Update app.py

app.py

@@ -9,13 +9,15 @@ from functools import lru_cache
 # CONFIG
 # ----------------------------
 JSON_FILE = "form-submissions-1.json"
-MODEL_ID = "HuggingFaceH4/sgpt-3.5-mini"
+# 🚩 FIX: Changed the model ID from an embedding model to a generative LLM.
+MODEL_ID = "google/flan-t5-large"
 HF_API_TOKEN = os.environ.get("HF_API_TOKEN")
 FILTERED_CSV = "/tmp/filtered_candidates.csv"
 OUTPUT_FILE = "/tmp/outputs.csv"
 BATCH_SIZE = 50
 
 if not HF_API_TOKEN:
+    # This check is good, but ensure the token is set in your environment (or space secrets)
     raise ValueError("HF_API_TOKEN not found in environment. Add it in Space Secrets.")
 
 CATEGORIES = {
@@ -54,9 +56,24 @@ Respond only 'Yes' if suitable, otherwise 'No'.
         )
         response.raise_for_status()
         result = response.json()
+
+        # Check for API error structure
         if isinstance(result, dict) and "error" in result:
+             print(f"LLM API Error: {result.get('error')}")
+             return "No"
+             
+        # Extract the generated text safely and clean it up
+        generated_text = result[0].get("generated_text", "No").strip().lower()
+        
+        # Flank-T5 often prepends the prompt or a part of it, so we only need the key decision word
+        # We look for 'yes' or 'no' anywhere in the response and prioritize 'yes' if found.
+        if "yes" in generated_text:
+            return "Yes"
+        elif "no" in generated_text:
+            return "No"
+        else:
             return "No"
-        return result[0].get("generated_text","No")
+            
     except Exception as e:
         print("LLM call failed:", e)
         return "No"
@@ -66,16 +83,29 @@ Respond only 'Yes' if suitable, otherwise 'No'.
 # ----------------------------
 def filter_by_roles(category_name):
     job_titles = CATEGORIES[category_name]
-    data = json.load(open(JSON_FILE, encoding="utf-8"))
+    # NOTE: Assuming 'form-submissions-1.json' exists in the current directory
+    try:
+        with open(JSON_FILE, encoding="utf-8") as f:
+            data = json.load(f)
+    except FileNotFoundError:
+        print(f"Error: JSON file '{JSON_FILE}' not found.")
+        return pd.DataFrame(), None
+        
     filtered = []
 
     for person in data:
         work_exps = person.get("work_experiences", [])
         if not work_exps:
             continue
-        non_fullstack_roles = [exp.get("roleName") for exp in work_exps if "full stack developer" not in exp.get("roleName","").lower()]
+        # Improved: Check if roleName is not None before calling .lower()
+        non_fullstack_roles = [
+            exp.get("roleName") for exp in work_exps 
+            if exp.get("roleName") and "full stack developer" not in exp.get("roleName").lower()
+        ]
         if not non_fullstack_roles:
             continue
+            
+        # Check for role match in the list of titles
         if any(role in job_titles for role in non_fullstack_roles):
             filtered.append({
                 "Name": person.get("name"),
@@ -101,21 +131,27 @@ def llm_recommendations(category_name):
     job_titles = CATEGORIES[category_name]
 
     if not os.path.exists(FILTERED_CSV):
-        return pd.DataFrame(), None
-
-    df_filtered = pd.read_csv(FILTERED_CSV)
-    df_filtered = df_filtered[df_filtered["Category"] == category_name]
+        # Re-run the filtering step if the CSV is missing
+        df_filtered, _ = filter_by_roles(category_name)
+        if df_filtered.empty:
+            return pd.DataFrame(), None
+    else:
+        df_filtered = pd.read_csv(FILTERED_CSV)
+        df_filtered = df_filtered[df_filtered["Category"] == category_name]
 
     if df_filtered.empty:
         return pd.DataFrame(), None
 
     recommended = []
-    filtered_candidates = df_filtered.to_dict(orient="records")
+    # Drop N/A values before converting to dict, otherwise json.dumps might fail if they are NaN
+    df_filtered_clean = df_filtered.fillna('N/A')
+    filtered_candidates = df_filtered_clean.to_dict(orient="records")
 
+    # The batching loop is fine, we will rely on the improved call_llm
     for i in range(0, len(filtered_candidates), BATCH_SIZE):
         batch = filtered_candidates[i:i+BATCH_SIZE]
         for person in batch:
-            # Only send necessary info
+            # Only send necessary info to save context length and cost
             candidate_info = {
                 "Name": person.get("Name"),
                 "Roles": person.get("Roles"),
@@ -123,7 +159,9 @@ def llm_recommendations(category_name):
             }
             candidate_str = json.dumps(candidate_info)
             response = call_llm(candidate_str, category_name, tuple(job_titles))
-            if response.strip().lower().startswith("yes"):
+            
+            # 🚩 IMPROVEMENT: The call_llm function now returns a clean 'Yes' or 'No'
+            if response == "Yes":
                 recommended.append(person)
 
     if not recommended:
@@ -133,9 +171,11 @@ def llm_recommendations(category_name):
     # Sort by numeric salary
     def parse_salary(s):
         try:
-            return float(s.replace("$","").replace(",",""))
+            # Remove currency symbols, commas, and convert to float
+            return float(str(s).replace("$","").replace(",","").replace("N/A", str(float('inf'))))
         except:
             return float('inf')
+            
     df_rec["Salary_sort"] = df_rec["Salary"].apply(parse_salary)
     df_rec = df_rec.sort_values("Salary_sort").drop(columns=["Salary_sort"])
     df_top5 = df_rec.head(5)
@@ -147,21 +187,30 @@ def llm_recommendations(category_name):
 # Show first 5 raw JSON candidates
 # ----------------------------
 def show_first_candidates():
-    data = json.load(open(JSON_FILE, encoding="utf-8"))
-    return pd.DataFrame(data[:5])
+    # NOTE: Assuming 'form-submissions-1.json' exists
+    try:
+        with open(JSON_FILE, encoding="utf-8") as f:
+             data = json.load(f)
+             return pd.DataFrame(data[:5])
+    except FileNotFoundError:
+        return pd.DataFrame({"Error": [f"JSON file '{JSON_FILE}' not found. Please ensure it is present."]})
+    except Exception as e:
+        return pd.DataFrame({"Error": [f"Failed to load JSON: {e}"]})
 
 # ----------------------------
 # Gradio interface
 # ----------------------------
 with gr.Blocks() as app:
-    gr.Markdown("### Raw JSON Preview: First 5 Candidates")
+    gr.Markdown("### Candidate Recommendation Engine")
+    
+    gr.Markdown("#### Raw JSON Preview: First 5 Candidates")
     gr.Dataframe(show_first_candidates(), label="First 5 JSON Entries")
 
     gr.Markdown("---")
     category_dropdown = gr.Dropdown(list(CATEGORIES.keys()), label="Select Category")
 
     # Step 1: Filter by roles
-    filter_button = gr.Button("Filter by Roles")
+    filter_button = gr.Button("1. Filter by Roles")
     filtered_df = gr.Dataframe(label="Filtered Candidates by Roles")
     download_filtered = gr.File(label="Download Filtered CSV", file_types=[".csv"])
     filter_button.click(filter_by_roles, inputs=[category_dropdown], outputs=[filtered_df, download_filtered])
@@ -169,10 +218,10 @@ with gr.Blocks() as app:
     gr.Markdown("---")
 
     # Step 2: LLM Recommendations
-    llm_button = gr.Button("Get LLM Recommendations")
+    llm_button = gr.Button("2. Get LLM Recommendations (Requires Step 1 to run first)")
     llm_df = gr.Dataframe(label="Top 5 Recommended Candidates")
     download_llm = gr.File(label="Download Recommendations CSV", file_types=[".csv"])
     llm_button.click(llm_recommendations, inputs=[category_dropdown], outputs=[llm_df, download_llm])
 
 if __name__ == "__main__":
-    app.launch()
+    app.launch()