File size: 4,348 Bytes
fca4fc0 |
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 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 |
import json
import os
import shutil
import random
from pathlib import Path
# Load the datasets
train = json.load(open('/home/jisoo/demo/VidChain/VTimeLLM/data/activitynet/mdpo-train.json'))
val = json.load(open('/home/jisoo/demo/VidChain/VTimeLLM/data/activitynet/val_2.json'))
# Source directory where videos are already stored
source_dir = Path('/hub_data2/jisoo/data/ActivityNet/videos')
# Create directories if they don't exist
train_dir = Path('/home/jisoo/demo/VidChain/VTimeLLM/data/activitynet/videos/train')
val_dir = Path('/home/jisoo/demo/VidChain/VTimeLLM/data/activitynet/videos/val')
train_dir.mkdir(exist_ok=True)
val_dir.mkdir(exist_ok=True)
def copy_video(video_id, output_dir):
"""Copy a video from the source directory to the target directory"""
# Look for mp4 files first, then other extensions
possible_extensions = ['.mp4', '.mkv', '.avi', '.mov']
for ext in possible_extensions:
source_path = source_dir / f"{video_id}{ext}"
if source_path.exists():
output_path = output_dir / f"{video_id}.mp4"
# Skip if file already exists
if output_path.exists():
print(f"Video {video_id} already exists in {output_dir}, skipping...")
return True
try:
print(f"Copying {video_id} to {output_dir}...")
shutil.copy2(source_path, output_path)
print(f"Successfully copied {video_id}")
return True
except Exception as e:
print(f"Error copying {video_id}: {e}")
return False
print(f"Video {video_id} not found in source directory")
return False
def copy_sample_videos_from_list(video_ids, output_dir, num_samples=25):
"""Copy a sample of videos from a list of video IDs"""
# Randomly sample videos
if len(video_ids) > num_samples:
sampled_ids = random.sample(video_ids, num_samples)
else:
sampled_ids = video_ids
print(f"Copying {len(sampled_ids)} videos to {output_dir}...")
successful_copies = 0
for video_id in sampled_ids:
if copy_video(video_id, output_dir):
successful_copies += 1
print(f"Successfully copied {successful_copies}/{len(sampled_ids)} videos")
return successful_copies
# Set random seed for reproducibility
random.seed(42)
# Check if source directory exists
if not source_dir.exists():
print(f"Error: Source directory {source_dir} does not exist!")
exit(1)
print(f"Source directory: {source_dir}")
print(f"Train directory: {train_dir}")
print(f"Validation directory: {val_dir}")
# Get video IDs from train dataset (it has a 'vid' field with a list of video IDs)
train_video_ids = train['vid'] if 'vid' in train else []
print(f"Found {len(train_video_ids)} video IDs in train dataset")
# Get video IDs from validation dataset (it has video IDs as keys)
val_video_ids = list(val.keys())
print(f"Found {len(val_video_ids)} video IDs in validation dataset")
# Copy sample videos
# print("\n=== Copying Train Videos ===")
# train_success = copy_sample_videos_from_list(train_video_ids, train_dir, num_samples=100)
print("\n=== Copying Validation Videos ===")
val_success = copy_sample_videos_from_list(val_video_ids, val_dir, num_samples=50)
print(f"\n=== Summary ===")
# print(f"Train videos copied: {train_success}/100")
print(f"Validation videos copied: {val_success}/25")
# print(f"Total videos copied: {train_success + val_success}/125")
# List the copied files
# print(f"\n=== Copied Train Videos ===")
# train_files = list(train_dir.glob("*.mp4"))
# for file in train_files:
# print(f" {file.name}")
print(f"\n=== Copied Validation Videos ===")
val_files = list(val_dir.glob("*.mp4"))
for file in val_files:
print(f" {file.name}")
# Show some statistics
print(f"\n=== Statistics ===")
# print(f"Total train files: {len(train_files)}")
print(f"Total validation files: {len(val_files)}")
# # Check file sizes
# if train_files:
# total_train_size = sum(f.stat().st_size for f in train_files)
# print(f"Total train directory size: {total_train_size / (1024*1024):.2f} MB")
if val_files:
total_val_size = sum(f.stat().st_size for f in val_files)
print(f"Total validation directory size: {total_val_size / (1024*1024):.2f} MB") |