Fix tensor parallelism with SGMV to use true rank of the LoRA after splitting #30

	name: Server Tests

	on:
	pull_request:
	paths:
	- ".github/workflows/server_tests.yaml"
	- "server/**"
	- "proto/**"

	concurrency:
	group: ${{ github.workflow }}-${{ github.head_ref \|\| github.run_id }}
	cancel-in-progress: true

	jobs:
	run_tests:
	runs-on: ubuntu-latest

	steps:
	- uses: actions/checkout@v2
	- name: Set up Python
	uses: actions/setup-python@v1
	with:
	python-version: 3.9
	- name: Install Protoc
	uses: arduino/setup-protoc@v1
	- name: Filter test dependencies
	run: \|
	# remove stanford-stk from test requirements as it cannot install correctly without GPUs
	sed -i '/stanford-stk/d' server/requirements.txt
	sed -i '/stanford-stk/d' server/pyproject.toml
	- name: Install
	run: \|
	make install-server install-custom-kernels
	- name: Run server tests
	run: \|
	pip install pytest
	export HUGGING_FACE_HUB_TOKEN=${{ secrets.HUGGING_FACE_HUB_TOKEN }}
	pytest -s -vv server/tests

Provide feedback