Gaudi3 CI (#38790)
This commit is contained in:
Ilyas Moutawwakil
committed by
GitHub
GitHub
parent
2166b6b4ff
commit
984ff89e73
121
.github/workflows/model_jobs_intel_gaudi.yml
vendored
Normal file
121
.github/workflows/model_jobs_intel_gaudi.yml
vendored
Normal file
@@ -0,0 +1,121 @@
|
||||
name: model jobs
|
||||
|
||||
on:
|
||||
workflow_call:
|
||||
inputs:
|
||||
folder_slices:
|
||||
required: true
|
||||
type: string
|
||||
slice_id:
|
||||
required: true
|
||||
type: number
|
||||
runner:
|
||||
required: true
|
||||
type: string
|
||||
machine_type:
|
||||
required: true
|
||||
type: string
|
||||
report_name_prefix:
|
||||
required: false
|
||||
default: run_models_gpu
|
||||
type: string
|
||||
|
||||
env:
|
||||
RUN_SLOW: yes
|
||||
PT_HPU_LAZY_MODE: 0
|
||||
TRANSFORMERS_IS_CI: yes
|
||||
PT_ENABLE_INT64_SUPPORT: 1
|
||||
HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
|
||||
SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
|
||||
HF_HOME: /mnt/cache/.cache/huggingface
|
||||
|
||||
jobs:
|
||||
run_models_gpu:
|
||||
name: " "
|
||||
strategy:
|
||||
max-parallel: 8
|
||||
fail-fast: false
|
||||
matrix:
|
||||
folders: ${{ fromJson(inputs.folder_slices)[inputs.slice_id] }}
|
||||
runs-on:
|
||||
group: ${{ inputs.runner }}
|
||||
container:
|
||||
image: vault.habana.ai/gaudi-docker/1.21.1/ubuntu22.04/habanalabs/pytorch-installer-2.6.0:latest
|
||||
options: --runtime=habana
|
||||
-v /mnt/cache/.cache/huggingface:/mnt/cache/.cache/huggingface
|
||||
--env OMPI_MCA_btl_vader_single_copy_mechanism=none
|
||||
--env HABANA_VISIBLE_DEVICES
|
||||
--env HABANA_VISIBLE_MODULES
|
||||
--cap-add=sys_nice
|
||||
--shm-size=64G
|
||||
steps:
|
||||
- name: Echo input and matrix info
|
||||
shell: bash
|
||||
run: |
|
||||
echo "${{ inputs.folder_slices }}"
|
||||
echo "${{ matrix.folders }}"
|
||||
echo "${{ toJson(fromJson(inputs.folder_slices)[inputs.slice_id]) }}"
|
||||
|
||||
- name: Echo folder ${{ matrix.folders }}
|
||||
shell: bash
|
||||
run: |
|
||||
echo "${{ matrix.folders }}"
|
||||
matrix_folders=${{ matrix.folders }}
|
||||
matrix_folders=${matrix_folders/'models/'/'models_'}
|
||||
echo "$matrix_folders"
|
||||
echo "matrix_folders=$matrix_folders" >> $GITHUB_ENV
|
||||
|
||||
- name: Checkout
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
pip install -e .[testing,torch] "numpy<2.0.0" scipy scikit-learn
|
||||
|
||||
- name: HL-SMI
|
||||
run: |
|
||||
hl-smi
|
||||
echo "HABANA_VISIBLE_DEVICES=${HABANA_VISIBLE_DEVICES}"
|
||||
echo "HABANA_VISIBLE_MODULES=${HABANA_VISIBLE_MODULES}"
|
||||
|
||||
- name: Environment
|
||||
run: python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: pip freeze
|
||||
|
||||
- name: Set `machine_type` for report and artifact names
|
||||
shell: bash
|
||||
run: |
|
||||
if [ "${{ inputs.machine_type }}" = "1gaudi" ]; then
|
||||
machine_type=single-gpu
|
||||
elif [ "${{ inputs.machine_type }}" = "2gaudi" ]; then
|
||||
machine_type=multi-gpu
|
||||
else
|
||||
machine_type=${{ inputs.machine_type }}
|
||||
fi
|
||||
echo "machine_type=$machine_type" >> $GITHUB_ENV
|
||||
|
||||
- name: Run all tests on Gaudi
|
||||
run: python3 -m pytest -v --make-reports=${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports tests/${{ matrix.folders }}
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat reports/${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports/failures_short.txt
|
||||
|
||||
- name: Run test
|
||||
shell: bash
|
||||
run: |
|
||||
mkdir -p reports/${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports
|
||||
echo "hello" > reports/${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports/hello.txt
|
||||
echo "${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports"
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ env.matrix_folders }}_test_reports
|
||||
path: reports/${{ env.machine_type }}_${{ inputs.report_name_prefix }}_${{ matrix.folders }}_test_reports
|
||||
345
.github/workflows/self-scheduled-intel-gaudi.yml
vendored
Normal file
345
.github/workflows/self-scheduled-intel-gaudi.yml
vendored
Normal file
@@ -0,0 +1,345 @@
|
||||
name: Self-hosted runner (scheduled-intel-gaudi)
|
||||
|
||||
on:
|
||||
workflow_call:
|
||||
inputs:
|
||||
job:
|
||||
required: true
|
||||
type: string
|
||||
slack_report_channel:
|
||||
required: true
|
||||
type: string
|
||||
runner_scale_set:
|
||||
required: true
|
||||
type: string
|
||||
ci_event:
|
||||
required: true
|
||||
type: string
|
||||
report_repo_id:
|
||||
required: true
|
||||
type: string
|
||||
|
||||
env:
|
||||
NUM_SLICES: 2
|
||||
RUN_SLOW: yes
|
||||
PT_HPU_LAZY_MODE: 0
|
||||
TRANSFORMERS_IS_CI: yes
|
||||
PT_ENABLE_INT64_SUPPORT: 1
|
||||
HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
|
||||
SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
|
||||
HF_HOME: /mnt/cache/.cache/huggingface
|
||||
|
||||
jobs:
|
||||
setup:
|
||||
if: contains(fromJSON('["run_models_gpu", "run_trainer_and_fsdp_gpu"]'), inputs.job)
|
||||
name: Setup
|
||||
runs-on: ubuntu-latest
|
||||
outputs:
|
||||
slice_ids: ${{ steps.set-matrix.outputs.slice_ids }}
|
||||
folder_slices: ${{ steps.set-matrix.outputs.folder_slices }}
|
||||
quantization_matrix: ${{ steps.set-matrix.outputs.quantization_matrix }}
|
||||
steps:
|
||||
- name: Checkout
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Set up Python
|
||||
uses: actions/setup-python@v5
|
||||
with:
|
||||
python-version: "3.10"
|
||||
|
||||
- id: set-matrix
|
||||
if: contains(fromJSON('["run_models_gpu", "run_trainer_and_fsdp_gpu"]'), inputs.job)
|
||||
name: Identify models to test
|
||||
working-directory: tests
|
||||
run: |
|
||||
if [ "${{ inputs.job }}" = "run_models_gpu" ]; then
|
||||
echo "folder_slices=$(python3 ../utils/split_model_tests.py --num_splits ${{ env.NUM_SLICES }})" >> $GITHUB_OUTPUT
|
||||
echo "slice_ids=$(python3 -c 'd = list(range(${{ env.NUM_SLICES }})); print(d)')" >> $GITHUB_OUTPUT
|
||||
elif [ "${{ inputs.job }}" = "run_trainer_and_fsdp_gpu" ]; then
|
||||
echo "folder_slices=[['trainer'], ['fsdp']]" >> $GITHUB_OUTPUT
|
||||
echo "slice_ids=[0, 1]" >> $GITHUB_OUTPUT
|
||||
fi
|
||||
|
||||
- id: set-matrix-quantization
|
||||
if: ${{ inputs.job == 'run_quantization_torch_gpu' }}
|
||||
name: Identify quantization method to test
|
||||
working-directory: tests
|
||||
run: |
|
||||
echo "quantization_matrix=$(python3 -c 'import os; tests = os.getcwd(); quantization_tests = os.listdir(os.path.join(tests, "quantization")); d = sorted(list(filter(os.path.isdir, [f"quantization/{x}" for x in quantization_tests]))) ; print(d)')" >> $GITHUB_OUTPUT
|
||||
|
||||
run_models_gpu:
|
||||
if: ${{ inputs.job == 'run_models_gpu' }}
|
||||
name: " "
|
||||
needs: setup
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [1gaudi, 2gaudi]
|
||||
slice_id: ${{ fromJSON(needs.setup.outputs.slice_ids) }}
|
||||
uses: ./.github/workflows/model_jobs_intel_gaudi.yml
|
||||
with:
|
||||
slice_id: ${{ matrix.slice_id }}
|
||||
machine_type: ${{ matrix.machine_type }}
|
||||
folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
runner: ${{ inputs.runner_scale_set }}-${{ matrix.machine_type }}
|
||||
report_name_prefix: run_models_gpu
|
||||
|
||||
secrets: inherit
|
||||
|
||||
run_trainer_and_fsdp_gpu:
|
||||
if: ${{ inputs.job == 'run_trainer_and_fsdp_gpu' }}
|
||||
name: " "
|
||||
needs: setup
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [1gaudi, 2gaudi]
|
||||
slice_id: ${{ fromJSON(needs.setup.outputs.slice_ids) }}
|
||||
uses: ./.github/workflows/model_jobs_intel_gaudi.yml
|
||||
with:
|
||||
slice_id: ${{ matrix.slice_id }}
|
||||
machine_type: ${{ matrix.machine_type }}
|
||||
folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
runner: ${{ inputs.runner_scale_set }}-${{ matrix.machine_type }}
|
||||
report_name_prefix: run_trainer_and_fsdp_gpu
|
||||
|
||||
secrets: inherit
|
||||
|
||||
run_pipelines_gpu:
|
||||
if: ${{ inputs.job == 'run_pipelines_gpu' }}
|
||||
name: Pipelines
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [1gaudi, 2gaudi]
|
||||
runs-on:
|
||||
group: ${{ inputs.runner_scale_set }}-${{ matrix.machine_type }}
|
||||
container:
|
||||
image: vault.habana.ai/gaudi-docker/1.21.1/ubuntu22.04/habanalabs/pytorch-installer-2.6.0:latest
|
||||
options: --runtime=habana
|
||||
-v /mnt/cache/.cache/huggingface:/mnt/cache/.cache/huggingface
|
||||
--env OMPI_MCA_btl_vader_single_copy_mechanism=none
|
||||
--env HABANA_VISIBLE_DEVICES
|
||||
--env HABANA_VISIBLE_MODULES
|
||||
--cap-add=sys_nice
|
||||
--shm-size=64G
|
||||
steps:
|
||||
- name: Checkout
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
pip install -e .[testing,torch] "numpy<2.0.0" scipy scikit-learn librosa soundfile
|
||||
|
||||
- name: HL-SMI
|
||||
run: |
|
||||
hl-smi
|
||||
echo "HABANA_VISIBLE_DEVICES=${HABANA_VISIBLE_DEVICES}"
|
||||
echo "HABANA_VISIBLE_MODULES=${HABANA_VISIBLE_MODULES}"
|
||||
|
||||
- name: Environment
|
||||
run: python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: pip freeze
|
||||
|
||||
- name: Set `machine_type` for report and artifact names
|
||||
shell: bash
|
||||
run: |
|
||||
if [ "${{ matrix.machine_type }}" = "1gaudi" ]; then
|
||||
machine_type=single-gpu
|
||||
elif [ "${{ matrix.machine_type }}" = "2gaudi" ]; then
|
||||
machine_type=multi-gpu
|
||||
else
|
||||
machine_type=${{ matrix.machine_type }}
|
||||
fi
|
||||
echo "machine_type=$machine_type" >> $GITHUB_ENV
|
||||
|
||||
- name: Run all pipeline tests on Intel Gaudi
|
||||
run: |
|
||||
python3 -m pytest -v --make-reports=${{ env.machine_type }}_run_pipelines_gpu_test_reports tests/pipelines -m "not not_device_test"
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: |
|
||||
cat reports/${{ env.machine_type }}_run_pipelines_gpu_test_reports/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ env.machine_type }}_run_pipelines_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ env.machine_type }}_run_pipelines_gpu_test_reports
|
||||
path: reports/${{ env.machine_type }}_run_pipelines_gpu_test_reports
|
||||
|
||||
run_examples_gpu:
|
||||
if: ${{ inputs.job == 'run_examples_gpu' }}
|
||||
name: Examples directory
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [1gaudi]
|
||||
runs-on:
|
||||
group: ${{ inputs.runner_scale_set }}-${{ matrix.machine_type }}
|
||||
container:
|
||||
image: vault.habana.ai/gaudi-docker/1.21.1/ubuntu22.04/habanalabs/pytorch-installer-2.6.0:latest
|
||||
options: --runtime=habana
|
||||
-v /mnt/cache/.cache/huggingface:/mnt/cache/.cache/huggingface
|
||||
--env OMPI_MCA_btl_vader_single_copy_mechanism=none
|
||||
--env HABANA_VISIBLE_DEVICES
|
||||
--env HABANA_VISIBLE_MODULES
|
||||
--cap-add=sys_nice
|
||||
--shm-size=64G
|
||||
steps:
|
||||
- name: Checkout
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
pip install -e .[testing,torch] "numpy<2.0.0" scipy scikit-learn librosa soundfile
|
||||
|
||||
- name: HL-SMI
|
||||
run: |
|
||||
hl-smi
|
||||
echo "HABANA_VISIBLE_DEVICES=${HABANA_VISIBLE_DEVICES}"
|
||||
echo "HABANA_VISIBLE_MODULES=${HABANA_VISIBLE_MODULES}"
|
||||
|
||||
- name: Environment
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: |
|
||||
pip freeze
|
||||
|
||||
- name: Set `machine_type` for report and artifact names
|
||||
shell: bash
|
||||
run: |
|
||||
if [ "${{ matrix.machine_type }}" = "1gaudi" ]; then
|
||||
machine_type=single-gpu
|
||||
elif [ "${{ matrix.machine_type }}" = "2gaudi" ]; then
|
||||
machine_type=multi-gpu
|
||||
else
|
||||
machine_type=${{ matrix.machine_type }}
|
||||
fi
|
||||
echo "machine_type=$machine_type" >> $GITHUB_ENV
|
||||
|
||||
- name: Run examples tests on Intel Gaudi
|
||||
run: |
|
||||
pip install -r examples/pytorch/_tests_requirements.txt
|
||||
python3 -m pytest -v --make-reports=${{ env.machine_type }}_run_examples_gpu_test_reports examples/pytorch -m "not not_device_test"
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: |
|
||||
cat reports/${{ env.machine_type }}_run_examples_gpu_test_reports/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ env.machine_type }}_run_examples_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ env.machine_type }}_run_examples_gpu_test_reports
|
||||
path: reports/${{ env.machine_type }}_run_examples_gpu_test_reports
|
||||
|
||||
run_deepspeed_gpu:
|
||||
if: ${{ inputs.job == 'run_deepspeed_gpu' }}
|
||||
name: Intel Gaudi deepspeed tests
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [1gaudi, 2gaudi]
|
||||
runs-on:
|
||||
group: ${{ inputs.runner_scale_set }}-${{ matrix.machine_type }}
|
||||
container:
|
||||
image: vault.habana.ai/gaudi-docker/1.21.1/ubuntu22.04/habanalabs/pytorch-installer-2.6.0:latest
|
||||
options: --runtime=habana
|
||||
-v /mnt/cache/.cache/huggingface:/mnt/cache/.cache/huggingface
|
||||
--env OMPI_MCA_btl_vader_single_copy_mechanism=none
|
||||
--env HABANA_VISIBLE_DEVICES
|
||||
--env HABANA_VISIBLE_MODULES
|
||||
--cap-add=sys_nice
|
||||
--shm-size=64G
|
||||
steps:
|
||||
- name: Checkout
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 0
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
pip install -e .[testing,torch] "numpy<2.0.0" scipy scikit-learn librosa soundfile
|
||||
pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.20.0
|
||||
|
||||
- name: HL-SMI
|
||||
run: |
|
||||
hl-smi
|
||||
echo "HABANA_VISIBLE_DEVICES=${HABANA_VISIBLE_DEVICES}"
|
||||
echo "HABANA_VISIBLE_MODULES=${HABANA_VISIBLE_MODULES}"
|
||||
|
||||
- name: Environment
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: |
|
||||
pip freeze
|
||||
|
||||
- name: Set `machine_type` for report and artifact names
|
||||
shell: bash
|
||||
run: |
|
||||
if [ "${{ matrix.machine_type }}" = "1gaudi" ]; then
|
||||
machine_type=single-gpu
|
||||
elif [ "${{ matrix.machine_type }}" = "2gaudi" ]; then
|
||||
machine_type=multi-gpu
|
||||
else
|
||||
machine_type=${{ matrix.machine_type }}
|
||||
fi
|
||||
echo "machine_type=$machine_type" >> $GITHUB_ENV
|
||||
|
||||
- name: Run all deepspeed tests on intel Gaudi
|
||||
run: |
|
||||
python3 -m pytest -v --make-reports=${{ env.machine_type }}_run_deepspeed_gpu_test_reports tests/deepspeed -m "not not_device_test"
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: |
|
||||
cat reports/${{ env.machine_type }}_run_deepspeed_gpu_test_reports/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ env.machine_type }}_run_deepspeed_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ env.machine_type }}_run_deepspeed_gpu_test_reports
|
||||
path: reports/${{ env.machine_type }}_run_deepspeed_gpu_test_reports
|
||||
|
||||
send_results:
|
||||
name: Slack Report
|
||||
needs:
|
||||
[
|
||||
setup,
|
||||
run_models_gpu,
|
||||
run_examples_gpu,
|
||||
run_pipelines_gpu,
|
||||
run_deepspeed_gpu,
|
||||
run_trainer_and_fsdp_gpu,
|
||||
]
|
||||
if: ${{ always() }}
|
||||
uses: ./.github/workflows/slack-report.yml
|
||||
with:
|
||||
job: ${{ inputs.job }}
|
||||
setup_status: ${{ needs.setup.result }}
|
||||
slack_report_channel: ${{ inputs.slack_report_channel }}
|
||||
quantization_matrix: ${{ needs.setup.outputs.quantization_matrix }}
|
||||
folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
report_repo_id: ${{ inputs.report_repo_id }}
|
||||
ci_event: ${{ inputs.ci_event }}
|
||||
|
||||
secrets: inherit
|
||||
67
.github/workflows/self-scheduled-intel-gaudi3-caller.yml
vendored
Normal file
67
.github/workflows/self-scheduled-intel-gaudi3-caller.yml
vendored
Normal file
@@ -0,0 +1,67 @@
|
||||
name: Self-hosted runner (Intel Gaudi3 scheduled CI caller)
|
||||
|
||||
on:
|
||||
repository_dispatch:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "17 2 * * *"
|
||||
|
||||
jobs:
|
||||
model-ci:
|
||||
name: Model CI
|
||||
uses: ./.github/workflows/self-scheduled-intel-gaudi.yml
|
||||
with:
|
||||
job: run_models_gpu
|
||||
ci_event: Scheduled CI (Intel) - Gaudi3
|
||||
runner_scale_set: itac-bm-emr-gaudi3-dell
|
||||
slack_report_channel: "#transformers-ci-daily-intel-gaudi3"
|
||||
report_repo_id: optimum-intel/transformers_daily_ci_intel_gaudi3
|
||||
|
||||
secrets: inherit
|
||||
|
||||
pipeline-ci:
|
||||
name: Pipeline CI
|
||||
uses: ./.github/workflows/self-scheduled-intel-gaudi.yml
|
||||
with:
|
||||
job: run_pipelines_gpu
|
||||
ci_event: Scheduled CI (Intel) - Gaudi3
|
||||
runner_scale_set: itac-bm-emr-gaudi3-dell
|
||||
slack_report_channel: "#transformers-ci-daily-intel-gaudi3"
|
||||
report_repo_id: optimum-intel/transformers_daily_ci_intel_gaudi3
|
||||
|
||||
secrets: inherit
|
||||
|
||||
example-ci:
|
||||
name: Example CI
|
||||
uses: ./.github/workflows/self-scheduled-intel-gaudi.yml
|
||||
with:
|
||||
job: run_examples_gpu
|
||||
ci_event: Scheduled CI (Intel) - Gaudi3
|
||||
runner_scale_set: itac-bm-emr-gaudi3-dell
|
||||
slack_report_channel: "#transformers-ci-daily-intel-gaudi3"
|
||||
report_repo_id: optimum-intel/transformers_daily_ci_intel_gaudi3
|
||||
|
||||
secrets: inherit
|
||||
|
||||
deepspeed-ci:
|
||||
name: DeepSpeed CI
|
||||
uses: ./.github/workflows/self-scheduled-intel-gaudi.yml
|
||||
with:
|
||||
job: run_deepspeed_gpu
|
||||
ci_event: Scheduled CI (Intel) - Gaudi3
|
||||
runner_scale_set: itac-bm-emr-gaudi3-dell
|
||||
slack_report_channel: "#transformers-ci-daily-intel-gaudi3"
|
||||
report_repo_id: optimum-intel/transformers_daily_ci_intel_gaudi3
|
||||
|
||||
secrets: inherit
|
||||
|
||||
trainer-fsdp-ci:
|
||||
name: Trainer/FSDP CI
|
||||
uses: ./.github/workflows/self-scheduled-intel-gaudi.yml
|
||||
with:
|
||||
job: run_trainer_and_fsdp_gpu
|
||||
ci_event: Scheduled CI (Intel) - Gaudi3
|
||||
runner_scale_set: itac-bm-emr-gaudi3-dell
|
||||
slack_report_channel: "#transformers-ci-daily-intel-gaudi3"
|
||||
report_repo_id: optimum-intel/transformers_daily_ci_intel_gaudi3
|
||||
secrets: inherit
|
||||
Reference in New Issue
Block a user