File size: 1,590 Bytes
427113a ba84673 b2d86ed 427113a 1347a75 ba84673 61b80cc ba84673 61b80cc ba84673 61b80cc ba84673 61b80cc 427113a 61b80cc 427113a 61b80cc | 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 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 | from transformers import Blip2Processor, Blip2ForConditionalGeneration
from typing import Dict, List, Any
from PIL import Image
from transformers import pipeline
import requests
import torch
class EndpointHandler():
def __init__(self, path=""):
"""
path:
"""
self.device = "cuda" if torch.cuda.is_available() else "cpu"
self.processor = Blip2Processor.from_pretrained(path)
self.model = Blip2ForConditionalGeneration.from_pretrained(path, torch_dtype=torch.float16).to(self.device)
def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
"""
data args:
inputs (:obj: `str` | `PIL.Image` | `np.array`)
kwargs
Return:
A :obj:`list` | `dict`: will be serialized and returned
"""
result = {}
inputs = data.pop("inputs", data)
image_url = inputs['image_url']
if "prompt" in inputs:
prompt = inputs["prompt"]
else:
prompt = None
image = Image.open(requests.get(image_url, stream=True).raw).convert('RGB')
if prompt:
processed_image = self.processor(images=image, text=prompt, return_tensors="pt").to(self.device, torch.float16)
else:
processed_image = self.processor(images=image, return_tensors="pt").to(self.device, torch.float16)
output = self.model.generate(**processed_image)
text_output = self.processor.decode(output[0], skip_special_tokens=True)
result["text_output"] = text_output
return result
|