from transformers import BlipProcessor, BlipForConditionalGeneration
from PIL import Image
import requests


class VisionEngine:

    def __init__(self):
        self.processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
        self.model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")

    def describe(self, image_url):

        raw_image = Image.open(requests.get(image_url, stream=True).raw).convert('RGB')

        inputs = self.processor(raw_image, return_tensors="pt")

        out = self.model.generate(**inputs)

        description = self.processor.decode(out[0], skip_special_tokens=True)

        return description