File size: 2,612 Bytes
2bed270
 
 
 
 
 
acc3215
 
 
2bed270
 
 
acc3215
 
b54826a
acc3215
 
2bed270
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
acc3215
 
 
 
 
 
 
2bed270
 
acc3215
 
2bed270
 
acc3215
2bed270
 
 
 
a4efcf4
2bed270
6e89f05
 
acc3215
 
3280a05
4f693a4
2bed270
 
 
 
 
 
 
 
 
 
acc3215
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
67
68
69
70
71
72
73
from typing import Dict, List, Any
from transformers import pipeline
import torch
import base64
from io import BytesIO
from PIL import Image
from diffusers import StableDiffusionXLImg2ImgPipeline
from diffusers.utils import load_image



class EndpointHandler():
    def __init__(self, path=""):                
        self.pipe = StableDiffusionXLImg2ImgPipeline.from_pretrained("stabilityai/sdxl-turbo", torch_dtype=torch.float16, variant="fp16")
        self.pipe.to("cuda")        
        self.pipe.upcast_vae()


    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
        """
       data args:
            inputs (:obj: `str`)
            date (:obj: `str`)
       Return:
            A :obj:`list` | `dict`: will be serialized and returned
        """
        # get inputs
        inputs = data.pop("inputs", data)
        encoded_image = data.pop("image", None)
        
        # hyperparamters
        num_inference_steps = data.pop("num_inference_steps", 25)
        guidance_scale = data.pop("guidance_scale", 7.5)
        negative_prompt = data.pop("negative_prompt", None)
        height = data.pop("height", None)
        width = data.pop("width", None)

        strength = data.pop("strength", 0.7)
        denoising_start = data.pop("denoising_start_step", 0)
        denoising_end = data.pop("denoising_start_step", 0)
        num_images_per_prompt = data.pop("num_images_per_prompt", 1)
        aesthetic_score = data.pop("aesthetic_score", 0.6)

        
        # process image
        if encoded_image is not None:
            image = self.decode_base64_image(encoded_image)            
        else:
            image = None
            
        
        # run inference pipeline
        out = self.pipe(inputs, 
                        image=image,             
                        strength=strength,
                        num_inference_steps=num_inference_steps,
                        denoising_start=denoising_start,
                        denoising_end=denoising_end,
                        num_images_per_prompt=num_images_per_prompt,
                        aesthetic_score=aesthetic_score,
                        guidance_scale=guidance_scale,                        
                        negative_prompt=negative_prompt
        )
            
        # return first generate PIL image
        return out.images[0]

    # helper to decode input image
    def decode_base64_image(self, image_string):
        base64_image = base64.b64decode(image_string)
        buffer = BytesIO(base64_image)
        image = Image.open(buffer)
        return image