load_data.py

# import torch
# import pandas as pd
# from preprocess import get_files, get_dirs
# import numpy as np


# # def load(img_path, label_path):
# #     label = pd.read_csv(label_path)
# #     types = ['Freeform', 'Northwind']
# #     paths, labels = [], []
# #     for t in types:
# #         dirs = get_dirs(img_path + t)
# #         for d in dirs:
# #             no = d[-5:]
# #             l = label[label['file'] == no]['label'].to_numpy()[0]
# #             files = get_files(d)
# #             for file in files:
# #                 paths.append(d + '/' + file)
# #                 labels.append(l)
# #     return paths, torch.from_numpy(np.array(labels)).view((len(labels), 1))


# def load(img_path, label_path):
#    label = pd.read_csv(label_path)
#    types = ['Freeform', 'Northwind']
#    paths, labels = [], []
#    for t in types:
#       dirs = get_dirs(img_path + t)
#       for d in dirs:
#          files = get_files(d)
#          for file in files:
#             no = file
#             l = label[label['file'] == no]['label'].to_numpy()[0]
#             paths.append(d + '/' + file)
#             labels.append(l)
#    return paths, torch.from_numpy(np.array(labels)).view((len(labels), 1))



# if __name__ == "__main__":
#    img_path = "your_img_path"
#    label_path = "G:\keep\label.csv"
#    paths, labels = load(img_path, label_path)
#    print(paths)
#    print(labels)



import torch
import pandas as pd
from preprocess import get_files, get_dirs
import numpy as np

def load(img_path, label_path):
    label = pd.read_csv(label_path)
    types = ['Freeform', 'Northwind']
    paths, labels = [], []
    for t in types:
        print(f"Processing type: {t}")
        dirs = get_dirs(img_path + '/' + t)
        for d in dirs:
            print(f"Processing directory: {d}")
            files = get_files(d)
            for file in files:
                print(f"Processing file: {file}")
                no = file
                try:
                    l = label[label['file'] == no]['label'].to_numpy()[0]
                    print(f"Found label: {l} for file {file}")
                    paths.append(d + '/' + file)
                    labels.append(l)
                except IndexError as e:
                    print(f"Label for file {file} not found.")
    return paths, torch.from_numpy(np.array(labels)).view((len(labels), 1))

if __name__ == "__main__":
    img_path = "G:\\keep\\test"  # Update this path with the correct one.
    label_path = "G:\\keep\\label.csv"  # Ensure this path is correct and accessible.
    paths, labels = load(img_path, label_path)
    print("Final paths:")
    print(paths)
    print("Final labels:")
    print(labels)