Skip to content
TheChatArchiveThe Chat Archive
gpt-4-0314· openaiHugging Face Suzhen/CodeChat

write a python script that takes an image and reduces its detentions until it is lower…

  1. User

    write a python script that takes an image and reduces its detentions until it is lower than 100kb

  2. Assistant

    To reduce the dimensions of an image until the file size is lower than 100 KB, you can utilize the PIL (Python Imaging Library) library, which is now known as Pillow. First, you need to install the library by running:

    pip install Pillow
    

    Here's a Python script that takes an image file as input, resizes it, and saves the result as a new image file:

    from PIL import Image
    import os
    
    def resize_image(input_image_path, output_image_path):
        image = Image.open(input_image_path)
    
        image_width, image_height = image.size
        factor = 1
    
        while True:
            factor = factor - 0.05
            new_image = image.resize((int(image_width * factor), int(image_height * factor)), Image.ANTIALIAS)
            
            new_image.save(output_image_path, quality=85, optimize=True)
    
            if os.stat(output_image_path).st_size < 100*1024:
                break
    
        print(f"Resized image saved to {output_image_path}")
    
    if __name__ == '__main__':
        resize_image("input_image.jpg", "output_image.jpg")
    

    Replace "input_image.jpg" with the path of your input image and "output_image.jpg" with the path where you want to save the resized image. The script will load the image, reduce its dimensions by a factor (5% at a time in this case), and save the new image with specified quality (85 in this case) and optimization. If the file size is still larger than 100 KB after resizing, the resizing process repeats until the file size is under the desired threshold.

    You may need to change the quality, optimization, and resizing factor in the script based on your specific use case and the sizes of your input images.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.