Fix age range handling and model architecture

add_debug.js

@@ -0,0 +1,21 @@
+const fs = require('fs');
+
+// Read the file
+let content = fs.readFileSync('app.py', 'utf8');
+
+// Add more detailed debugging for model loading
+content = content.replace(
+  'if model_files_exist:',
+  'if model_files_exist:\n    print(f"🔍 Checking model files:")\n    print(f"   - MODEL_PATH: {MODEL_PATH} (exists: {os.path.exists(MODEL_PATH)})")\n    print(f"   - ENCODERS_DIR: {ENCODERS_DIR} (exists: {os.path.exists(ENCODERS_DIR)})")\n    print(f"   - PRODUCT_MASTER_PATH: {PRODUCT_MASTER_PATH} (exists: {os.path.exists(PRODUCT_MASTER_PATH)})")\n    \n    # Check file sizes\n    if os.path.exists(MODEL_PATH):\n        model_size = os.path.getsize(MODEL_PATH) / (1024*1024)  # MB\n        print(f"   - Model file size: {model_size:.2f} MB")\n    \n    if os.path.exists(ENCODERS_DIR):\n        encoder_files = [f for f in os.listdir(ENCODERS_DIR) if f.endswith(".json")]\n        print(f"   - Encoder files found: {encoder_files}")'
+);
+
+// Add more detailed error logging in the exception handler
+content = content.replace(
+  '    except Exception as e:\n        print(f"❌ Failed to load Yasai CID model: {e}")',
+  '    except Exception as e:\n        print(f"❌ Failed to load Yasai CID model: {e}")\n        import traceback\n        print(f"❌ Full traceback: {traceback.format_exc()}")'
+);
+
+// Write the updated content back
+fs.writeFileSync('app.py', content);
+
+console.log('Successfully added debugging information to app.py');

app.py

@@ -22,6 +22,19 @@ model_files_exist = all([
 ])
 
 if model_files_exist:
+    print(f"🔍 Checking model files:")
+    print(f"   - MODEL_PATH: {MODEL_PATH} (exists: {os.path.exists(MODEL_PATH)})")
+    print(f"   - ENCODERS_DIR: {ENCODERS_DIR} (exists: {os.path.exists(ENCODERS_DIR)})")
+    print(f"   - PRODUCT_MASTER_PATH: {PRODUCT_MASTER_PATH} (exists: {os.path.exists(PRODUCT_MASTER_PATH)})")
+    
+    # Check file sizes
+    if os.path.exists(MODEL_PATH):
+        model_size = os.path.getsize(MODEL_PATH) / (1024*1024)  # MB
+        print(f"   - Model file size: {model_size:.2f} MB")
+    
+    if os.path.exists(ENCODERS_DIR):
+        encoder_files = [f for f in os.listdir(ENCODERS_DIR) if f.endswith(".json")]
+        print(f"   - Encoder files found: {encoder_files}")
     try:
         from inference_yasai_cid import YasaiCIDInferenceEngine
         engine = YasaiCIDInferenceEngine(
@@ -32,6 +45,8 @@ if model_files_exist:
         print("✅ Yasai CID model loaded successfully!")
     except Exception as e:
         print(f"❌ Failed to load Yasai CID model: {e}")
+        import traceback
+        print(f"❌ Full traceback: {traceback.format_exc()}")
         engine = None
 else:
     print("⚠️  Model files not found. This is a template - add your model files to:")

fix_requirements.js

@@ -0,0 +1,15 @@
+const fs = require('fs');
+
+// Read the file
+let content = fs.readFileSync('requirements.txt', 'utf8');
+
+// Update rtdl to use a specific version that's compatible with Hugging Face
+content = content.replace(
+  'rtdl ',
+  'rtdl>=0.0.1rc0,<0.1.0'
+);
+
+// Write the updated content back
+fs.writeFileSync('requirements.txt', content);
+
+console.log('Successfully updated requirements.txt with rtdl version');

inference_yasai_cid.py

@@ -57,7 +57,7 @@ class YasaiCIDInferenceEngine:
 			return None
 		# Use training-script hyperparameters
 		model = FTTransformer.make_baseline(
-			n_num_features=8,
+			n_num_features=5,  # Updated: 5 numerical features (age ranges are now categorical)
 			cat_cardinalities=self.cat_cardinalities,
 			d_out=len(self.all_cids),
 			d_token=768,  # Use the actual saved model's d_token
@@ -106,9 +106,24 @@ class YasaiCIDInferenceEngine:
 				X_cat.append(self._encode_categorical(self.cat_encoders[col], '__UNKNOWN__'))
 		X_cat = torch.tensor([X_cat], dtype=torch.long)
 
-		# Numerical features (8 features to match training script)
-		num_cols = ['LAT', 'LONG', 'DELIVERY_NUM', 'MEDIAN_GENDER_RATIO', 'MODE_TOP_AGE_RANGE_1','TOTAL_VOLUME']
-		X_num = torch.tensor([[float(df.get(c, pd.Series([0])).iloc[0]) for c in num_cols]], dtype=torch.float32)
+		# Numerical features (5 features to match training script - age ranges are now categorical)
+		# Remove age range fields from numerical features since they're now categorical
+		num_cols = ['LAT', 'LONG', 'DELIVERY_NUM', 'MEDIAN_GENDER_RATIO', 'TOTAL_VOLUME']
+		X_num = []
+		for col in num_cols:
+			if col in df.columns:
+				try:
+					X_num.append(float(df[col].iloc[0]))
+				except (ValueError, TypeError):
+					X_num.append(0.0)
+			else:
+				# Provide default values for missing fields
+				if col == 'TOTAL_VOLUME':
+					X_num.append(0.0)  # Default total volume
+				else:
+					X_num.append(0.0)
+		
+		X_num = torch.tensor([X_num], dtype=torch.float32)
 		return X_cat, X_num
 
 	def predict(self, data: Dict) -> List[Dict]: