File size: 5,562 Bytes
536d153
f87d76f
 
 
67d94fb
536d153
 
 
 
67d94fb
 
 
 
 
 
536d153
 
 
 
 
 
 
 
 
 
 
 
0adf250
536d153
 
 
 
 
 
 
0adf250
536d153
 
 
 
 
 
 
 
 
 
 
 
 
eb7ac3e
e96f8bf
 
 
 
 
eb7ac3e
 
 
 
 
 
 
 
 
 
e96f8bf
536d153
 
 
e96f8bf
536d153
 
e96f8bf
 
 
 
97a751a
536d153
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
eb7ac3e
 
 
 
536d153
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
eb7ac3e
536d153
 
 
 
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
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
import gradio as gr
import os
import numpy as np
import random
from huggingface_hub import login
import torch
from diffusers import StableDiffusionXLPipeline, AutoencoderKL
from blora_utils import BLOCKS, filter_lora, scale_lora

hf_token = os.environ.get("HF_TOKEN")
login(token=hf_token)

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 1024

vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
pipeline = StableDiffusionXLPipeline.from_pretrained(
        "stabilityai/stable-diffusion-xl-base-1.0",
        vae=vae,
        torch_dtype=torch.float16,
    ).to("cuda")

def load_b_lora_to_unet(pipe, content_lora_model_id: str = '', style_lora_model_id: str = '', content_alpha: float = 1.,
                            style_alpha: float = 1.) -> None:
        try:
            # Get Content B-LoRA SD
            if content_lora_model_id:
                content_B_LoRA_sd, _ = pipe.lora_state_dict(content_lora_model_id, use_auth_token=True)
                content_B_LoRA = filter_lora(content_B_LoRA_sd, BLOCKS['content'])
                content_B_LoRA = scale_lora(content_B_LoRA, content_alpha)
            else:
                content_B_LoRA = {}

            # Get Style B-LoRA SD
            if style_lora_model_id:
                style_B_LoRA_sd, _ = pipe.lora_state_dict(style_lora_model_id, use_auth_token=True)
                style_B_LoRA = filter_lora(style_B_LoRA_sd, BLOCKS['style'])
                style_B_LoRA = scale_lora(style_B_LoRA, style_alpha)
            else:
                style_B_LoRA = {}

            # Merge B-LoRAs SD
            res_lora = {**content_B_LoRA, **style_B_LoRA}

            # Load
            pipe.load_lora_into_unet(res_lora, None, pipe.unet)
        except Exception as e:
            raise type(e)(f'failed to load_b_lora_to_unet, due to: {e}')

def main(content_b_lora, style_b_lora, prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):

    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
        
    generator = torch.Generator().manual_seed(seed)

    if content_b_lora is None:
        content_B_LoRA_path = ''
    else:
        content_B_LoRA_path = content_b_lora

    if style_b_lora is None:
        style_B_LoRA_path = ''
    else:
        style_B_LoRA_path = style_b_lora
    
    content_alpha,style_alpha = 1,1.1

    load_b_lora_to_unet(pipeline, content_B_LoRA_path, style_B_LoRA_path, content_alpha, style_alpha)
    prompt = prompt
    image = pipeline(
        prompt,
        generator=generator, 
        num_images_per_prompt=1,
        width = width, 
        height = height,
    ).images[0]
    
    pipeline.unload_lora_weights()
    
    return image

css="""
#col-container {
    margin: 0 auto;
    max-width: 520px;
}
"""

if torch.cuda.is_available():
    power_device = "GPU"
else:
    power_device = "CPU"

with gr.Blocks(css=css) as demo:
    
    with gr.Column(elem_id="col-container"):
        gr.Markdown(f"""
        # Text-to-Image Gradio Template
        Currently running on {power_device}.
        """)

        with gr.Row():
            content_b_lora = gr.Textbox(label="B-LoRa for content")
            style_b_lora = gr.Textbox(label="B-LoRa for style")
        
        with gr.Row():
            
            prompt = gr.Text(
                label="Prompt",
                show_label=False,
                max_lines=1,
                placeholder="Enter your prompt",
                container=False,
            )
            
            run_button = gr.Button("Run", scale=0)
        
        result = gr.Image(label="Result", show_label=False)

        with gr.Accordion("Advanced Settings", open=False):
            
            negative_prompt = gr.Text(
                label="Negative prompt",
                max_lines=1,
                placeholder="Enter a negative prompt",
                visible=False,
            )
            
            seed = gr.Slider(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=0,
            )
            
            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
            
            with gr.Row():
                
                width = gr.Slider(
                    label="Width",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=512,
                )
                
                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=512,
                )
            
            with gr.Row():
                
                guidance_scale = gr.Slider(
                    label="Guidance scale",
                    minimum=0.0,
                    maximum=10.0,
                    step=0.1,
                    value=0.0,
                )
                
                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=50,
                )

    run_button.click(
        fn = main,
        inputs = [content_b_lora, style_b_lora, prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
        outputs = [result]
    )

demo.queue().launch()