Update app.py

app.py

@@ -318,15 +318,16 @@ def evaluate_answers():
             [f"Question {i+1}: {entry['question']}\nCandidate's Answer {i+1}: {str(entry.get('response', '[No response provided]'))}" 
              for i, entry in enumerate(st.session_state["answers"])]
         )
+        #full_prompt_for_hr_evaluation = f"{hr_prompt_template}\n\nCandidate's Interview Answers:\n{candidate_responses_formatted_hr}"
         full_prompt_for_hr_evaluation = f"{hr_prompt_template}\n\nCandidate's Interview Answers (Consider all of these for holistic parameter scoring):\n{candidate_responses_formatted_hr}"
 
         try:
             response_content = model.generate_content(full_prompt_for_hr_evaluation)
             full_llm_response_text = response_content.text.strip()
-            
-            print("--- LLM Output for HR Score Extraction ---")
+            print("--- FULL LLM SOFT SKILLS RESPONSE ---")
             print(full_llm_response_text)
-            print("-----------------------------------------")
+            print("------ END RESPONSE ------")
+            print("--- AI Full Response for Soft Skills ---\n", full_llm_response_text, "\n------------------------")
 
             hr_parameter_scores_parsed_dict = {} # To store parsed scores for each HR param
             total_weighted_score_percentage = 0.0
@@ -357,8 +358,14 @@ def evaluate_answers():
                 total_weighted_score_percentage += (param_score / 5.0) * config_data["weight_normalized"] # Use normalized weight
             
             st.session_state["hr_parameter_scores_dict"] = hr_parameter_scores_parsed_dict # Store for table display
-            st.session_state["overall_score"] = round(total_weighted_score_percentage, 1) 
-            st.session_state["percentage_score"] = round(total_weighted_score_percentage, 1)
+
+            num_qs_in_session = len(st.session_state.get("answers", []))
+            max_possible_score = num_qs_in_session * 5.0  # Each Q worth 5
+            actual_score = (total_weighted_score_percentage / 100.0) * max_possible_score
+
+            st.session_state["overall_score"] = round(actual_score, 1)
+            st.session_state["percentage_score"] = round((actual_score / max_possible_score) * 100, 1)
+
             
             # Construct the feedback to be displayed: Parsed scores + Qualitative from LLM
             # The full_llm_response_text might still be useful if qualitative parsing is tricky
@@ -1002,10 +1009,45 @@ if st.session_state.get("show_summary", False):
     if num_qs_in_session == 0 and st.session_state.get("answers"): # Fallback if no generated_questions but answers exist
         num_qs_in_session = len(st.session_state.answers)
 
-    max_score_possible_for_session = num_qs_in_session * 5.0
+    if st.session_state["selected_domain"] == "Soft Skills":
+        num_qs_in_session = len(st.session_state.get("answers", []))
+        max_score_possible_for_session = num_qs_in_session * 5.0
+
+    else:
+        if st.session_state["selected_domain"] == "Soft Skills":
+            num_hr_params = len(st.session_state.get("hr_parameter_scores_dict", {}))
+            max_score_possible_for_session = num_hr_params * 5.0
+        else:
+            max_score_possible_for_session = num_qs_in_session * 5.0
+
+    #max_score_possible_for_session = num_qs_in_session * 5.0
     current_percentage_score = st.session_state.get('percentage_score', 0.0)
     current_overall_score = st.session_state.get('overall_score', 0.0)
 
+    if st.session_state["selected_domain"] == "Soft Skills":
+    hr_table_data = []
+    for param, config in HR_PARAMETERS_CONFIG.items():
+        score = st.session_state.get("hr_parameter_scores_dict", {}).get(param, 0.0)
+        weight_percent = config["weight_original"]
+        contribution = (score / 5.0) * config["weight_normalized"]
+        hr_table_data.append({
+            "Parameter": param,
+            "Weight (Original %)": f"{weight_percent}%",
+            "Score (1–5)": round(score, 1),
+            "Contribution to Final %": f"{contribution:.1f}%"
+        })
+
+    hr_table_data.append({
+        "Parameter": "Total",
+        "Weight (Original %)": "100%",
+        "Score (1–5)": "",
+        "Contribution to Final %": f"{current_percentage_score:.1f}%"
+    })
+
+    r_df = pd.DataFrame(hr_table_data)
+    st.markdown("### 🧾 Soft Skills Evaluation Breakdown")
+    st.dataframe(hr_df, use_container_width=True)
+
     # Display the calculated score and percentage bar first in a card
     st.markdown(f"""
         <div class='summary-card' style="margin-bottom: 20px;">