File size: 1,257 Bytes
f076f26 bfbe48c f076f26 |
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 |
from typing import Dict, List, Any
from PIL import Image
from io import BytesIO
import torch
import base64
from diffusers import StableDiffusionInstructPix2PixPipeline, EulerAncestralDiscreteScheduler
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
class EndpointHandler():
def __init__(self, path=""):
model_id = "timbrooks/instruct-pix2pix"
self.pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, safety_checker=None)
self.pipe.to(device)
self.pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(self.pipe.scheduler.config)
def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
"""
data args:
inputs (:obj:`string`)
parameters (:obj:)
Return:
A :obj:`string`:. Base64 encoded image string
"""
inputs = data.pop("inputs", data)
# decode base64 image to PIL
image = Image.open(BytesIO(base64.b64decode(inputs['image'])))
prompt = inputs['prompt']
images = self.pipe(prompt, image=image, num_inference_steps=10, image_guidance_scale=1).images
return images[0] |