File size: 4,075 Bytes
863f08e
 
 
 
 
 
 
863413c
863f08e
 
e61c05b
863f08e
 
 
 
 
 
e61c05b
 
863f08e
 
 
 
 
 
 
863413c
 
 
 
53ddaff
863413c
 
 
124feae
863f08e
 
 
e61c05b
 
 
 
 
 
 
 
 
 
e5f2339
863413c
 
e61c05b
 
 
863413c
cec3a8b
863f08e
 
 
 
 
 
 
 
 
 
863413c
 
863f08e
 
863413c
 
863f08e
863413c
 
863f08e
 
863413c
 
 
 
 
863f08e
863413c
 
 
 
 
863f08e
863413c
 
863f08e
 
 
 
863413c
863f08e
 
863413c
 
 
 
 
 
 
 
863f08e
53ddaff
 
 
 
 
 
 
 
 
863413c
863f08e
 
 
 
 
 
863413c
863f08e
863413c
 
863f08e
 
 
 
 
863413c
863f08e
 
863413c
 
 
863f08e
 
 
 
 
 
 
 
 
124feae
 
 
863f08e
cec3a8b
863f08e
 
 
df4b467
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
import gradio as gr
import numpy as np
import random

# import spaces #[uncomment to use ZeroGPU]
from diffusers import DiffusionPipeline
import torch
from typing import Optional

device = "cuda" if torch.cuda.is_available() else "cpu"
model_id_default = "CompVis/stable-diffusion-v1-4"

if torch.cuda.is_available():
    torch_dtype = torch.float16
else:
    torch_dtype = torch.float32

pipe_default = DiffusionPipeline.from_pretrained(model_id_default, torch_dtype=torch_dtype)
pipe_default = pipe_default.to(device)

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


# @spaces.GPU #[uncomment to use ZeroGPU]
def infer(
    prompt: str,
    negative_prompt: str,
    width: int,
    height: int,
    num_inference_steps: Optional[int] = 20,
    model_id: Optional[str] = 'CompVis/stable-diffusion-v1-4',
    seed: Optional[int] = 42,
    guidance_scale: Optional[float] = 7.0,
    progress=gr.Progress(track_tqdm=True),
):
    generator = torch.Generator().manual_seed(seed)

    params = {
        'prompt': prompt,
        'negative_prompt': negative_prompt,
        'guidance_scale': guidance_scale,
        'num_inference_steps': num_inference_steps,
        'width': width,
        'height': height,
        'generator': generator,
    }

    if model_id != model_id_default:
        pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch_dtype)
        pipe = pipe.to(device)
        image = pipe(**params).images[0]
    else:
        image = pipe_default(**params).images[0]

    return image

css = """
#col-container {
    margin: 0 auto;
    max-width: 640px;
}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        
        gr.Markdown(" # DEMO Text-to-Image")

        with gr.Row():
            model_id = gr.Textbox(
                label="Model ID",
                max_lines=1,
                placeholder="Enter model id like 'CompVis/stable-diffusion-v1-4'",
                value="CompVis/stable-diffusion-v1-4"
            )

        prompt = gr.Textbox(
            label="Prompt",
            max_lines=1,
            placeholder="Enter your prompt",
        )

        negative_prompt = gr.Textbox(
            label="Negative prompt",
            max_lines=1,
            placeholder="Enter a negative prompt",
        )

        with gr.Row():
            seed = gr.Number(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=42,
            )

        with gr.Row():
            guidance_scale = gr.Slider(
                label="Guidance scale",
                minimum=0.0,
                maximum=10.0,
                step=0.1,
                value=7.0,
            )

        with gr.Row():
            num_inference_steps = gr.Slider(
                label="Number of inference steps",
                minimum=1,
                maximum=50,
                step=1,
                value=20,
            )

        with gr.Accordion("Optional Settings", open=False):
            with gr.Row():
                width = gr.Slider(
                    label="Width",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )
            
            with gr.Row():
                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )

        run_button = gr.Button("Run", scale=1, variant="primary")
        result = gr.Image(label="Result", show_label=False)
    
    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[
            prompt,
            negative_prompt,
            width,
            height,
            num_inference_steps,
            model_id,
            seed,
            guidance_scale,
        ],
        outputs=[result],
    )

if __name__ == "__main__":
    demo.launch()