app.py

import gradio as gr
from huggingface_hub import InferenceClient
import tempfile
import random
import math
from PIL import Image, ImageOps
import logging

logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)


client = InferenceClient(model="timbrooks/instruct-pix2pix")
# client = InferenceClient()
def model_call(prompt, input_image):
    with tempfile.NamedTemporaryFile(suffix=".jpg", delete=True) as temp_file:
        seed = random.randint(0, 100000)
        text_cfg_scale = round(random.uniform(6.0, 9.0), ndigits=2)
        image_cfg_scale = round(random.uniform(1.2, 1.8), ndigits=2)

        width, height = input_image.size
        factor = 512 / max(width, height)
        factor = math.ceil(min(width, height) * factor / 64) * 64 / min(width, height)
        width = int((width * factor) // 64) * 64
        height = int((height * factor) // 64) * 64
        image = ImageOps.fit(input_image, (width, height), method=Image.Resampling.LANCZOS)

        image.save(temp_file.name)
        print('test log here')
        logger.info('logger test')
        return client.image_to_image(temp_file.name, prompt=prompt)#, guidance_scale =7.5, num_inference_steps=50)

iface = gr.Interface(
    fn=model_call,
     inputs=[gr.inputs.Textbox(label="How to transform this image?"), gr.inputs.Image(type="pil")],
      outputs=gr.outputs.Image(type="pil")
).launch(debug=True)