Upload folder using huggingface_hub

SuperKart_Sales_Prediction_Model.joblib

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:703e82419781df3324bbcb71e8815ac468fc4cba261bb7a157e02ed502ea5fd1
-size 473604
+oid sha256:9c298d589a2158eb513cb52191144518a2acab2cb0c04f1df14fca0f712fa4a1
+size 4

app.py

@@ -1,52 +1,55 @@
 
-import os
 import joblib
-from flask import Flask, request, jsonify
 import pandas as pd
 import numpy as np
-import warnings
-warnings.filterwarnings("ignore")
+from flask import Flask, request, jsonify
 
-# Define the path to the serialised model
-MODEL_PATH = "/content/Backend_files/SuperKart_Sales_Prediction_Model.joblib"
+app = Flask(__name__)
 
-# Load the trained model pipeline
-try:
-    model_pipeline = joblib.load(MODEL_PATH)
-    print(f"Model loaded successfully from {MODEL_PATH}")
-except Exception as e:
-    model_pipeline = None
-    print(f"Error loading model: {e}")
+# Define the path to the serialized model
+MODEL_PATH = "/content/Backend_files/SuperKart_Sales_Prediction_Model.joblib"
 
-# Initialize the Flask application
-app = Flask(__name__)
+#from joblib import load
 
-# Define a route for the home page
-@app.route("/", methods=["GET"])
-def home():
-    return "Welcome to the SuperKart Sales Prediction App!"
+model_pipeline = load("/content/Backend_files/SuperKart_Sales_Prediction_Model.joblib")
 
-# Define an endpoint for making predictions
-@app.route("/predict", methods=["POST"])
+@app.route('/predict', methods=['POST'])
 def predict():
-    if model_pipeline is None:
-        return jsonify({"error": "Model not loaded"}), 500
+    data = request.json
+    features = pd.DataFrame([data])
+    
+    prediction = model_pipeline.predict(features)
+    return jsonify({'prediction': prediction.tolist()})
+
+
+    # Get JSON request data
+    data = request.json
+    sample_df = pd.DataFrame([data])  # Convert JSON to DataFrame
+
+    # Feature Engineering (Ensure required columns are present)
+    sample_df['Store_Age'] = 2025 - sample_df['Store_Establishment_Year']
+    sample_df['Is_Tier1'] = sample_df['Store_Location_City_Type'].apply(lambda x: 1 if x == "Tier 1" else 0)
+    sample_df['MRP_Category'] = pd.cut(sample_df['Product_MRP'], bins=[0, 100, 200, 300], labels=['Low', 'Medium', 'High'])
+
+    # Ensure categorical features match model expectations
+    categorical_cols = ['Product_Sugar_Content', 'Product_Type', 'Store_Size', 'Store_Location_City_Type', 'Store_Type']
+    for col in categorical_cols:
+        if col in sample_df.columns:
+            sample_df[col] = sample_df[col].astype('category')
 
-    try:
-        # Get JSON data from the request
-        data = request.get_json()
-        if not data:
-            return jsonify({"error": "No data provided"}), 400
+    # Ensure feature names match the trained model
+    required_features = model_pipeline.feature_names_in_
+    missing_features = set(required_features) - set(sample_df.columns)
 
-        # Extract features from the JSON data
-        input_df = pd.DataFrame([data])
+    if missing_features:
+        print(f"Warning: Missing features detected: {missing_features}")
+        for feature in missing_features:
+            sample_df[feature] = 0  # Assign default values for missing features
 
-        prediction = model_pipeline.predict(input_df)
-        return jsonify({"prediction": prediction.tolist()})
+    # Make a prediction
+    prediction = model_pipeline.predict(sample_df[required_features])[0]
+    return jsonify({'prediction': float(prediction)})
 
-    except Exception as e:
-        return jsonify({"error": f'Error during prediction: {e}'}), 500
+if __name__ == '__main__':
+    app.run(debug=True)
 
-if __name__ == "__main__":  # Correct indentation
-    port = int(os.environ.get("PORT", 5000))
-    app.run(host="0.0.0.0", port=5000, debug=True)