Sphere360 / toolset /crawl /channel /get_test_list.py
omniaudio's picture
Upload folder using huggingface_hub
77dbe7c verified
import os
import pandas as pd
import random
import argparse
def process_csv(input_folder, output_folder, sample_size=10):
"""
Process CSV files in the input folder and randomly select a specified number of video_ids.
Args:
input_folder: Path to folder containing input CSV files
output_folder: Path to folder where processed files will be saved
sample_size: Number of video_ids to randomly select (default: 10)
"""
# Get all CSV files in input folder
csv_files = [f for f in os.listdir(input_folder) if f.endswith('.csv')]
# Process each CSV file
for csv_file in csv_files:
# Build input and output file paths
input_file_path = os.path.join(input_folder, csv_file)
output_file_path = os.path.join(output_folder, csv_file)
# Read CSV file
df = pd.read_csv(input_file_path)
# Check if 'video_id' column exists
if 'video_id' not in df.columns:
print(f"Warning: 'video_id' column not found in {csv_file}. Skipping...")
continue
# Randomly select specified number of video_ids
all_video_ids = df['video_id'].tolist()
if len(all_video_ids) > sample_size:
selected_video_ids = random.sample(all_video_ids, sample_size)
else:
selected_video_ids = all_video_ids
selected_file_ids = [video_id + '_5' for video_id in selected_video_ids]
# Create new DataFrame with selected file_ids
selected_df = pd.DataFrame({'file_id': selected_file_ids})
# Save results to output folder
selected_df.to_csv(output_file_path, index=False)
print(f"Processed {csv_file}, selected {len(selected_video_ids)} video_ids, saved to {output_file_path}")
def main():
# Create argument parser
parser = argparse.ArgumentParser(
description="Process CSV files and randomly select video_ids."
)
# Add command line arguments
parser.add_argument('-i', '--input-folder',
type=str,
help="Input folder containing CSV files",
required=True)
parser.add_argument('-o', '--output-folder',
type=str,
help="Output folder to save the result CSV files",
required=True)
parser.add_argument('-n', '--sample-size',
type=int,
default=10,
help="Number of video_ids to randomly select (default: 10)")
# Parse arguments
args = parser.parse_args()
# Create output folder if it doesn't exist
os.makedirs(args.output_folder, exist_ok=True)
# Process CSV files
process_csv(args.input_folder, args.output_folder, args.sample_size)
if __name__ == '__main__':
main()