Update app.py

app.py

@@ -2,13 +2,15 @@ from flask import Flask, request, jsonify, send_from_directory
 from flask_cors import CORS
 import joblib
 import numpy as np
-import xgboost as xgb
+import json
 import math
+import xgboost as xgb
+import re
 
 app = Flask(__name__, static_folder='.', static_url_path='/')
 CORS(app)
 
-# Load models
+# Load ML models
 try:
     rf = joblib.load("rf_model.pkl")
     xgb_model = xgb.Booster()
@@ -18,7 +20,7 @@ except Exception as e:
     print(f"❌ Error loading models: {e}")
     raise e
 
-# Hardcoded product catalog
+# Product catalog
 tile_catalog = [
     {"name": "Travertino Light Beige", "type": "Floor", "price": 780, "coverage": 15.5, "size": "1200x1200 MM", "url": "https://arqonz.ae/products/6775"},
     {"name": "Marquina Glossy Black", "type": "Wall", "price": 620, "coverage": 12.0, "size": "600x600 MM", "url": "https://arqonz.ae/products/6776"},
@@ -47,26 +49,25 @@ def recommend():
         tile_type = data.get("tile_type", "").strip().lower()
         coverage = float(data.get("coverage", 1))
         area = float(data.get("area", 1))
-        price_range = data.get("price_range", [3, 10000])
+        price_range = data.get("price_range", [1, 10000])
         preferred_sizes = data.get("preferred_sizes", [])
 
         if area <= 0 or coverage <= 0:
-            return jsonify({"error": "Invalid area or coverage"}), 400
+            return jsonify({"error": "Please enter valid area and coverage values."}), 400
 
         features = prepare_features(tile_type, coverage, area, price_range)
         xgb_pred = xgb_model.predict(xgb.DMatrix(features))[0]
         rf_pred = rf.predict_proba(features)[0][1]
         score = (xgb_pred + rf_pred) / 2
 
-        matches = filter_products(tile_type, price_range, preferred_sizes)
-
+        products = filter_products(tile_type, price_range, preferred_sizes)
         return jsonify({
             "recommendation_score": round(float(score), 3),
-            "recommended_products": matches[:4],
-            "total_matches": len(matches)
+            "recommended_products": products[:4],
+            "total_matches": len(products),
         })
     except Exception as e:
-        print("❌ /recommend error:", e)
+        print("❌ Error in /recommend:", str(e))
         return jsonify({"error": "Server error"}), 500
 
 def prepare_features(tile_type, coverage, area, price_range):
@@ -85,22 +86,23 @@ def filter_products(tile_type, price_range, preferred_sizes):
         if not (min_price <= p["price"] <= max_price):
             continue
 
-        size_match = False
-        for size in preferred_sizes:
-            if similar_size(p["size"], size):
-                size_match = True
+        match_found = False
+        for pref in preferred_sizes:
+            if similar_size_mm(p["size"], pref):
+                match_found = True
                 break
-        if not size_match:
+
+        if not match_found:
             continue
 
         price_score = 1 - (p["price"] - min_price) / (max_price - min_price + 1e-6)
         filtered.append({**p, "recommendation_score": round(price_score, 2)})
     return sorted(filtered, key=lambda x: x["recommendation_score"], reverse=True)
 
-def similar_size(size_a, size_b, tolerance=10):
+def similar_size_mm(a, b, tolerance=10):
     try:
-        w1, h1 = map(int, size_a.lower().replace("mm", "").split("x"))
-        w2, h2 = map(int, size_b.lower().replace("mm", "").split("x"))
+        w1, h1 = map(int, a.lower().replace("mm", "").split("x"))
+        w2, h2 = map(int, b.lower().replace("mm", "").split("x"))
         return abs(w1 - w2) <= tolerance and abs(h1 - h2) <= tolerance
     except:
         return False