mehdidc/DALLE_clip_score

Simple script to compute CLIP-based scores given a DALL-e trained model.

21
/ 100
Experimental

This tool helps researchers and developers evaluate the quality of images generated by a DALL-E model. You provide your trained DALL-E model and a dataset of real images with their corresponding text captions. The tool then calculates various "CLIP scores" that indicate how well the generated images match their intended captions, and also how they compare to real images. This is useful for anyone working on text-to-image generation models to understand and improve their performance.

No commits in the last 6 months.

Use this if you are developing or fine-tuning a text-to-image generation model, specifically DALL-E, and need quantitative metrics to assess the image-text alignment of your model's outputs.

Not ideal if you are looking for a tool to generate images directly or to evaluate image quality based on human perception rather than a model-based metric.

AI-model-evaluation generative-AI text-to-image deep-learning-research computer-vision
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 6 / 25

How are scores calculated?

Stars

29

Forks

2

Language

Python

License

Last pushed

Jun 13, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/mehdidc/DALLE_clip_score"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.