app.py

import os
import sys

import gradio as gr
import numpy as np
import shutil

import copy
import json
import gc
import random
from PIL import Image

'''
models
images
custom.css
sd_cfg.json
'''

'''
if not os.path.exists("sd-ggml-cpp-dp"):
    os.system("git clone https://huggingface.co/svjack/sd-ggml-cpp-dp")
else:
    shutil.rmtree("sd-ggml-cpp-dp")
    os.system("git clone https://huggingface.co/svjack/sd-ggml-cpp-dp")
assert os.path.exists("sd-ggml-cpp-dp")
os.chdir("sd-ggml-cpp-dp")
'''

os.system("pip install huggingface_hub")
#### https://huggingface.co/svjack/sd-ggml-cpp-dp/resolve/main/models/Cyberpunk_Anime_Diffusion-ggml-model_q4_0.bin
def make_and_download_clean_dir(repo_name = "svjack/sd-ggml", 
                               rp_tgt_tail_dict = {
                                   "models": "wget https://huggingface.co/{}/resolve/main/{}/{}"
                               }
                               ):
    import shutil
    import os
    from tqdm import tqdm
    from huggingface_hub import HfFileSystem
    fs = HfFileSystem()
    req_dir = repo_name.split("/")[-1]
    if os.path.exists(req_dir):
        shutil.rmtree(req_dir)
    os.mkdir(req_dir)
    os.chdir(req_dir)
    fd_list = fs.ls(repo_name, detail = False)
    fd_clean_list = list(filter(lambda x: not x.split("/")[-1].startswith("."), fd_list))
    for path in tqdm(fd_clean_list):
        src = path
        tgt = src.split("/")[-1]
        print("downloading {} to {}".format(src, tgt))
        if tgt not in rp_tgt_tail_dict:
            fs.download(
               src, tgt, recursive = True
            )
        else:
            tgt_cmd_format = rp_tgt_tail_dict[tgt]
            os.mkdir(tgt)
            os.chdir(tgt)
            sub_fd_list = fs.ls(src, detail = False)
            for sub_file in tqdm(sub_fd_list):
                tgt_cmd = tgt_cmd_format.format(
                repo_name, tgt, sub_file.split("/")[-1]
                )
                print("run {}".format(tgt_cmd))
                os.system(tgt_cmd)
            os.chdir("..")
    os.chdir("..")
make_and_download_clean_dir("svjack/sd-ggml")
os.chdir("sd-ggml")

assert os.path.exists("stable-diffusion.cpp")
os.system("cmake stable-diffusion.cpp")
os.system("cmake --build . --config Release")
assert os.path.exists("bin")

def process(model_path ,prompt, num_samples, image_resolution, sample_steps, seed,):
    from PIL import Image
    from uuid import uuid1
    output_path = "output_image_dir"
    if not os.path.exists(output_path):
        os.mkdir(output_path)
    else:
        shutil.rmtree(output_path)
        os.mkdir(output_path)
    assert os.path.exists(output_path)

    run_format = './bin/sd -m {} --sampling-method "dpm++2mv2" -o "{}/{}.png" -p "{}" --steps {} -H {} -W {} -s {}'
    images = []
    for i in range(num_samples):
        uid = str(uuid1())
        run_cmd = run_format.format(model_path, output_path,
        uid, prompt, sample_steps, image_resolution,
        image_resolution, seed + i)
        print("run cmd: {}".format(run_cmd))
        os.system(run_cmd)
        assert os.path.exists(os.path.join(output_path, "{}.png".format(uid)))
        image = Image.open(os.path.join(output_path, "{}.png".format(uid)))
        images.append(np.asarray(image))
    results = images
    return results

model_list = list(map(lambda x: os.path.join("models", x), os.listdir("models")))
assert model_list

sdxl_loras_raw = []
with open("sd_cfg.json", "r") as file:
    data = json.load(file)
    sdxl_loras_raw = [
        {
            "image": item["image"],
            "title": item["title"],
            "repo": item["repo"],
            "trigger_word": item["trigger_word"],
            "model_path": item["model_path"]
            #"weights": item["weights"],
            #"is_compatible": item["is_compatible"],
            #"is_pivotal": item.get("is_pivotal", False),
            #"text_embedding_weights": item.get("text_embedding_weights", None),
            #"likes": item.get("likes", 0),
            #"downloads": item.get("downloads", 0),
            #"is_nc": item.get("is_nc", False)
        }
        for item in data
    ]

sdxl_loras_raw = list(filter(lambda d: d["model_path"] in model_list, sdxl_loras_raw))
assert sdxl_loras_raw


def update_selection(selected_state: gr.SelectData, sdxl_loras):
    lora_repo = sdxl_loras[selected_state.index]["repo"]
    instance_prompt = sdxl_loras[selected_state.index]["trigger_word"]
    new_placeholder = "Type a prompt. This applies for all prompts, no need for a trigger word" if instance_prompt == "" else "Type a prompt to use your selected model"
    #weight_name = sdxl_loras[selected_state.index]["weights"]
    updated_text = f"### Selected: [{lora_repo}](https://huggingface.co/{lora_repo}) ✨ "
    is_compatible = True
    is_pivotal = True

    use_with_diffusers = f'''
    ## Using [`{lora_repo}`](https://huggingface.co/{lora_repo})

    ## Use it with diffusers:
    '''
    use_with_uis = f'''
    ## Use it with Comfy UI, Invoke AI, SD.Next, AUTO1111:

    ### Download the `*.safetensors` weights of [here](https://huggingface.co/{lora_repo})

    - [ComfyUI guide](https://comfyanonymous.github.io/ComfyUI_examples/lora/)
    - [Invoke AI guide](https://invoke-ai.github.io/InvokeAI/features/CONCEPTS/?h=lora#using-loras)
    - [SD.Next guide](https://github.com/vladmandic/automatic)
    - [AUTOMATIC1111 guide](https://stable-diffusion-art.com/lora/)
    '''
    return (
        updated_text,
        instance_prompt,
        gr.update(placeholder=new_placeholder),
        selected_state,
        use_with_diffusers,
        use_with_uis,
    )

def check_selected(selected_state):
    if not selected_state:
        raise gr.Error("You must select a Model")

def shuffle_gallery(sdxl_loras):
    random.shuffle(sdxl_loras)
    return [(item["image"], item["title"]) for item in sdxl_loras], sdxl_loras

def swap_gallery(order, sdxl_loras):
    if(order == "random"):
        return shuffle_gallery(sdxl_loras)
    else:
        #sorted_gallery = sorted(sdxl_loras, key=lambda x: x.get(order, 0), reverse=True)
        sorted_gallery = sorted(sdxl_loras, key=lambda x: x["title"], reverse=False)
        return [(item["image"], item["title"]) for item in sorted_gallery], sorted_gallery

'''
def run_lora(prompt, negative, lora_scale, selected_state, sdxl_loras,
        progress=gr.Progress(track_tqdm=True)):
'''
def run_lora(prompt, selected_state, sdxl_loras,
        image_resolution, sample_steps, seed,
        progress=gr.Progress(track_tqdm=True)):
    #global last_lora, last_merged, last_fused, pipe

    '''
    if negative == "":
        negative = None
    '''

    if not selected_state:
        raise gr.Error("You must select a Model")
    repo_name = sdxl_loras[selected_state.index]["repo"]
    model_path = sdxl_loras[selected_state.index]["model_path"]
    #weight_name = sdxl_loras[selected_state.index]["weights"]

    '''
    image = pipe(
        prompt=prompt,
        negative_prompt=negative,
        width=1024,
        height=1024,
        num_inference_steps=20,
        guidance_scale=7.5,
    ).images[0]
    last_lora = repo_name
    gc.collect()
    '''
    num_samples = 1
    #### image_resolution : 512
    #### sample_steps : 8
    #### seed : 20
    image = process(model_path ,prompt, num_samples, image_resolution, sample_steps, seed,)[0]
    image = Image.fromarray(image.astype(np.uint8))
    #return image, gr.update(visible=True)
    return image

with gr.Blocks(css="custom.css") as demo:
    #with gr.Blocks() as demo:
    gr_sdxl_loras = gr.State(value=sdxl_loras_raw)
    title = gr.HTML(
        """<h1><img src="https://i.imgur.com/vT48NAO.png" alt="SD"> StableDiffusion GGML Explorer</h1>""",
        elem_id="title",
    )

    selected_state = gr.State()
    with gr.Row(elem_id="main_app"):
        with gr.Box(elem_id="gallery_box"):
            order_gallery = gr.Radio(choices=["random", "alphabetical"],
            value="random", label="Order by", elem_id="order_radio")
            gallery = gr.Gallery(
                #value=[(item["image"], item["title"]) for item in sdxl_loras_raw],
                label="SD Model Gallery",
                allow_preview=True,
                #rows = 1,
                columns=3,
                #scale = 3,
                min_width = 256,
                #object_fit = "scale-down",
                elem_id="gallery",
                show_share_button=False,
                height=512
            )
        with gr.Column():
            prompt_title = gr.Markdown(
                value="### Click on a Model in the gallery to select it",
                visible=True,
                elem_id="selected_model",
            )
            with gr.Row():
                prompt = gr.Textbox(label="Prompt", show_label=False, lines=1, max_lines=1,
                placeholder="Type a prompt after selecting a Model", elem_id="prompt")
                button = gr.Button("Run", elem_id="run_button")
            '''
            with gr.Group(elem_id="share-btn-container", visible=False) as share_group:
                community_icon = gr.HTML(community_icon_html)
                loading_icon = gr.HTML(loading_icon_html)
                share_button = gr.Button("Share to community", elem_id="share-btn")
            '''
            result = gr.Image(
                interactive=False, label="Generated Image", elem_id="result-image"
            )
            with gr.Accordion("Advanced options", open=False):
                #negative = gr.Textbox(label="Negative Prompt")
                #weight = gr.Slider(0, 10, value=0.8, step=0.1, label="LoRA weight")
                #negative = ""
                image_resolution = gr.Slider(label="Image Resolution", minimum=256, maximum=768, value=512, step=256)
                sample_steps = gr.Slider(label="Steps", minimum=1, maximum=100, value=8, step=1)
                seed = gr.Slider(label="Seed", minimum=-1, maximum=2147483647, step=1, randomize=True)

    order_gallery.change(
        fn=swap_gallery,
        inputs=[order_gallery, gr_sdxl_loras],
        outputs=[gallery, gr_sdxl_loras],
        queue=False
    )
    gallery.select(
        fn=update_selection,
        inputs=[gr_sdxl_loras],
        #outputs=[prompt_title, prompt, prompt, selected_state, use_diffusers, use_uis],
        outputs=[prompt_title, prompt, prompt, selected_state,],
        queue=False,
        show_progress=False
    )
    prompt.submit(
        fn=check_selected,
        inputs=[selected_state],
        queue=False,
        show_progress=False
    ).success(
        fn=run_lora,
        #inputs=[prompt, negative, weight, selected_state, gr_sdxl_loras],
        inputs=[prompt, selected_state, gr_sdxl_loras, image_resolution, sample_steps, seed],
        #outputs=[result, share_group],
        #outputs=[result,],
        outputs = result
    )
    button.click(
        fn=check_selected,
        inputs=[selected_state],
        queue=False,
        show_progress=False
    ).success(
        fn=run_lora,
        #inputs=[prompt, negative, weight, selected_state, gr_sdxl_loras],
        inputs=[prompt, selected_state, gr_sdxl_loras, image_resolution, sample_steps, seed],
        #outputs=[result, share_group],
        #outputs=[result,],
        outputs = result
    )
    #share_button.click(None, [], [], _js=share_js)
    demo.load(fn=shuffle_gallery, inputs=[gr_sdxl_loras], outputs=[gallery, gr_sdxl_loras], queue=False)
demo.queue(max_size=20)
demo.launch()