Delete index.py

index.py

@@ -1,55 +0,0 @@
-from flask import Flask, request, jsonify
-from flask_cors import CORS
-import joblib
-import pandas as pd
-import os
-
-# Initialize the Flask app
-app = Flask(__name__)
-# This is the key change: it automatically handles CORS for all routes
-CORS(app)
-
-# --- Load the model and columns ---
-# Vercel copies the files to a temporary directory, so we find the path
-base_path = os.path.dirname(__file__)
-model_path = os.path.join(base_path, 'alumni_match_model.joblib')
-columns_path = os.path.join(base_path, 'model_feature_columns.joblib')
-
-model = joblib.load(model_path)
-model_columns = joblib.load(columns_path)
-
-
-@app.route('/api', methods=['POST'])
-def handle_request():
-    # Get the JSON data sent from the React app
-    incoming_data = request.get_json()
-
-    # --- PREPARE THE DATA FOR PREDICTION ---
-    df = pd.DataFrame([incoming_data])
-
-    def count_common_skills(row):
-        viewer_skills = set(str(row['viewer_skills']).lower().split('|'))
-        target_skills = set(str(row['target_skills']).lower().split('|'))
-        return len(viewer_skills.intersection(target_skills))
-
-    df['common_skills_count'] = df.apply(count_common_skills, axis=1)
-    df['branch_match'] = (df['viewer_branch'].str.lower() == df['target_branch'].str.lower()).astype(int)
-
-    for col in model_columns:
-        if col.startswith('company_'):
-            df[col] = 0
-
-    company_col_name = f"company_{incoming_data['target_company']}"
-    if company_col_name in df.columns:
-        df[company_col_name] = 1
-
-    # Reorder columns to match the model's training order
-    final_df = df[model_columns]
-
-    # --- MAKE THE PREDICTION ---
-    prediction_proba = model.predict_proba(final_df)
-    match_probability = prediction_proba[0][1]
-    final_score = round(match_probability * 10)
-
-    # Send the score back as a JSON response
-    return jsonify({'score': final_score})