:31 +0000 Subject: [PATCH 4/5] Propagate repo id variable --- .github/workflows/model_jobs.yml | 3 +++ .github/workflows/self-scheduled.yml | 1 + 2 files changed, 4 insertions(+) diff --git a/.github/workflows/model_jobs.yml b/.github/workflows/model_jobs.yml index 3d22bff3823c..7e30cde735fa 100644 --- a/.github/workflows/model_jobs.yml +++ b/.github/workflows/model_jobs.yml @@ -28,6 +28,9 @@ on: runner_type: required: false type: string + report_repo_id: + required: false + type: string env: HF_HOME: /mnt/cache diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml index 7c3be3616ea5..a5dbc9d59a82 100644 --- a/.github/workflows/self-scheduled.yml +++ b/.github/workflows/self-scheduled.yml @@ -130,6 +130,7 @@ jobs: docker: ${{ inputs.docker }} commit_sha: ${{ inputs.commit_sha || github.sha }} runner_type: ${{ inputs.runner_type }} + report_repo_id: ${{ inputs.report_repo_id }} secrets: inherit run_trainer_and_fsdp_gpu: From acf9a6b3721a4d61f8dc22faa6a213bd1831b65c Mon Sep 17 00:00:00 2001 From: Akos Hadnagy Date: Tue, 2 Sep 2025 07:40:56 +0000 Subject: [PATCH 5/5] assifgn runner_type is self-scheduled-caller --- .github/workflows/self-nightly-caller.yml | 1 - .github/workflows/self-scheduled-caller.yml | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/self-nightly-caller.yml b/.github/workflows/self-nightly-caller.yml index 4950eb43276e..48edaf8187fe 100644 --- a/.github/workflows/self-nightly-caller.yml +++ b/.github/workflows/self-nightly-caller.yml @@ -49,7 +49,6 @@ jobs: slack_report_channel: "#transformers-ci-past-future" docker: huggingface/transformers-all-latest-torch-nightly-gpu ci_event: Nightly CI - runner_type: "a10" report_repo_id: hf-internal-testing/transformers_daily_ci_with_torch_nightly commit_sha: ${{ github.event.workflow_run.head_sha || github.sha }} secrets: inherit diff --git a/.github/workflows/self-scheduled-caller.yml b/.github/workflows/self-scheduled-caller.yml index d709c562251e..78c7f3c60f23 100644 --- a/.github/workflows/self-scheduled-caller.yml +++ b/.github/workflows/self-scheduled-caller.yml @@ -52,6 +52,7 @@ jobs: slack_report_channel: "#transformers-ci-daily-models" docker: huggingface/transformers-all-latest-gpu ci_event: Daily CI + runner_type: "a10" report_repo_id: hf-internal-testing/transformers_daily_ci commit_sha: ${{ github.sha }} secrets: inherit