File size: 2,030 Bytes
489c564
dd90f68
49ec14a
ef3e238
f124040
 
d582d22
f124040
dc0dc06
f124040
ef3e238
49ec14a
9adf938
8e48d3e
fadfca7
8d0ce31
8e48d3e
cfa5afa
9adf938
 
4e07626
 
 
 
 
8e5c897
4e07626
19dae2f
2b84e99
19dae2f
8e48d3e
4a6889b
4e07626
 
 
cd9db19
4e8b19e
cd9db19
8e48d3e
79f02c0
4e8b19e
 
79f02c0
8e48d3e
79f02c0
823a19b
6b5a3ad
c9de4a3
8e48d3e
 
4e07626
74e89b2
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
import gradio as gr
from transformers import pipeline

examples = [
    'Alisher Navoiy – ulug‘ o‘zbek va boshqa turkiy xalqlarning <mask>, mutafakkiri va davlat arbobi bo‘lgan.',
    'Oʻzbekistonning poytaxti <mask> shahri boʻlib, davlat tili oʻzbek tili hisoblanadi.',
    'Registon maydoni - tarixda shaharning ilm-fan, siyosat va <mask> markazi boʻlgan.',
    'Venera - Quyosh tizimidagi o‘z o‘qi atrofida soat sohasi farqli ravishda aylanadigan yagona <mask>.',
    'Kuchli yomg‘irlar tufayli bir qator <mask> kuchli sel oqishi kuzatildi.',
    'Oʻzbekiston iqtisodiyoti bozor <mask> bosqichma-bosqich oʻtadi, tashqi savdo siyosati import oʻrnini bosishga asoslangan.'
]

models = [
    "sinonimayzer/UzRoBERTa-v1", 
    "sinonimayzer/UzRoBERTa-v2", 
    "sinonimayzer/UzRoBERTa-v2", 
    "rifkat/uztext-3Gb-BPE-Roberta",
    "tahrirchi/tahrirchi-bert-base", 
]

def df(arr):
    d = {}
    for val in arr:
        d[val['token_str']] = val['score']
    return d
    
def fn(text):
    arr = []
    for model in models:
        arr.append(df(pipeline("fill-mask", model=model)(text)))
    return arr[0], arr[1], arr[2], arr[3], arr[4]
    
with gr.Blocks() as demo:
    with gr.Row():
        with gr.Column():
            output0 = gr.Label(label=models[0])
        with gr.Column():
            output1 = gr.Label(label=models[1])
        with gr.Column():
            output2 = gr.Label(label=models[2])
    with gr.Row():
        with gr.Column():
            output3 = gr.Label(label=models[3])
        with gr.Column():
            output4 = gr.Label(label=models[4])
        with gr.Column():
            input = gr.Textbox(label="Input", value=examples[0], lines=8, max_lines=8)
            btn = gr.Button("Check")
    gr.Examples(examples, fn=fn, inputs=[input], outputs=[output0, output1, output2, output3, output4], cache_examples=True, batch=True)
    btn.click(fn, inputs=[input], outputs=[output0, output1, output2, output3, output4])
if __name__ == "__main__":
    demo.queue().launch()