File size: 795 Bytes
c37097e 5d88e42 c37097e ce487ea f23e620 ce487ea 2477b23 ce487ea c37097e ec4ea82 56aa136 fbfd467 c37097e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
from PIL import Image
import requests
import gradio as gr
from transformers import BlipProcessor, BlipForConditionalGeneration
model_id = "Salesforce/blip-image-captioning-base"
model = BlipForConditionalGeneration.from_pretrained(model_id)
processor = BlipProcessor.from_pretrained(model_id)
def launch(input):
outputs = []
for file in input:
inputs = processor(Image.open(file.name), return_tensors="pt")
out = model.generate(**inputs)
outputs.append(processor.decode(out[0], skip_special_tokens=True))
return outputs
description = "Simple BLIP image captioning app that supports multiple images as input."
iface = gr.Interface(launch, description=description, inputs=gr.inputs.File(file_count="multiple"), outputs="text")
iface.launch() |