Hugging Face's logo

Spaces:
fffiloni
/
CLIP-Interrogator-2
Running on T4

App Files Community
86
CLIP-Interrogator-2
File size: 4,898 Bytes 
0e633ca
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
de4a512
 
 
 
 
 
 
 
 
 
df4cffb
de4a512
 
 
df4cffb
de4a512
 
 
 
 
df4cffb
de4a512
 
 
 
 
 
 
 
 
 
 
 
 
 
 
df4cffb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0e633ca
df4cffb
 
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
 
#@title Setup
import os, subprocess

def setup():
    install_cmds = [
        ['pip', 'install', 'ftfy', 'gradio', 'regex', 'tqdm', 'transformers==4.21.2', 'timm', 'fairscale', 'requests'],
        ['pip', 'install', 'open_clip_torch'],
        ['pip', 'install', '-e', 'git+https://github.com/pharmapsychotic/BLIP.git@lib#egg=blip'],
        ['git', 'clone', '-b', 'open-clip', 'https://github.com/pharmapsychotic/clip-interrogator.git']
    ]
    for cmd in install_cmds:
        print(subprocess.run(cmd, stdout=subprocess.PIPE).stdout.decode('utf-8'))

setup()

# download cache files
print("Download preprocessed cache files...")
CACHE_URLS = [
    'https://huggingface.co/pharma/ci-preprocess/resolve/main/ViT-H-14_laion2b_s32b_b79k_artists.pkl',
    'https://huggingface.co/pharma/ci-preprocess/resolve/main/ViT-H-14_laion2b_s32b_b79k_flavors.pkl',
    'https://huggingface.co/pharma/ci-preprocess/resolve/main/ViT-H-14_laion2b_s32b_b79k_mediums.pkl',
    'https://huggingface.co/pharma/ci-preprocess/resolve/main/ViT-H-14_laion2b_s32b_b79k_movements.pkl',
    'https://huggingface.co/pharma/ci-preprocess/resolve/main/ViT-H-14_laion2b_s32b_b79k_trendings.pkl',
]
os.makedirs('cache', exist_ok=True)
for url in CACHE_URLS:
    print(subprocess.run(['wget', url, '-P', 'cache'], stdout=subprocess.PIPE).stdout.decode('utf-8'))

import sys
sys.path.append('src/blip')
sys.path.append('clip-interrogator')

import gradio as gr
from clip_interrogator import Config, Interrogator

config = Config()
config.blip_offload = True
config.chunk_size = 2048
config.flavor_intermediate_count = 512
config.blip_num_beams = 64

ci = Interrogator(config)

def inference(image, mode, best_max_flavors):
    image = image.convert('RGB')
    if mode == 'best':
        return ci.interrogate(image, max_flavors=int(best_max_flavors))
    elif mode == 'classic':
        return ci.interrogate_classic(image)
    else:
        return ci.interrogate_fast(image)

title = """
    <div style="text-align: center; max-width: 650px; margin: 0 auto;">
        <div
        style="
            display: inline-flex;
            align-items: center;
            gap: 0.8rem;
            font-size: 1.75rem;
        "
        >
        <h1 style="font-weight: 600; margin-bottom: 7px;">
            CLIP Interrogator 2.1
        </h1>
        </div>
        <p style="margin-bottom: 10px;font-size: 16px;font-weight: 100;line-height: 1.5em;">
        Want to figure out what a good prompt might be to create new images like an existing one? The CLIP Interrogator is here to get you answers!
        <br />This version is specialized for producing nice prompts for use with Stable Diffusion 2.0 using the ViT-H-14 OpenCLIP model!
        </p>
    </div>
"""

article = """
<div style="text-align: center; max-width: 650px; margin: 0 auto;">
    
    <p>
    Server busy? You can also run on <a href="https://colab.research.google.com/github/pharmapsychotic/clip-interrogator/blob/open-clip/clip_interrogator.ipynb">Google Colab</a>
    </p>
    <p>
    Has this been helpful to you? Follow Pharma on twitter 
    <a href="https://twitter.com/pharmapsychotic">@pharmapsychotic</a> 
    and check out more tools at his
    <a href="https://pharmapsychotic.com/tools.html">Ai generative art tools list</a>
    </p>
</div>
"""

css = '''
#col-container {max-width: 700px; margin-left: auto; margin-right: auto;}
a {text-decoration-line: underline; font-weight: 600;}
.animate-spin {
    animation: spin 1s linear infinite;
}
@keyframes spin {
    from {
        transform: rotate(0deg);
    }
    to {
        transform: rotate(360deg);
    }
}
#share-btn-container {
    display: flex; padding-left: 0.5rem !important; padding-right: 0.5rem !important; background-color: #000000; justify-content: center; align-items: center; border-radius: 9999px !important; width: 13rem;
}
#share-btn {
    all: initial; color: #ffffff;font-weight: 600; cursor:pointer; font-family: 'IBM Plex Sans', sans-serif; margin-left: 0.5rem !important; padding-top: 0.25rem !important; padding-bottom: 0.25rem !important;
}
#share-btn * {
    all: unset;
}
#share-btn-container div:nth-child(-n+2){
    width: auto !important;
    min-height: 0px !important;
}
#share-btn-container .wrap {
    display: none !important;
}
'''

with gr.Blocks(css=css) as block:
    with gr.Column(elem_id="col-container"):
        gr.HTML(title)

        input_image = gr.Image(type='pil', elem_id="input-img")
        mode_input = gr.Radio(['best', 'classic', 'fast'], label='', value='best')
        flavor_input = gr.Number(value=4, label='best mode max flavors')
        submit_btn = gr.Button("Submit")
        output_text = gr.Textbox(label="Output", elem_id="output-txt")
       
        gr.HTML(article)

    submit_btn.click(fn=inference, inputs=[input_image,mode_input,flavor_input], outputs=[output_text])

block.queue(max_size=32).launch(show_api=False)