-
Thanks for your contribution, nice work.
{'index': 22147, 'question': 'Please write a poem based on the image.', 'answer': 'J', 'A': "Promise the joy that I've longed to hear", 'B': 'The essence of…
-
### COCO is a large-scale object detection, segmentation, and captioning dataset. COCO has several features: Object segmentation, Recognition in context, Superpixel stuff segmentation, 330K images (>…
-
Thanks for this great work. It would be nice to try it.
Could you please provide CoCa pre-trained weights on LAION-2B and CC3M? Many thanks.
-
Hi @ruotianluo
Thank you for your fantastic work. I followed your implementation to train the captioning model (TopDown) with the top 36 features I downloaded from the author. And below is the res…
-
Hello author, this is a great job! Thank you for your contribution!
May I ask if this is a fine-tuning guide for the BLIP2 model? What is the main aspect of fine-tuning? Is it an image capture task o…
-
Running Command: bash process-youtube-video.sh
raise ValueError(str(e))
ValueError: Dimension 0 in both shapes must be equal, but are 2617 and 2718. Shapes are [2617,256] and [2718,256]. for 'Ass…
-
Issue created by marthinwurer#1160.
-
Hi,
I am a user of the Lavis library. Now I am doing some research for the hardware requirements when using BLIP2 model. Do you know the hardware requirements for the BLIP2 model?
I am looking for…
-
How do you add metadata to ImageClassification tasks as shown in the Demo ImageClassification task below? It seems you can only upload images.
 with blip2 for images captioning.
Sometimes, the generated text includes irrelevant or unwarranted intellectual property, such as '…