app.py

import gradio as gr
import requests
import time
import random
import json
import base64
import os
from transformers import pipeline, set_seed
from io import BytesIO
import io
import html
import PIL
from PIL import Image
import re
import cv2
import numpy as np

def upscale_image(input_image, radio_input):
    upscale_factor = radio_input
    output_image = cv2.resize(input_image, None, fx = upscale_factor, fy = upscale_factor, interpolation = cv2.INTER_CUBIC)
    return [output_image], output_image

def send_prompt(prompt):
    return prompt

def query(payload, model):
    HF_TOKEN = os.getenv("HF_TOKEN")
    headers = {"Authorization": f"Bearer {HF_TOKEN}"}
    url = "https://api-inference.huggingface.co/models/"
    API_URL = f"{url}{model}"
    response = requests.post(API_URL, headers=headers, json=payload)
    return response.content

def hf_inference(prompt, negative, model, steps, sampler, guidance, width, height, seed, progress=gr.Progress()):
    try:
        progress(0, desc="Starting")
        images=[]
        time.sleep(2.5)
        progress(0.05)
        progress(0.25, desc="Generating")
        image_bytes = query(payload={
            "inputs": f"{prompt}",
            "parameters": {
                "negative_prompt": f"{negative}",
                "num_inference_steps": steps,
                "guidance_scale": guidance,
                "width": width, "height": height,
                "seed": seed,
            },
        }, model=model)
        progress(0.75, desc="Opening image")
        time.sleep(1)
        image = Image.open(io.BytesIO(image_bytes))
        images.append(image)
        progress(0.99, desc="Sending image")
        time.sleep(0.5)
        return images
    except PIL.UnidentifiedImageError:
        gr.Warning("This model is not loaded now. Try others models.")






class Prodia:
    def __init__(self, api_key, base=None):
        self.base = base or "https://api.prodia.com/v1"
        self.headers = {
            "X-Prodia-Key": api_key
        }

    def generate(self, params):
        response = self._post(f"{self.base}/sd/generate", params)
        return response.json()

    def transform(self, params):
        response = self._post(f"{self.base}/sd/transform", params)
        return response.json()

    def controlnet(self, params):
        response = self._post(f"{self.base}/sd/controlnet", params)
        return response.json()

    def get_job(self, job_id):
        response = self._get(f"{self.base}/job/{job_id}")
        return response.json()

    def wait(self, job):
        job_result = job

        while job_result['status'] not in ['succeeded', 'failed']:
            time.sleep(0.25)
            job_result = self.get_job(job['job'])

        return job_result

    def list_models(self):
        response = self._get(f"{self.base}/sd/models")
        return response.json()

    def list_samplers(self):
        response = self._get(f"{self.base}/sd/samplers")
        return response.json()

    def _post(self, url, params):
        headers = {
            **self.headers,
            "Content-Type": "application/json"
        }
        response = requests.post(url, headers=headers, data=json.dumps(params))

        if response.status_code != 200:
            raise Exception(f"Bad Prodia Response: {response.status_code}")

        return response

    def _get(self, url):
        response = requests.get(url, headers=self.headers)

        if response.status_code != 200:
            raise Exception(f"Bad Prodia Response: {response.status_code}")

        return response


def image_to_base64(image):
    # Convert the image to bytes
    buffered = BytesIO()
    image.save(buffered, format="PNG")  # You can change format to PNG if needed

    # Encode the bytes to base64
    img_str = base64.b64encode(buffered.getvalue())

    return img_str.decode('utf-8')  # Convert bytes to string


def remove_id_and_ext(text):
    text = re.sub(r'\[.*\]$', '', text)
    extension = text[-12:].strip()
    if extension == "safetensors":
        text = text[:-13]
    elif extension == "ckpt":
        text = text[:-4]
    return text


def get_data(text):
    results = {}
    patterns = {
        'prompt': r'(.*)',
        'negative_prompt': r'Negative prompt: (.*)',
        'steps': r'Steps: (\d+),',
        'seed': r'Seed: (\d+),',
        'sampler': r'Sampler:\s*([^\s,]+(?:\s+[^\s,]+)*)',
        'model': r'Model:\s*([^\s,]+)',
        'cfg_scale': r'CFG scale:\s*([\d\.]+)',
        'size': r'Size:\s*([0-9]+x[0-9]+)'
    }
    for key in ['prompt', 'negative_prompt', 'steps', 'seed', 'sampler', 'model', 'cfg_scale', 'size']:
        match = re.search(patterns[key], text)
        if match:
            results[key] = match.group(1)
        else:
            results[key] = None
    if results['size'] is not None:
        w, h = results['size'].split("x")
        results['w'] = w
        results['h'] = h
    else:
        results['w'] = None
        results['h'] = None
    return results


def send_to_img2img_def(images):
    return images


def send_to_txt2img(image):
    result = {tabs: gr.update(selected="t2i")}

    try:
        text = image.info['parameters']
        data = get_data(text)
        result[prompt] = gr.update(value=data['prompt'])
        result[negative_prompt] = gr.update(value=data['negative_prompt']) if data[
                                                                                  'negative_prompt'] is not None else gr.update()
        result[steps] = gr.update(value=int(data['steps'])) if data['steps'] is not None else gr.update()
        result[seed] = gr.update(value=int(data['seed'])) if data['seed'] is not None else gr.update()
        result[cfg_scale] = gr.update(value=float(data['cfg_scale'])) if data['cfg_scale'] is not None else gr.update()
        result[width] = gr.update(value=int(data['w'])) if data['w'] is not None else gr.update()
        result[height] = gr.update(value=int(data['h'])) if data['h'] is not None else gr.update()
        result[sampler] = gr.update(value=data['sampler']) if data['sampler'] is not None else gr.update()
        if model in model_names:
            result[model] = gr.update(value=model_names[model])
        else:
            result[model] = gr.update()
        return result

    except Exception as e:
        print(e)

        return result


prodia_client = Prodia(api_key=os.getenv("PRODIA_API_KEY"))
model_list = prodia_client.list_models()
model_names = {}

for model_name in model_list:
    name_without_ext = remove_id_and_ext(model_name)
    model_names[name_without_ext] = model_name


def txt2img(prompt, negative_prompt, model, steps, sampler, cfg_scale, width, height, upscale, seed, progress=gr.Progress()):
    progress(0, desc="Starting")
    time.sleep(2.5)
    progress(0.25, desc="Generating")
    result = prodia_client.generate({
        "prompt": prompt,
        "negative_prompt": negative_prompt,
        "model": model,
        "steps": steps,
        "sampler": sampler,
        "cfg_scale": cfg_scale,
        "width": width,
        "height": height,
        "upscale": upscale,
        "seed": seed
    })
    progress(0.75, desc="Opening image")
    job = prodia_client.wait(result)
    progress(0.99, desc="Sending image")
    return [job["imageUrl"]], job["imageUrl"]


def img2img(input_image, denoising, prompt, negative_prompt, model, steps, sampler, cfg_scale, width, height, upscale, seed, progress=gr.Progress()):
    progress(0, desc="Starting")
    time.sleep(1.5)
    progress(0.10, desc="Uploading input image")
    time.sleep(1.5)
    progress(0.25, desc="Generating")
    result = prodia_client.transform({
        "imageData": image_to_base64(input_image),
        "denoising_strength": denoising,
        "prompt": prompt,
        "negative_prompt": negative_prompt,
        "model": model,
        "steps": steps,
        "sampler": sampler,
        "cfg_scale": cfg_scale,
        "width": width,
        "height": height,
        "upscale": upscale,
        "seed": seed
    })
    progress(0.75, desc="Opening image")

    job = prodia_client.wait(result)
    progress(0.99, desc="Sending image")
    time.sleep(0.5)
    return [job["imageUrl"]], job["imageUrl"]


css = """
#generate {
    height: 100%;
}
"""

def change_checkpoint(name, progress=gr.Progress()):
    progress(0, desc="Starting")
    time.sleep(0.5)
    progress(0.25, desc="Changing")
    time.sleep(1)
    progress(0.95, desc="Loading to GPU")
    time.sleep(0.5)
    progress(0.100, desc="Ready")
    return name
    

with gr.Blocks(css=css) as demo:
    with gr.Row():
        with gr.Column(scale=6):
            model = gr.Dropdown(interactive=True, show_label=True,
                                label="Stable Diffusion Checkpoint", choices=prodia_client.list_models())

    with gr.Tabs() as tabs:
        with gr.Tab("txt2img", id='t2i'):
            with gr.Row():
                with gr.Column(scale=6, min_width=600):
                    prompt = gr.Textbox(placeholder="Prompt", show_label=False, lines=3)
                    negative_prompt = gr.Textbox(placeholder="Negative Prompt", show_label=False, lines=3,
                                                 value="(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation")
                with gr.Column():
                    text_button = gr.Button("Generate", variant='primary', elem_id="generate")

            with gr.Row():
                with gr.Column(scale=3):
                    with gr.Tab("Generation"):
                        with gr.Row():
                            with gr.Column(scale=1):
                                sampler = gr.Dropdown(value="DPM++ 2M Karras", show_label=True, label="Sampling Method",
                                                      choices=prodia_client.list_samplers())
    
                            with gr.Column(scale=1):
                                steps = gr.Slider(label="Sampling Steps", minimum=1, maximum=25, value=20, step=1)
                            with gr.Column(scale=1):
                                upscale = gr.Checkbox(label="Upscale", value=False, interactive=True)
                        with gr.Row():
                            with gr.Column(scale=1):
                                width = gr.Slider(label="Width", maximum=1024, value=512, step=8)
                                height = gr.Slider(label="Height", maximum=1024, value=512, step=8)

                            with gr.Column(scale=1):
                                batch_size = gr.Slider(label="Batch Size", maximum=1, value=1)
                                batch_count = gr.Slider(label="Batch Count", maximum=1, value=1)

                        cfg_scale = gr.Slider(label="CFG Scale", minimum=1, maximum=20, value=8, step=1)
                        seed = gr.Number(label="Seed", value=-1)

                with gr.Column(scale=2):
                    image_output = gr.Gallery(show_label=False, rows=2, preview=True)
                    send_to_img2img = gr.Button(value="Send OUTPUT IMAGE to img2img")
                    send_to_png = gr.Button(value="Send OUTPUT IMAGE to PNG Info")
                    past_url = gr.Textbox(visible=False, interactive=False)

            text_button.click(txt2img, inputs=[prompt, negative_prompt, model, steps, sampler, cfg_scale, width, height, upscale,
                                               seed], outputs=[image_output, past_url], concurrency_limit=64)

        with gr.Tab("img2img", id='i2i'):
            with gr.Row():
                with gr.Column(scale=6, min_width=600):
                    i2i_prompt = gr.Textbox(placeholder="Prompt", show_label=False, lines=3)
                    i2i_negative_prompt = gr.Textbox(placeholder="Negative Prompt", show_label=False, lines=3,
                                                     value="(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation")
                with gr.Column():
                    i2i_text_button = gr.Button("Generate", variant='primary', elem_id="generate")

            with gr.Row():
                with gr.Column(scale=3):
                    with gr.Tab("Generation"):
                        i2i_image_input = gr.Image(type="pil", interactive=True)

                        with gr.Row():
                            with gr.Column(scale=1):
                                i2i_sampler = gr.Dropdown(value="Euler a", show_label=True, label="Sampling Method",
                                                          choices=prodia_client.list_samplers())

                            with gr.Column(scale=1):
                                i2i_steps = gr.Slider(label="Sampling Steps", minimum=1, maximum=25, value=20, step=1)
                            with gr.Column(scale=1):
                                i2i_upscale = gr.Checkbox(label="Upscale", value=False, interactive=True)
                        with gr.Row():
                            with gr.Column(scale=1):
                                i2i_width = gr.Slider(label="Width", maximum=1024, value=512, step=8)
                                i2i_height = gr.Slider(label="Height", maximum=1024, value=512, step=8)

                            with gr.Column(scale=1):
                                i2i_batch_size = gr.Slider(label="Batch Size", maximum=1, value=1)
                                i2i_batch_count = gr.Slider(label="Batch Count", maximum=1, value=1)

                        i2i_cfg_scale = gr.Slider(label="CFG Scale", minimum=1, maximum=20, value=7, step=1)
                        i2i_denoising = gr.Slider(label="Denoising Strength", minimum=0, maximum=1, value=0.7, step=0.1)
                        i2i_seed = gr.Number(label="Seed", value=-1)
                        

                with gr.Column(scale=2):
                    i2i_image_output = gr.Gallery(show_label=False, rows=2, preview=True)
                    send_to_png_i2i = gr.Button(value="Send INPUT IMAGE to PNG Info")
                    i2i_past_url = gr.Textbox(visible=False, interactive=False)

            i2i_text_button.click(img2img, inputs=[i2i_image_input, i2i_denoising, i2i_prompt, i2i_negative_prompt,
                                                   model, i2i_steps, i2i_sampler, i2i_cfg_scale, i2i_width, i2i_height, i2i_upscale,
                                                   i2i_seed], outputs=[i2i_image_output, i2i_past_url], concurrency_limit=64)
        send_to_img2img.click(send_to_img2img_def, inputs=past_url, outputs=i2i_image_input)

        with gr.Tab("PNG Info"):
            def plaintext_to_html(text, classname=None):
                content = "<br>\n".join(html.escape(x) for x in text.split('\n'))

                return f"<p class='{classname}'>{content}</p>" if classname else f"<p>{content}</p>"


            def get_exif_data(path):
                image = Image.open(path)
                items = image.info

                info = ''
                for key, text in items.items():
                    info += f"""
                    <div>
                    <p><b>{plaintext_to_html(str(key))}</b></p>
                    <p>{plaintext_to_html(str(text))}</p>
                    </div>
                    """.strip() + "\n"

                if len(info) == 0:
                    message = "Nothing found in the image."
                    info = f"<div><p>{message}<p></div>"

                return info


            with gr.Row():
                with gr.Column():
                    image_input = gr.Image(type="filepath", interactive=True)
            png_button = gr.Button("Get Info", variant="primary")
            with gr.Row():
                with gr.Column():
                    exif_output = gr.HTML(label="EXIF Data")
                    send_to_txt2img_btn = gr.Button("Send PARAMETRS to txt2img")
                    send_to_img2img_png = gr.Button("Send IMAGE to img2img")

            image_input.upload(get_exif_data, inputs=[image_input], outputs=exif_output)
            png_button.click(get_exif_data, inputs=[image_input], outputs=exif_output)
            send_to_txt2img_btn.click(send_to_txt2img, inputs=[image_input], outputs=[tabs, prompt, negative_prompt,
                                                                                      steps, seed, model, sampler,
                                                                                      width, height, cfg_scale],
                                      concurrency_limit=64)
        send_to_png.click(send_to_img2img_def, inputs=past_url, outputs=image_input)
        send_to_img2img_png.click(send_to_img2img_def, inputs=past_url, outputs=i2i_image_input)
        send_to_png_i2i.click(send_to_img2img_def, inputs=i2i_past_url, outputs=image_input)
        with gr.Tab("HuggingFace Inference"):
            with gr.Row():
                gr.Markdown("Add your model from HF.co, enter model ID.")
                hf_model = gr.Dropdown(label="HuggingFace checkpoint", choices=["runwayml/stable-diffusion-v1-5", "stabilityai/stable-diffusion-2-1", "dataautogpt3/OpenDalleV1.1", "CompVis/stable-diffusion-v1-4", "playgroundai/playground-v2-1024px-aesthetic", "prompthero/openjourney", "openskyml/dreamdrop-v1", "SG161222/Realistic_Vision_V1.4", "digiplay/AbsoluteReality_v1.8.1", "openskyml/dalle-3-xl", "Lykon/dreamshaper-7", "Pclanglais/Mickey-1928"], value="runwayml/stable-diffusion-v1-5", allow_custom_value=True, interactive=True)
            with gr.Row():
                with gr.Column(scale=6, min_width=600):
                    hf_prompt = gr.Textbox(placeholder="Prompt", show_label=False, lines=3)
                    hf_negative_prompt = gr.Textbox(placeholder="Negative Prompt", show_label=False, lines=3,
                                                 value="(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation")
                with gr.Column():
                    hf_text_button = gr.Button("Generate with HF", variant='primary', elem_id="generate")

            with gr.Row():
                with gr.Column(scale=3):
                    with gr.Tab("Generation"):
                        with gr.Row():

                            with gr.Column(scale=1):
                                hf_steps = gr.Slider(label="Sampling Steps", minimum=1, maximum=25, value=20, step=1)

                        with gr.Row():
                            with gr.Column(scale=1):
                                hf_width = gr.Slider(label="Width", maximum=1024, value=512, step=8)
                                hf_height = gr.Slider(label="Height", maximum=1024, value=512, step=8)

                            with gr.Column(scale=1):
                                hf_batch_size = gr.Slider(label="Batch Size", maximum=1, value=1)
                                hf_batch_count = gr.Slider(label="Batch Count", maximum=1, value=1)

                        hf_cfg_scale = gr.Slider(label="CFG Scale", minimum=1, maximum=20, value=8, step=1)
                        hf_seed = gr.Number(label="Seed", value=-1)

                with gr.Column(scale=2):
                    hf_image_output = gr.Gallery(show_label=False, preview=True, rows=2, allow_preview=True)
                    #hf_send_to_img2img = gr.Button(value="Send to img2img")

            hf_text_button.click(hf_inference, inputs=[hf_prompt, hf_negative_prompt, hf_model, hf_steps, sampler, hf_cfg_scale, hf_width, hf_height,
                                               hf_seed], outputs=hf_image_output, concurrency_limit=64)
        with gr.Tab("Prompt Generator"):
            gpt2_pipe = pipeline('text-generation', model='Gustavosta/MagicPrompt-Stable-Diffusion', tokenizer='gpt2')
            with open("ideas.txt", "r") as f:
                line = f.readlines()
            def prompt_gen(starting_text):
                seed = random.randint(100, 1000000)
                set_seed(seed)
                if starting_text == "":
                    starting_text: str = line[random.randrange(0, len(line))].replace("\n", "").lower().capitalize()
                    starting_text: str = re.sub(r"[,:\-–.!;?_]", '', starting_text)
                response = gpt2_pipe(starting_text, max_length=(len(starting_text) + random.randint(60, 90)), num_return_sequences=1)
                response_list = []
                for x in response:
                    resp = x['generated_text'].strip()
                    if resp != starting_text and len(resp) > (len(starting_text) + 4) and resp.endswith((":", "-", "—")) is False:
                        response_list.append(resp+'\n')
                response_end = "\n".join(response_list)
                response_end = re.sub('[^ ]+\.[^ ]+','', response_end)
                response_end = response_end.replace("<", "").replace(">", "")
                if response_end != "":
                    return response_end
            idea = gr.Textbox(show_label=False, placeholder="Enter your idea", max_lines=3, lines=1, interactive=True)
            prompt_button = gr.Button(value="Generate Prompt", variant="primary")
            prompt_full = gr.Textbox(label="Full Prompt", max_lines=8, lines=3, interactive=False, show_copy_button=True)
            send_to_txt2img_prompt = gr.Button(value="Send FULL PROMPT to txt2img")
            send_to_img2img_prompt = gr.Button(value="Send FULL PROMPT to img2img")

            send_to_txt2img_prompt.click(send_prompt, inputs=prompt_full, outputs=prompt)
            send_to_img2img_prompt.click(send_prompt, inputs=prompt_full, outputs=i2i_prompt)
            idea.submit(prompt_gen, inputs=idea, outputs=prompt_full)
            prompt_button.click(prompt_gen, inputs=idea, outputs=prompt_full)
        with gr.Tab("Upscaler"):
            gr.HTML(value="""
            <h1><center>Upscaler</center></h1>
            """)
            with gr.Row():
                us_input = gr.Image(show_label=False, interactive=True, scale=10)
            us_radio = gr.Radio(label="Upscale Levels", choices=[2, 4, 6, 8, 10], value=2, scale=5)
            us_button = gr.Button(value="Generate with Upscaler", variant="primary", scale=5)
            with gr.Column(scale=1):
                us_output = gr.Gallery(show_label=False, rows=2, preview=True, scale=1)
            send_to_img2img_us = gr.Button(value="Send UPSCALED IMAGE to img2img")
            us_past = gr.Image(interactive=False, visible=False)
        
            us_button.click(upscale_image, inputs=[us_input, us_radio], outputs=[us_output, us_past])
            send_to_img2img_us.click(send_to_img2img_def, inputs=us_past, outputs=i2i_image_input)

        with gr.Tab("BLIP"):
            with gr.Tab("Base"):
                gr.load("models/Salesforce/blip-image-captioning-base", title="BLIP-base")
            with gr.Tab("Large"):
                gr.load("models/Salesforce/blip-image-captioning-large", title="BLIP-large")
        with gr.Tab("Classification"):
            gr.load("models/google/vit-base-patch16-224", title="ViT Classification")
        #with gr.Tab("Segmentation"):
        #    gr.load("models/mattmdjaga/segformer_b2_clothes", title="SegFormer Segmentation")
        with gr.Tab("Visual Question Answering"):
            gr.load("models/dandelin/vilt-b32-finetuned-vqa", title="ViLT VQA")

        
        model.change(change_checkpoint, inputs=model, outputs=model)
        hf_model.change(change_checkpoint, inputs=hf_model, outputs=hf_model)
        
        
demo.queue(max_size=80, api_open=False).launch(max_threads=256, show_api=False)