File size: 2,026 Bytes
fb5d697 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 | # split clips into train / val / test set
# From the list, split clips into train / val / test directories
import csv
import shutil
import os
# read clip_uid csv file save as list
def read_uid_csv(filename):
with open(filename, 'r') as f:
reader = csv.reader(f)
uid_list = list(reader)
return uid_list
def split_clips(clips_path, clips_split_path):
subset = ['train', 'val', 'test']
for split in subset:
if split not in clip_csv_dict.keys():
print("{} split is not in clip_csv_dict".format(split))
continue
clip_uids_list = read_uid_csv(clip_csv_dict[split])[0]
os.makedirs(os.path.join(clips_split_path, split),exist_ok=True)
# copy videos from clips_path into clips_split_path
for clip_uid in clip_uids_list:
clip_filename = clip_uid + '.mp4'
src = os.path.join(clips_path, clip_filename)
dst = os.path.join(clips_split_path, split, clip_filename)
shutil.copy(src, dst)
print('copied {} to {}'.format(src, dst))
clip_csv_dict = {
'train': 'train_clip_uids.csv',
'val': 'val_clip_uids.csv',
'test': 'test_clip_uids.csv'
}
ego4d_path = '/home/junhyeok/datasets/ego4d_dataset/v2/'
clips_path = os.path.join(ego4d_path, 'clips')
clips_split_path = os.path.join(ego4d_path, 'splits')
os.makedirs(clips_split_path, exist_ok=True)
def convert_csv_to_lines_and_save(csv_file_path, output_file_path):
with open(csv_file_path, 'r') as file:
csv_content = file.read().strip()
lines = csv_content.split(',')
# Join the lines with newline character '\n' to create the desired format
converted_text = '\n'.join(lines)
with open(output_file_path, 'w') as file:
file.write(converted_text)
print("Conversion completed. The result is saved to '{}'.".format(output_file_path))
convert_csv_to_lines_and_save('train_clip_uids.csv', 'train_clip_uids.txt')
convert_csv_to_lines_and_save('val_clip_uids.csv', 'val_clip_uids.txt')
convert_csv_to_lines_and_save('test_clip_uids.csv', 'test_clip_uids.txt')
|