Repository navigation

Fix FFN depth-scaled initialization target #18738

	name: CPU Unit Test

	on:
	push:
	branches: [ main ]
	paths-ignore:
	- 'torchtitan/experiments/**'
	pull_request:
	paths-ignore:
	- 'torchtitan/experiments/**'

	concurrency:
	group: unit-test${{ github.workflow }}-${{ github.ref == 'refs/heads/main' && github.run_number \|\| github.ref }}
	cancel-in-progress: true

	permissions:
	id-token: write
	contents: read

	jobs:
	set-matrix:
	uses: ./.github/workflows/set-matrix.yaml

	build-test:
	needs: set-matrix
	uses: pytorch/test-infra/.github/workflows/linux_job_v3.yml@main
	with:
	runner: mt-l-x86iavx512-8-64
	docker-image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/torchtitan/torchtitan-ubuntu-22.04-clang12:${{ needs.set-matrix.outputs.docker-hash }}
	repository: pytorch/torchtitan
	script: \|
	set -eux

	# The generic Linux job chooses to use base env, not the one setup by the image
	CONDA_ENV=$(conda env list --json \| jq -r ".envs \| .[-1]")
	conda activate "${CONDA_ENV}"

	pip config --user set global.progress_bar off

	pip install --force-reinstall --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/cpu

	USE_CPP=0 python -m pip install --pre torchao --index-url https://download.pytorch.org/whl/nightly/cpu

	# Write coverage data/report to /tmp (world-writable): the mounted workspace and
	# RUNNER_TEMP are owned by the host uid, which the container user (ci-user) can't write.
	export COVERAGE_FILE=/tmp/.coverage
	pytest tests/unit_tests/cpu --cov=. --cov-report=xml:/tmp/coverage.xml --durations=20 -vv