Update main.py

main.py

@@ -27,22 +27,16 @@ print("model size ====> :", file_size.st_size, "bytes")
 
 # Assume 'model' is already defined somewhere else
 
+@app.route('/compare', methods=['POST'])
 @app.route('/compare', methods=['POST'])
 def compare():
-    data = request.json
-
-    # Validation and data extraction
-    if 'jobs_skills' not in data or 'employee_skills' not in data:
-        return jsonify({"error": "Missing 'jobs_skills' or 'employee_skills' in request"}), 400
+    employee_skills = request.json.get('jobs_skills')  
+    jobs_skills = request.json.get('employee_skills')
 
-    jobs_skills = data['jobs_skills']
-    employee_skills = data['employee_skills']
 
+    # Validation
     if not isinstance(jobs_skills, list) or not all(isinstance(skill, str) for skill in jobs_skills):
-        return jsonify({"error": "'jobs_skills' must be a list of strings"}), 400
-
-    if not isinstance(employee_skills, list) or not all(isinstance(skill, str) for skill in employee_skills):
-        return jsonify({"error": "'employee_skills' must be a list of strings"}), 400
+        raise ValueError("jobs_skills must be a list of strings")
 
     # Encoding skills into embeddings
     job_embeddings = model.encode(jobs_skills)
@@ -54,10 +48,9 @@ def compare():
 
     for i, job_e in enumerate(job_embeddings):
         job_e_tensor = torch.from_numpy(job_e).unsqueeze(0)
-        similarity_score = cosine_similarity(employee_embeddings_tensor, job_e_tensor, dense_output=True)
+        similarity_score = cosine_similarity(employee_embeddings_tensor, job_e_tensor, dim=1)
         similarity_scores.append({"job": jobs_skills[i], "similarity_score": similarity_score.item()})
 
     return jsonify(similarity_scores)
-
 if __name__ == '__main__':
-    app.run(debug=True)
+    app.run()