some minor patches

.github/actions/nm-install-test-whl/action.yml

@@ -44,14 +44,12 @@ runs:
         pip3 install coverage
         pip3 install pytest-cov
         pip3 install pytest-xdist
-        pip3 install --index-url http://${{ inputs.pypi }}:8080/ --trusted-host ${{ inputs.pypi }} nm-magic-wand-nightly
-        pip3 list
+        pip3 install -r requirements-dev.txt
         BASE=$(./.github/scripts/convert-version ${{ inputs.python }})
         WHL=$(find . -type f -iname "*${BASE}*.whl")
         WHL_BASENAME=$(basename ${WHL})
         echo "whl=${WHL_BASENAME}" >> "$GITHUB_OUTPUT"
-        pip3 install ${WHL}
-        pip3 install -r requirements-dev.txt
+        pip3 install ${WHL}[sparse]
         # report magic_wand version
         MAGIC_WAND=$(pip3 show nm-magic-wand-nightly | grep "Version" | cut -d' ' -f2)
         echo "magic_wand=${MAGIC_WAND}" >> "$GITHUB_OUTPUT"

.github/actions/nm-install-whl/action.yml

@@ -19,7 +19,7 @@ runs:
         COMMIT=${{ github.sha }}
         VENV="${{ env.VENV_BASE }}-${COMMIT:0:7}"
         source $(pyenv root)/versions/${{ inputs.python }}/envs/${VENV}/bin/activate
-        pip3 install -r requirements-cuda.txt -r requirements-dev.txt
+        pip3 install -r requirements-dev.txt
         BASE=$(./.github/scripts/convert-version ${{ inputs.python }})
         WHL=$(find . -type f -iname "*${BASE}*.whl")
         WHL_BASENAME=$(basename ${WHL})

.github/workflows/build-test.yml

@@ -172,6 +172,7 @@ jobs:
             timeout: ${{ inputs.benchmark_timeout }}
             gitref: ${{ github.ref }}
             python: ${{ inputs.python }}
+            whl: ${{ needs.BUILD.outputs.whl }}
             # Always push if it is a scheduled job
             push_benchmark_results_to_gh_pages: "${{ github.event_name == 'schedule' || inputs.push_benchmark_results_to_gh_pages }}"
         secrets: inherit

.github/workflows/nm-benchmark.yml

@@ -23,6 +23,10 @@ on:
         description: "python version, e.g. 3.10.12"
         type: string
         required: true
+      whl:
+        description: "whl to test (variable appears late binding so unusable outside 'download artifact')"
+        type: string
+        required: true
       push_benchmark_results_to_gh_pages:
         description: "When set to true, the workflow pushes all benchmarking results to gh-pages UI"
         type: string
@@ -51,6 +55,10 @@ on:
         description: "python version, e.g. 3.10.12"
         type: string
         required: true
+      whl:
+        description: "whl to test (variable appears late binding so unusable outside 'download artifact')"
+        type: string
+        required: true
       push_benchmark_results_to_gh_pages:
         description: "When set to true, the workflow pushes all benchmarking results to gh-pages UI"
         type: choice
@@ -85,8 +93,8 @@ jobs:
         uses: ./.github/actions/nm-set-env/
         with:
           hf_token: ${{ secrets.NM_HF_TOKEN }}
-          Gi_per_thread: ${{ inputs.Gi_per_thread }}
-          nvcc_threads: ${{ inputs.nvcc_threads }}
+          Gi_per_thread: 1
+          nvcc_threads: 0
 
       - name: set python
         id: set_python

.github/workflows/test.yml

@@ -82,6 +82,7 @@ jobs:
               with:
                 hf_token: ${{ secrets.NM_HF_TOKEN }}
                 Gi_per_thread: 1
+                nvcc_threads: 0
 
             - name: set python
               id: set_python

neuralmagic/benchmarks/requirements-benchmark.txt

@@ -2,4 +2,3 @@
 requests
 aiohttp
 datasets
-nm-magic-wand-nightly