Fallback to pytorch engine when the model is quantized by smooth quant #1116

	name: windows-x64-gpu
	on:
	push:
	paths:
	- '.github/workflows/windows-x64-gpu.yml'
	- 'src/**'
	- 'CMakeLists.txt'
	- 'cmake/**'
	- 'examples/**'
	- '3rdparty/**'
	- 'tests/csrc/**'
	pull_request:
	paths:
	- '.github/workflows/windows-x64-gpu.yml'
	- 'src/**'
	- 'CMakeLists.txt'
	- 'cmake/**'
	- 'examples/**'
	- '3rdparty/**'
	- 'tests/csrc/**'
	concurrency:
	group: windows-x64-gpu-${{ github.ref }}
	cancel-in-progress: true
	permissions:
	contents: read

	jobs:
	build:
	strategy:
	matrix:
	cudaver: [11.8.0, 12.1.0]
	name: cuda-${{ matrix.cudaver }}
	runs-on: windows-latest
	steps:
	- name: Checkout repository
	uses: actions/checkout@v3
	- name: Set up python
	uses: actions/setup-python@v4
	with:
	python-version: '3.8'
	- name: Install python packages
	run: \|
	pip install -r requirements/build.txt
	pip install wheel
	- name: Setup CUDA Toolkit
	id: cuda-toolkit
	shell: pwsh
	run: ./builder/windows/setup_cuda.ps1
	env:
	INPUT_CUDA_VERSION: ${{ matrix.cudaver }}
	- name: Build wheel
	run: \|
	$env:BUILD_TEST="ON"
	mkdir build
	cd build
	..\builder\windows\generate.ps1
	cmake --build . --config Release -- /m /v:q
	if (-Not $?) {
	echo "build failed"
	exit 1
	}
	cmake --install . --config Release
	cd ..
	rm build -Force -Recurse
	python setup.py bdist_wheel -d build/wheel

Provide feedback