How do i generate a caption
So far, I have the CLI running but it only lets me chat with the model and does not generate the meme
It is so far only describing what the llava sees in the image
Hi, the model should generate a caption for the image. Can you please provide your sample image, the model's output, and your expected output?
Hi,
This is the code I am running on an a100 :
CUDA_VISIBLE_DEVICES=0 python -m llava.serve.cli --model-path AmirHossein1378/LLaVA-1.5-7b-meme-captioner --image-file pic.jpeg
It shows a user prompt :
Human: "generate a funny caption "
ouput:
Assistant: A woman is wearing rainbow sunglasses. Meme poster is trying to convey that the woman is gay.
Here a link to the image : https://www.instagram.com/unitedcoffeehouserewind/p/C4Pz1HNxLBI/?img_index=1
I was expecting a meme like caption .
The output of the model is as expected. Please note that here we're referring to the model creating a concise description of the meme, including its purpose and target audience, rather than generating the text that should appear within the meme itself.
Got it! Thanks a lot