batman

car_price_guesser/data_analysis_temp.py

@@ -0,0 +1,27 @@
+import pandas as pd
+import numpy as np
+
+# Load data
+file_path = "New_York_cars.csv"
+print(f"Loading {file_path}...")
+df = pd.read_csv(file_path)
+
+print("Columns:", df.columns.tolist())
+print("\nShape:", df.shape)
+
+print("\nSample 'money' values:", df["money"].head(10).tolist())
+print("\nSample 'Mileage' values:", df["Mileage"].head(10).tolist())
+
+# Check info
+print("\nInfo:")
+print(df.info())
+
+# Check missing
+print("\nMissing values:\n", df.isnull().sum())
+
+# Check 'money' column type and issues
+# It seems to be int64 based on previous head, but let's verify if there are non-numeric
+print("\nMoney description:\n", df["money"].describe())
+
+# Check 'Mileage'
+print("\nMileage description:\n", df["Mileage"].describe())