fix: end to end job had a series of erros

.github/workflows/instructlab-knowledge-e2e.yml

@@ -8,11 +8,10 @@ on:
     pull_request:
         branches:
           - "main"
-        types: [opened, reopened]
 
 env:
     # dependencies that need to be installed using pip for testing to work
-    TESTING_PIP_DEPENDENCIES: "papermill nbformat"
+    TESTING_PIP_DEPENDENCIES: "papermill nbformat ipykernel"
 
 jobs:
     end_to_end:
@@ -23,18 +22,19 @@ jobs:
         runs-on: ubuntu-latest
         env:
             # customize the workflow here
-            API_KEY: ""
             API_URL: "http://127.0.0.1:11434/v1"
             MODEL_ID: "Mixtral-8x7B" # must be open-AI compatible when using inference mock
         steps:
             - uses: actions/checkout@v3
             - name: Set up Python
-              uses: actions/setup-python@v3
+              uses: actions/setup-python@v5
               with:
                 python-version: '3.11'
                 cache: pip
             - name: Install Testing Tools
-              run: pip install ${{ env.TESTING_PIP_DEPENDENCIES }}
+              run: |
+                  pip install ${{ env.TESTING_PIP_DEPENDENCIES }}
+                  ipython kernel install --name "python3" --user
             - name: Validate The Notebook
               working-directory: ./notebooks/instructlab-knowledge
               run: |
@@ -44,15 +44,10 @@ jobs:
               working-directory: ./tests/inference-mock
               run: |
                 pip install -r requirements.txt
-                python python -m flask run &
+                nohup python app.py &
                 sleep 1
                 echo "Inference mock server started on port 11434"
             - name: Run End To End Tests
               working-directory: ./notebooks/instructlab-knowledge
               # NOTE: for this to work, cells with parameters need to be tagged as parameters in the target notebooks
-              run: papermill ${{ matrix.notebooks_to_test }} ${{ matrix.notebooks_to_test }}.tmp -p API_KEY $API_KEY -p API_URL $API_URL -p MODEL_ID $MODEL_ID
-            - name: Terminate Inference Mock Server
-              if: always()
-              run: |
-                pkill -f "python -m flask"
-                echo "Mock server terminated"
+              run: papermill ${{ matrix.notebooks_to_test }} ${{ matrix.notebooks_to_test }}.tmp -p API_URL $API_URL -p MODEL_ID $MODEL_ID

notebooks/instructlab-knowledge/instructlab-knowledge.ipynb

@@ -367,14 +367,14 @@
    },
    "outputs": [],
    "source": [
-    "API_KEY = \"\"  # the API access key for your account ( empty if not required )\n",
+    "API_KEY = \"none\"  # the API access key for your account ( cannot be empty )\n",
     "API_URL = \"http://127.0.0.1:11434/v1\"  # the URL of your model's API\n",
     "MODEL_ID = \"granite3.3\" # the name of your model"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "id": "b702267e-f550-4bc2-bce4-c0fcecbbd292",
    "metadata": {},
    "outputs": [],
@@ -426,15 +426,13 @@
     "import random #TODO: replace random sampling with subset selection\n",
     "\n",
     "for doc, chunks in dataset.items(): # TODO: multiple file support\n",
-    "    generate_options.generated_file = AUTHORING_OUTPUT_DIR / f\"qagen-{doc}.json\" \n",
+    "    generate_options.generated_file = AUTHORING_OUTPUT_DIR / f\"qagen-{doc}.json\"\n",
     "    gen = Generator(generate_options=generate_options)\n",
-    "    \n",
     "    print(f\"processing chunks that looks like:\\n{chunks[0].text}\")\n",
     "    selected_chunks = random.sample(chunks, NUM_CHUNKS_TO_SELECT_FOR_AUTHORING)\n",
     "    print(f\"Selected {len(selected_chunks)} contexts\")\n",
     "\n",
     "    Path.unlink(generate_options.generated_file, missing_ok=True)\n",
-    "    breakpoint()\n",
     "    results = gen.generate_from_chunks(selected_chunks) # automatically saves to file\n",
     "\n",
     "    print(f\"{doc}: {results.status}\")\n",

tests/inference-mock/app.py

@@ -58,7 +58,6 @@ def mock_generate():
     response = jsonify(completion_response(completion, request.json['model']))
     app.logger.debug(f"response: {response}")
     return response
-
 
 if __name__ == '__main__':
     app.run(debug=True, port=port_number)