File size: 2,854 Bytes
ed105db 589bc03 ed105db b29140f ed105db cec4136 ed105db cc479e4 c84ac25 ed105db c84ac25 ed105db c84ac25 ed105db cc479e4 ed105db 043d595 ed105db cc479e4 043d595 ed105db 589bc03 ed105db |
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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
import requests
from typing import Dict, Any
from PIL import Image
import torch
import base64
import io
from transformers import BlipForConditionalGeneration, BlipProcessor
import logging
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# Configure logging
logging.basicConfig(level=logging.DEBUG)
# Configure logging
logging.basicConfig(level=logging.ERROR)
# Configure logging
logging.basicConfig(level=logging.WARNING)
class EndpointHandler():
def __init__(self, path=""):
self.processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
self.model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large").to(device)
self.model.eval()
def __call__(self, data: Dict[str, Any]) -> Dict[str, Any]:
logging.error(f"----------This is an error message {str(data)}")
input_data = data.get("inputs", {})
logging.warning(f"------input_data-- {str(input_data)}")
# encoded_images = input_data.get("images")
#logging.warning(f"---encoded_images----- {str(encoded_images)}")
if not input_data:
#logging.warning(f"---encoded_images--not provided in if block--- {str(encoded_images)}")
return {"captions": [], "error": "No images provided"}
try:
logging.warning(f"---encoded_images-- provided in try block--- {str(input_data)}")
byteImgIO = io.BytesIO()
byteImg = Image.open(input_data)
byteImg.save(byteImgIO, "PNG")
byteImgIO.seek(0)
byteImg = byteImgIO.read()
# Non test code
dataBytesIO = io.BytesIO(byteImg)
raw_images =[Image.open(dataBytesIO)]
logging.warning(f"----raw_images----0--- {str(raw_images)}")
# Check if any images were successfully decoded
if not raw_images:
print("No valid images found.")
processed_inputs = [
self.processor(image, return_tensors="pt") for image in zip(raw_images)
]
processed_inputs = {
"pixel_values": torch.cat([inp["pixel_values"] for inp in processed_inputs], dim=0).to(device),
"max_new_tokens":40
}
with torch.no_grad():
out = self.model.generate(**processed_inputs)
captions = self.processor.batch_decode(out, skip_special_tokens=True)
logging.warning(f"----captions---- {str(captions)}")
print("caption is here-------",captions)
return {"captions": captions}
except Exception as e:
print(f"Error during processing: {str(e)}")
logging.error(f"Error during processing: ----------------{str(e)}")
return {"captions": [], "error": str(e)}
|