clip-gpt-captioning
clip-gpt-captioning copied to clipboard
CLIPxGPT Captioner is Image Captioning Model based on OpenAI's CLIP and GPT-2.
![image](https://github.com/jmisilo/clip-gpt-captioning/assets/141383792/e22a83b0-0cf2-4370-9b3e-b595a1805a1d) At this point, it stops moving. What is the reason for this?
Can this project batch process pedestrian images? (i.e. generating corresponding text descriptions for pedestrian images) Can good captions be achieved for low resolution images? Looking forward to your reply! Thank...
how do I put this model in kohya_ss? I insert the link, but it doesn't work, I insert the name and it doesn't work either. ![Screenshot_1](https://github.com/jmisilo/clip-gpt-captioning/assets/121365756/06ee0724-5b7f-47f3-8767-2d72962d264d) ![Screenshot_2](https://github.com/jmisilo/clip-gpt-captioning/assets/121365756/cce0a6e2-056e-4d6d-8f42-d612d5126763)
Hi, Thanks for your nice work. I want to ask what are the training and validation losses for flickr 30k dataset in your experiments. I experimented with your code on...