Datasets:
Tags:
Not-For-All-Audiences
Upload Drive to WebP.ipynb
Browse files- Drive to WebP.ipynb +1 -1
Drive to WebP.ipynb
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"
|
|
|
|
| 1 |
+
{"cells":[{"cell_type":"code","execution_count":null,"metadata":{"colab":{"background_save":true},"id":"6HquiGjaJiYK"},"outputs":[],"source":["# Install required libraries\n","#!pip install pillow google-colab\n","\n","from google.colab import drive\n","from PIL import Image\n","import os\n","import zipfile\n","from pathlib import Path\n","import shutil\n","\n","# Mount Google Drive\n","drive.mount('/content/drive')\n","\n","# Define paths and settings\n","source_dir = '/content/drive/MyDrive/Saved from Chrome/'\n","output_zip = '/content/images_lossless_1024_webp.zip'\n","webp_dir = '/content/temp_lossless_1024_webp'\n","MAX_SIZE = 1024 # Maximum dimension (width or height)\n","\n","# Create temporary directory for resized lossless webp files\n","os.makedirs(webp_dir, exist_ok=True)\n","\n","# Get all image files\n","image_extensions = {'.jpg', '.jpeg', '.png', '.bmp', '.tiff', '.gif', '.webp'}\n","image_files = []\n","\n","print(\"🔍 Scanning for images...\")\n","for file_path in Path(source_dir).rglob('*'):\n"," if file_path.is_file() and file_path.suffix.lower() in image_extensions:\n"," image_files.append(file_path)\n","\n","print(f\"✅ Found {len(image_files)} image files\")\n","\n","# Convert, resize, and save as lossless WebP\n","webp_files = []\n","failed_conversions = []\n","resize_stats = {'resized': 0, 'unchanged': 0}\n","\n","print(\"🔄 Resizing to 1024px max dimension & converting to lossless WebP...\")\n","for i, img_path in enumerate(image_files, 1):\n"," try:\n"," # Open image\n"," with Image.open(img_path) as img:\n"," original_width, original_height = img.size\n","\n"," # Resize if either dimension exceeds MAX_SIZE\n"," if original_width > MAX_SIZE or original_height > MAX_SIZE:\n"," # Calculate scaling factor to fit within MAX_SIZE\n"," scale = min(MAX_SIZE / original_width, MAX_SIZE / original_height)\n"," new_width = int(original_width * scale)\n"," new_height = int(original_height * scale)\n","\n"," # Resize using high-quality LANCZOS resampling\n"," img_resized = img.resize((new_width, new_height), Image.Resampling.LANCZOS)\n"," resize_stats['resized'] += 1\n"," else:\n"," img_resized = img\n"," resize_stats['unchanged'] += 1\n","\n"," # Preserve original mode (including transparency) for lossless conversion\n"," # Save as lossless WebP\n"," webp_path = os.path.join(webp_dir, f\"{i:04d}.webp\")\n"," img_resized.save(webp_path, 'WEBP', lossless=True, method=6)\n"," webp_files.append(webp_path)\n","\n"," if i % 10 == 0 or i == len(image_files):\n"," print(f\" Processed {i}/{len(image_files)} images\")\n","\n"," except Exception as e:\n"," error_msg = f\"Error processing {img_path.name}: {str(e)}\"\n"," print(f\"❌ {error_msg}\")\n"," failed_conversions.append((img_path, str(e)))\n"," continue\n","\n","print(f\"\\n📏 Resize Statistics:\")\n","print(f\" Images resized: {resize_stats['resized']}\")\n","print(f\" Images unchanged: {resize_stats['unchanged']}\")\n","print(f\"✅ Successfully converted {len(webp_files)} images to 1024px lossless WebP\")\n","\n","# Create zip file with maximum compression\n","print(\"📦 Creating ZIP archive...\")\n","with zipfile.ZipFile(output_zip, 'w', zipfile.ZIP_DEFLATED, compresslevel=9) as zipf:\n"," for webp_file in webp_files:\n"," arcname = os.path.basename(webp_file)\n"," zipf.write(webp_file, arcname)\n"," if len(webp_files) <= 20 or webp_files.index(webp_file) % 20 == 0:\n"," print(f\" Added {arcname}\")\n","\n","# Calculate statistics\n","original_size = sum(os.path.getsize(f) for f in image_files)\n","webp_size = sum(os.path.getsize(f) for f in webp_files)\n","zip_size = os.path.getsize(output_zip)\n","\n","size_reduction = (1 - webp_size/original_size) * 100 if original_size > 0 else 0\n","\n","print(f\"\\n📊 Conversion Statistics:\")\n","print(f\" Original images size: {original_size / (1024*1024):.2f} MB\")\n","print(f\" 1024px Lossless WebP: {webp_size / (1024*1024):.2f} MB\")\n","print(f\" Size reduction: {size_reduction:.1f}%\")\n","print(f\" Final ZIP size: {zip_size / (1024*1024):.2f} MB\")\n","\n","# Copy zip back to Google Drive\n","drive_output = '/content/drive/MyDrive/images_1024_lossless_webp.zip'\n","shutil.copy2(output_zip, drive_output)\n","print(f\"💾 ZIP saved to Google Drive: {drive_output}\")\n","\n","# Clean up temporary directory\n","shutil.rmtree(webp_dir)\n","print(\"🧹 Temporary files cleaned up\")\n","\n","# Display sample of files in zip\n","print(\"\\n📋 Sample files in ZIP:\")\n","with zipfile.ZipFile(output_zip, 'r') as zipf:\n"," file_list = zipf.namelist()\n"," for i, filename in enumerate(file_list[:10], 1):\n"," print(f\" {i}. {filename}\")\n"," if len(file_list) > 10:\n"," print(f\" ... and {len(file_list) - 10} more files\")\n","\n","print(f\"\\n🎯 AI-Training Ready (1024px Optimized)!\")\n","print(f\" • ✅ 1024px max resolution - standardized for ML\")\n","print(f\" • ✅ 100% lossless quality - NO compression artifacts\")\n","print(f\" • ✅ Preserves transparency (RGBA support)\")\n","print(f\" • ✅ High-quality LANCZOS resampling\")\n","print(f\" • ✅ Sequential numbering: 0001.webp, 0002.webp, etc.\")\n","print(f\" • 🎯 Perfect for CNNs, GANs, object detection, segmentation\")\n","print(f\" • ⚡ Consistent input size for faster training\")\n","print(f\" • 💾 Significant storage savings vs original high-res images\")\n","\n","if failed_conversions:\n"," print(f\"\\n⚠️ Failed conversions (check originals):\")\n"," for orig_path, error in failed_conversions[:5]:\n"," print(f\" • {orig_path.name}: {error}\")\n"," if len(failed_conversions) > 5:\n"," print(f\" ... and {len(failed_conversions) - 5} more\")\n","\n","print(f\"\\n🔧 Technical Details:\")\n","print(f\" • Max dimension: {MAX_SIZE}px (maintains aspect ratio)\")\n","print(f\" • Resampling: LANCZOS (highest quality)\")\n","print(f\" • Lossless WebP: Perfect fidelity, ~50% smaller than PNG\")"]},{"cell_type":"markdown","metadata":{"id":"lz_4gtiQAdEP"},"source":["# 📸 Google Drive Image Converter to WebP\n","\n","**Convert your Chrome-saved images to compact WebP format with automatic indexing!**\n","\n","This Google Colab notebook transforms all images from your `MyDrive/Saved from Chrome/` folder into optimized WebP files, named sequentially as `1.webp`, `2.webp`, `3.webp`, etc., and packages them into a single ZIP archive.\n","\n","## ✨ Features\n","- **Batch Processing**: Automatically finds and converts JPG, PNG, GIF, BMP, TIFF, and existing WebP files\n","- **Smart Conversion**: Handles transparency by converting RGBA/P modes to RGB with white background\n","- **High-Quality WebP**: Uses 85% quality setting for optimal size/quality balance\n","- **Recursive Search**: Processes images in subfolders too\n","- **Error-Resilient**: Skips corrupt files and continues processing\n","- **Dual Output**: Saves ZIP locally in Colab AND copies to your Google Drive root\n","- **Progress Tracking**: Shows conversion progress and final statistics\n","\n","## 🚀 How It Works\n","1. Mounts your Google Drive\n","2. Scans for image files in \"Saved from Chrome/\" and subdirectories\n","3. Converts each image to WebP with sequential numbering\n","4. Creates a compressed ZIP archive\n","5. Cleans up temporary files and reports file sizes\n","\n","## 💾 Output\n","- `images_webp.zip` containing `1.webp`, `2.webp`, `3.webp`, etc.\n","- Significantly smaller file sizes (WebP compression typically 25-35% better than JPEG)\n","- Ready for web use, storage, or sharing\n","\n","## 🎯 Perfect For\n","- Organizing Chrome download clutter\n","- Web developers needing optimized images\n","- Anyone wanting to save Google Drive storage space\n","- Creating indexed image collections for ML datasets or galleries\n","\n","**Run this notebook to instantly compress and organize your Chrome-saved images into a clean, numbered WebP collection!**\n","\n","*Note: Requires Google Drive authentication when first run.*"]}],"metadata":{"colab":{"provenance":[{"file_id":"https://huggingface.co/datasets/codeShare/lora-training-data/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1760450712160},{"file_id":"https://huggingface.co/datasets/codeShare/lora-training-data/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1756712618300},{"file_id":"https://huggingface.co/codeShare/JupyterNotebooks/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1747490904984},{"file_id":"https://huggingface.co/codeShare/JupyterNotebooks/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1740037333374},{"file_id":"https://huggingface.co/codeShare/JupyterNotebooks/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1736477078136},{"file_id":"https://huggingface.co/codeShare/JupyterNotebooks/blob/main/YT-playlist-to-mp3.ipynb","timestamp":1725365086834}]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"}},"nbformat":4,"nbformat_minor":0}
|