Got_classification_folders2.py

import os
import shutil
import pandas as pd

# Preprocess images
data_dir = 'train_classified'
# Load CSV file with filenames and labels
train_df = pd.read_csv('train_updated.csv')

# Initialize count for skipped images
num_skipped = 0

# Iterate over each row in the CSV file
for index, row in train_df.iterrows():
    # Extract filename and label from the current row
    filename = str(row[1])  # Assuming the file names are in the second column (index 1)
    label = str(row[3])     # Assuming the classifications are in the fourth column (index 3)

    # Create directory if it doesn't exist
    label_dir = os.path.join(data_dir, label)
    if len(label) == 1:
        label = f'0{label}'  # Add leading zero if folder name has only one digit
    label_dir = os.path.join(data_dir, label)
    if not os.path.exists(label_dir):
        os.makedirs(label_dir)

    # Move image to its corresponding directory if the file exists
    source_path = os.path.join(data_dir, filename)
    if os.path.exists(source_path):
        destination_path = os.path.join(label_dir, filename)
        shutil.move(source_path, destination_path)
    else:
        print(f"Warning: File '{filename}' mentioned in the CSV was not found.")
        num_skipped += 1

print("Images have been moved to their respective folders.")
print(f"Skipped {num_skipped} images.")