Port changes from Milestone-5

.gitignore

@@ -188,5 +188,3 @@ cython_debug/
 # PyPI configuration file
 .pypirc
 .github/copilot-instructions.md
-
-docs/img/

Dockerfile

@@ -10,6 +10,7 @@ ENV PYTHONDONTWRITEBYTECODE=1 \
 # Install system dependencies
 RUN apt-get update && apt-get install -y \
     git \
+    dos2unix \
     && rm -rf /var/lib/apt/lists/*
 
 # Create a non-root user
@@ -21,12 +22,24 @@ WORKDIR /app
 # Copy requirements first for caching
 COPY requirements.txt .
 
-# Install dependencies
-RUN pip install --no-cache-dir -r requirements.txt
+# Remove -e . from requirements.txt to avoid installing the project before copying it
+# and install dependencies
+RUN sed -i '/-e \./d' requirements.txt && \
+    pip install --no-cache-dir -r requirements.txt
 
 # Copy the rest of the application
 COPY --chown=user:user . .
 
+# Ensure the user has permissions on the app directory (needed for dvc init if .dvc is missing)
+RUN chown -R user:user /app
+
+# Fix line endings and permissions for the start script
+RUN dos2unix scripts/start_space.sh && \
+    chmod +x scripts/start_space.sh
+
+# Install the project itself
+RUN pip install --no-cache-dir .
+
 # Make start script executable
 RUN chmod +x scripts/start_space.sh
 

hopcroft_skill_classification_tool_competition/config.py

@@ -30,10 +30,10 @@ EMBEDDING_MODEL_NAME = "all-MiniLM-L6-v2"
 # API Configuration - which model to use for predictions
 API_CONFIG = {
     # Model file to load (without path, just filename)
-    "model_name": "random_forest_embedding_gridsearch.pkl",
+    "model_name": "random_forest_tfidf_gridsearch.pkl",
     # Feature type: "tfidf" or "embedding"
     # This determines how text is transformed before prediction
-    "feature_type": "embedding",
+    "feature_type": "tfidf",
 }
 
 # Training configuration

hopcroft_skill_classification_tool_competition/main.py

@@ -194,6 +194,8 @@ async def predict_skills(issue: IssueInput) -> PredictionRecord:
         )
 
     except Exception as e:
+        import traceback
+        traceback.print_exc()
         raise HTTPException(
             status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
             detail=f"Prediction failed: {str(e)}",

requirements.txt

@@ -13,6 +13,7 @@ seaborn
 
 # Data versioning
 dvc
+dvc-s3
 mlflow==2.16.0
 protobuf==4.25.3
 
@@ -25,6 +26,7 @@ fastapi[standard]>=0.115.0
 pydantic>=2.0.0
 uvicorn>=0.30.0
 httpx>=0.27.0
+streamlit>=1.28.0
 
 # Development tools
 ipython

scripts/start_space.sh

@@ -3,6 +3,13 @@
 # Fail on error
 set -e
 
+# Ensure DVC is initialized (in case .dvc folder was not copied)
+if [ ! -d ".dvc" ]; then
+    echo "Initializing DVC..."
+    dvc init --no-scm
+    dvc remote add -d origin https://dagshub.com/se4ai2526-uniba/Hopcroft.dvc
+fi
+
 # Determine credentials
 # Prefer specific DAGSHUB vars, fallback to MLFLOW vars (often the same for DagsHub)
 USER=${DAGSHUB_USERNAME:-$MLFLOW_TRACKING_USERNAME}
@@ -20,9 +27,17 @@ fi
 
 echo "Pulling models from DVC..."
 # Pull only the necessary files for inference
-dvc pull models/random_forest_tfidf_gridsearch.pkl \
-         models/tfidf_vectorizer.pkl \
-         models/label_names.pkl
+dvc pull models/random_forest_tfidf_gridsearch.pkl.dvc \
+         models/tfidf_vectorizer.pkl.dvc \
+         models/label_names.pkl.dvc
+
+echo "Starting FastAPI application in background..."
+uvicorn hopcroft_skill_classification_tool_competition.main:app --host 0.0.0.0 --port 8000 &
+
+# Wait for API to start
+echo "Waiting for API to start..."
+sleep 10
 
-echo "Starting FastAPI application..."
-uvicorn hopcroft_skill_classification_tool_competition.main:app --host 0.0.0.0 --port 7860
+echo "Starting Streamlit application..."
+export API_BASE_URL="http://localhost:8000"
+streamlit run hopcroft_skill_classification_tool_competition/streamlit_app.py --server.port 7860 --server.address 0.0.0.0