Upload folder using huggingface_hub

Random_Forest_Model.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ff344efb433298a18c3abe1a7ef6baecc036bb16dc12bb1cfb90b7d3ab0d554
-size 86813571
+oid sha256:d39188f07d0b74e2aae8f6127bf6a6adf0b7d030d929eaabefbff0825484e7fa
+size 86814371

app.py

@@ -1,19 +1,53 @@
 # Backend_files/app.y
+import joblib
 import pandas as pd
 from flask import Flask, request, jsonify
 from flask_cors import CORS
-import joblib
-
+from sklearn.base import BaseEstimator, TransformerMixin
 # Import the classes so joblib can find them during unpickling
-from custom_transformers import StoreAgeAdder, OutlierCapper
-
-# Load the trained model
-Random_Forest_Loaded_Model = joblib.load('Random_Forest_Model.pkl')
-
+#from backend.files.custom_transformers import StoreAgeAdder, OutlierCapper
+
+class StoreAgeAdder(BaseEstimator, TransformerMixin):
+    def __init__(self, current_year=2025):
+        self.current_year = current_year
+
+    def fit(self, X, y=None):
+        return self
+
+    def transform(self, X):
+        X = X.copy()
+        X['store_age'] = self.current_year - X['Store_Establishment_Year']
+        return X.drop(columns='Store_Establishment_Year')
+
+class OutlierCapper(BaseEstimator, TransformerMixin):
+    def __init__(self, factor=1.5):
+        self.factor = factor
+        self.bounds = {}
+
+    def fit(self, X, y=None):
+        for col in X.columns:
+            Q1 = X[col].quantile(0.25)
+            Q3 = X[col].quantile(0.75)
+            IQR = Q3 - Q1
+            lower = Q1 - self.factor * IQR
+            upper = Q3 + self.factor * IQR
+            self.bounds[col] = (lower, upper)
+        return self
+
+    def transform(self, X):
+        X = X.copy()
+        for col in X.columns:
+            lower, upper = self.bounds[col]
+            X[col] = np.clip(X[col], lower, upper)
+        return X
+        
 # Initialize flas app with name
 Superkart_Sales_Predictor_API = Flask("Superkart Sales Predictor")
 CORS(Superkart_Sales_Predictor_API)  # Enable CORS for frontend integration (optional)
 
+# Load the trained model
+Random_Forest_Loaded_Model = joblib.load('Random_Forest_Model.pkl')
+
 # Define a route for the home page (GET request)
 @Superkart_Sales_Predictor_API.get('/')
 def home():
@@ -99,4 +133,4 @@ def predict_batch():
 
 # Run flask in debug mode
 if __name__ == '__main__':
-    Superkart_Sales_Predictor_API.run(debug=False, host='0.0.0.0', port=7860) 
+    Superkart_Sales_Predictor_API.run(debug=False, host='0.0.0.0', port=7860)

custom_transformer.py

@@ -0,0 +1,36 @@
+import numpy as np
+from sklearn.base import BaseEstimator, TransformerMixin
+
+class StoreAgeAdder(BaseEstimator, TransformerMixin):
+    def __init__(self, current_year=2025):
+        self.current_year = current_year
+
+    def fit(self, X, y=None):
+        return self
+
+    def transform(self, X):
+        X = X.copy()
+        X['store_age'] = self.current_year - X['Store_Establishment_Year']
+        return X.drop(columns='Store_Establishment_Year')
+
+class OutlierCapper(BaseEstimator, TransformerMixin):
+    def __init__(self, factor=1.5):
+        self.factor = factor
+        self.bounds = {}
+
+    def fit(self, X, y=None):
+        for col in X.columns:
+            Q1 = X[col].quantile(0.25)
+            Q3 = X[col].quantile(0.75)
+            IQR = Q3 - Q1
+            lower = Q1 - self.factor * IQR
+            upper = Q3 + self.factor * IQR
+            self.bounds[col] = (lower, upper)
+        return self
+
+    def transform(self, X):
+        X = X.copy()
+        for col in X.columns:
+            lower, upper = self.bounds[col]
+            X[col] = np.clip(X[col], lower, upper)
+        return X

requirements.txt

@@ -1,11 +1,10 @@
 Flask==3.1.1
 flask-cors==6.0.1
 geopandas==1.0.1
-joblib==1.5.1
+joblib==1.4.2
 pandas==2.2.2
 pandas-datareader==0.10.0
 pandas-gbq==0.29.1
 pandas-stubs==1.2.0.62
 scikit-learn==1.6.1
 sklearn-pandas==2.2.0
-gunicorn==21.2.0  # or latest