Update inference.py

inference.py

@@ -1,63 +1,63 @@
-# inference.py
-import joblib
-import numpy as np
-import os
-
-class EagleBlendPredictor:
-    def __init__(self, model_dir=r"C:\Users\Otto Henry\CodingWorld\Hackerthon\Fuel-Blend-Properties-Prediction\dev\models"):
-        """
-        Load fitted scaler, PCA transformer, and trained XGBoost multioutput model.
-        """
-        self.model_dir = model_dir
-        
-        # Load scaler
-        scaler_path = os.path.join(model_dir, "scaler.joblib")
-        self.scaler = joblib.load(scaler_path)
-
-        # Load PCA
-        pca_path = os.path.join(model_dir, "pca.joblib") 
-        self.pca = joblib.load(pca_path)
-
-        # Load trained model
-        model_path = os.path.join(model_dir, "xmodel.joblib")
-        self.model = joblib.load(model_path)
-
-    def predict_all(self, X_new):
-        """
-        Make predictions on new data using the trained scaler, PCA, and model.
-        
-        Parameters:
-        - X_new: array-like of shape (n_samples, n_features)
-        
-        Returns:
-        - predictions: numpy array of shape (n_samples, n_outputs)
-        """
-        # Convert input to NumPy array
-        X_new = np.array(X_new)
-
-        # Step 1: Scale data
-        X_scaled = self.scaler.transform(X_new)
-
-        # Step 2: PCA transform
-        X_pca = self.pca.transform(X_scaled)
-
-        # Step 3: Predict
-        predictions = self.model.predict(X_pca)
-
-        return predictions
-
-
-# if __name__ == "__main__":
-#     # Example usage
-#     # Create the inference object
-#     predictor = EagleBlendPredictor(model_dir="models")
-
-#     # Example new data (must have same number of features as training data)
-#     sample_data = [
-#         [0.5, 1.2, 3.3, 4.1, 5.5],  # Replace with actual feature values
-#         [1.5, 2.1, 0.3, 4.5, 2.5]
-#     ]
-
-#     # Get predictions
-#     preds = predictor.predict_all(sample_data)
-#     print("Predictions:\n", preds)
+# inference.py
+import joblib
+import numpy as np
+import os
+
+class EagleBlendPredictor:
+    def __init__(self, model_dir="Models"):
+        """
+        Load fitted scaler, PCA transformer, and trained XGBoost multioutput model.
+        """
+        self.model_dir = model_dir
+        
+        # Load scaler
+        scaler_path = os.path.join(model_dir, "scaler.joblib")
+        self.scaler = joblib.load(scaler_path)
+
+        # Load PCA
+        pca_path = os.path.join(model_dir, "pca.joblib") 
+        self.pca = joblib.load(pca_path)
+
+        # Load trained model
+        model_path = os.path.join(model_dir, "xmodel.joblib")
+        self.model = joblib.load(model_path)
+
+    def predict_all(self, X_new):
+        """
+        Make predictions on new data using the trained scaler, PCA, and model.
+        
+        Parameters:
+        - X_new: array-like of shape (n_samples, n_features)
+        
+        Returns:
+        - predictions: numpy array of shape (n_samples, n_outputs)
+        """
+        # Convert input to NumPy array
+        X_new = np.array(X_new)
+
+        # Step 1: Scale data
+        X_scaled = self.scaler.transform(X_new)
+
+        # Step 2: PCA transform
+        X_pca = self.pca.transform(X_scaled)
+
+        # Step 3: Predict
+        predictions = self.model.predict(X_pca)
+
+        return predictions
+
+
+# if __name__ == "__main__":
+#     # Example usage
+#     # Create the inference object
+#     predictor = EagleBlendPredictor(model_dir="models")
+
+#     # Example new data (must have same number of features as training data)
+#     sample_data = [
+#         [0.5, 1.2, 3.3, 4.1, 5.5],  # Replace with actual feature values
+#         [1.5, 2.1, 0.3, 4.5, 2.5]
+#     ]
+
+#     # Get predictions
+#     preds = predictor.predict_all(sample_data)
+#     print("Predictions:\n", preds)