feat(earnings): Add unique commentary for all content formats

api/main.py

@@ -995,33 +995,29 @@ Task Title: "{request.title}"
 @app.post("/predict/earning-opportunities", response_model=EarningOpportunityResponse, summary="Finds the best earning opportunities for an influencer")
 async def predict_earning_opportunities(request: EarningOpportunityRequest):
     """
-    [SMART COMMENTARY VERSION] Based on follower count, the model estimates a performance
-    score, and this function adds dynamic, helpful commentary.
+    [FINAL POLISHED VERSION] Uses the model for a score and adds dynamic, helpful
+    commentary for every content format.
     """
-    print(f"\n✅ Received request on /predict/earning-opportunities (SMART COMMENTARY)")
+    print(f"\n✅ Received request on /predict/earning-opportunities (FINAL POLISH)")
     if _earnings_optimizer is None or _earnings_encoder is None:
         raise HTTPException(status_code=503, detail="Earning Optimizer model or encoder is not available.")
 
     try:
+        # This part remains the same: preparing data and getting a score from the model
         scenarios_list = [
             {'campaign_niche': niche, 'content_format': c_format, 'follower_count': request.follower_count}
             for niche in ['Tech', 'Fashion', 'Food', 'Gaming', 'General']
             for c_format in ['Reel', 'Post', 'Story']
         ]
         df_scenarios = pd.DataFrame(scenarios_list)
-
-        print("   - Manually encoding data using saved encoder...")
         categorical_features = ['campaign_niche', 'content_format']
         encoded_cats = _earnings_encoder.transform(df_scenarios[categorical_features])
         encoded_df = pd.DataFrame(encoded_cats, columns=_earnings_encoder.get_feature_names_out(categorical_features))
-
         numerical_features = df_scenarios[['follower_count']].reset_index(drop=True)
         X_final_to_predict = pd.concat([encoded_df, numerical_features], axis=1)
-
-        print(f"   - Predicting scores for {len(X_final_to_predict)} scenarios...")
         predicted_scores = _earnings_optimizer.predict(X_final_to_predict)
 
-        # === ✨ THE SMART COMMENTARY LOGIC STARTS HERE ✨ ===
+        # === ✨ FINAL POLISH: MORE DYNAMIC COMMENTARY LOGIC ✨ ===
         results = []
         for i, scenario in enumerate(scenarios_list):
             score = float(predicted_scores[i])
@@ -1030,19 +1026,21 @@ async def predict_earning_opportunities(request: EarningOpportunityRequest):
             
             # Default commentary based on score
             if score > 0.75:
-                comment = "This is an excellent match! High potential for success."
+                comment = "Excellent match! This area has high potential for you."
             elif score < 0.4:
-                comment = "This might be a challenging area for you."
+                comment = "This could be a challenging area to grow in."
             else:
                 comment = "This is a solid opportunity worth exploring."
             
-            # Add dynamic, helpful tips based on format
+            # Add dynamic, helpful tips for EVERY format
             if c_format == 'Reel':
-                comment += " Reels are great for grabbing quick attention."
-            elif c_format == 'Post' and niche in ['Fashion', 'Food']:
-                comment += " A high-quality photo carousel could work very well here."
-            elif c_format == 'Story' and niche == 'Gaming':
-                comment += " Use Stories for interactive polls and behind-the-scenes content."
+                comment += " Reels are perfect for reaching a wider audience with trending audio."
+            elif c_format == 'Post':
+                # Ab yeh tip hamesha 'Post' ke saath aayegi
+                comment += " Use high-quality visuals and a strong caption for best results with posts."
+            elif c_format == 'Story':
+                # Ab yeh tip hamesha 'Story' ke saath aayegi
+                comment += " Stories are great for engaging your current followers with interactive polls or Q&As."
             
             results.append(Opportunity(
                 campaign_niche=niche,
@@ -1050,7 +1048,7 @@ async def predict_earning_opportunities(request: EarningOpportunityRequest):
                 estimated_score=score,
                 commentary=comment
             ))
-        # === ✨ END OF SMART COMMENTARY LOGIC ✨ ===
+        # === ✨ END OF FINAL POLISH ✨ ===
         
         sorted_results = sorted(results, key=lambda x: x.estimated_score, reverse=True)
         return EarningOpportunityResponse(opportunities=sorted_results[:5])
@@ -1059,6 +1057,7 @@ async def predict_earning_opportunities(request: EarningOpportunityRequest):
         print("🚨 An error occurred in /predict/earning-opportunities endpoint:")
         traceback.print_exc()
         raise HTTPException(status_code=500, detail=str(e))
+    
 
 @app.post("/predict/post-performance", response_model=PostPerformanceResponse, summary="Predicts likes and comments for a new post")
 async def predict_post_performance(request: PostPerformanceRequest):