dataset_properties_get_row_features_count

import pandas as pd
import sys
import os

def analyze_csv(csv_path):
    # Read CSV file
    try:
        # Skip first row as header
        df = pd.read_csv(csv_path, header=0)
    except Exception as e:
        print(f"Error reading CSV file: {e}")
        sys.exit(1)

    # Get filename without extension
    base_filename = os.path.splitext(os.path.basename(csv_path))[0]
    output_filename = f"{base_filename}_info.txt"

    # Get required information
    total_rows = len(df)
    total_columns = len(df.columns)

    # Last column analysis
    last_column = df.iloc[:, -1]
    unique_classes_last = last_column.nunique()
    class_counts_last = last_column.value_counts().to_dict()

    # Second-to-last column analysis
    second_last_column = df.iloc[:, -2]
    unique_classes_second_last = second_last_column.nunique()
    class_counts_second_last = second_last_column.value_counts().to_dict()

    # Format class counts for last column
    class_count_str_last = "\n".join(f"    {class_name}: {count}" for class_name, count in class_counts_last.items())

    # Format class counts for second-to-last column
    class_count_str_second_last = "\n".join(f"    {class_name}: {count}" for class_name, count in class_counts_second_last.items())

    # Create output content
    output_content = f"""0. Filename - {base_filename}
1. Count the number of rows in the dataset : {total_rows}
2. Count the number of columns in the dataset : {total_columns}
3. In the last column the number of unique classes : {unique_classes_last}
4. In the last number the total count of each classes :
{class_count_str_last}
5. In the second last column the number of unique classes : {unique_classes_second_last}
6. In the second last column the total count of each classes :
{class_count_str_second_last}"""

    # Write to output file
    try:
        with open(output_filename, 'w') as f:
            f.write(output_content)
        print(f"Analysis completed. Results written to {output_filename}")
    except Exception as e:
        print(f"Error writing output file: {e}")
        sys.exit(1)

def main():
    if len(sys.argv) != 2:
        print("Usage: python script.py <path_to_csv_file>")
        sys.exit(1)

    csv_path = sys.argv[1]
    if not os.path.exists(csv_path):
        print(f"Error: File {csv_path} does not exist")
        sys.exit(1)

    analyze_csv(csv_path)

if __name__ == "__main__":
    main()