|
|
import base64 |
|
|
from io import BytesIO |
|
|
from typing import Dict, Any |
|
|
|
|
|
import torch |
|
|
from PIL import Image |
|
|
from io import BytesIO |
|
|
from diffusers import StableDiffusionImg2ImgPipeline |
|
|
import requests |
|
|
|
|
|
|
|
|
|
|
|
def decode_base64_image(image_string): |
|
|
base64_image = base64.b64decode(image_string) |
|
|
buffer = BytesIO(base64_image) |
|
|
return Image.open(buffer) |
|
|
|
|
|
|
|
|
class EndpointHandler: |
|
|
def __init__(self, path=""): |
|
|
self.pipe = StableDiffusionImg2ImgPipeline.from_pretrained(path, |
|
|
torch_dtype=torch.float16, revision="fp16") |
|
|
self.pipe = self.pipe.to("cuda") |
|
|
|
|
|
def __call__(self, data: Any) -> Dict[str, str]: |
|
|
""" |
|
|
Return predict value. |
|
|
:param data: A dictionary contains `inputs` and optional `image` field. |
|
|
:return: A dictionary with `image` field contains image in base64. |
|
|
""" |
|
|
prompts = data.pop("inputs", None) |
|
|
url = data.pop("image", None) |
|
|
seed = data.pop("seed", 0) |
|
|
width = data.pop("width", 0) |
|
|
height = data.pop("height", 0) |
|
|
response = requests.get(url) |
|
|
init_image = Image.open(BytesIO(response.content)).convert("RGB") |
|
|
|
|
|
init_image.thumbnail((width, height)) |
|
|
|
|
|
generator = torch.Generator(device="cuda").manual_seed(seed) |
|
|
images = self.pipe(prompts, image=init_image,generator = generator, **data).images |
|
|
img_strs = [] |
|
|
|
|
|
for image in images: |
|
|
buffered = BytesIO() |
|
|
image.save(buffered, format="png") |
|
|
img_str = base64.b64encode(buffered.getvalue()) |
|
|
img_strs.append(img_str) |
|
|
|
|
|
if len(img_strs) > 1 : |
|
|
return {"images": [img_str.decode() for img_str in img_strs] } |
|
|
else: |
|
|
return {"image": img_strs[0].decode() } |
|
|
|
