File size: 8,420 Bytes
a83f80b
4f2c346
 
509d266
 
4f2c346
81eed54
8404815
509d266
6671a55
a83f80b
9660558
 
 
 
7c020ac
4f2c346
 
917f1e0
6c5db00
 
 
 
 
 
 
 
ec39d42
6c5db00
 
 
7c020ac
 
9660558
509d266
 
fe7c35d
509d266
 
 
 
9660558
 
 
509d266
 
ec39d42
 
 
 
 
 
 
fe7c35d
 
 
 
 
 
 
 
 
 
 
 
 
81eed54
fe7c35d
81eed54
 
fe7c35d
 
81eed54
fe7c35d
 
 
 
c239d0a
ec39d42
c239d0a
63644af
ec39d42
63644af
bde04fe
c239d0a
ec39d42
c239d0a
ec39d42
c239d0a
ec39d42
c239d0a
ec39d42
c239d0a
 
 
 
ec39d42
c239d0a
fe7c35d
 
ec39d42
 
c239d0a
fe7c35d
 
ec39d42
509d266
ec39d42
63644af
c239d0a
509d266
 
 
 
81eed54
 
 
 
 
63644af
 
 
 
 
 
 
 
 
dd21895
 
 
 
 
63644af
dd21895
 
 
 
 
63644af
 
 
509d266
917f1e0
c239d0a
63644af
 
 
 
509d266
 
 
 
837f208
 
 
81eed54
ec39d42
63644af
cee83d4
 
 
 
 
 
 
63644af
641f6f7
6ddfa3e
c239d0a
 
 
 
 
 
 
 
f6c0686
 
641f6f7
 
 
 
f6c0686
 
c7317db
f6c0686
 
641f6f7
 
 
 
f6c0686
c239d0a
 
 
641f6f7
6ddfa3e
c239d0a
641f6f7
c239d0a
 
 
 
 
 
 
f6c0686
c239d0a
 
 
 
63644af
7c020ac
fe7c35d
213ce70
c239d0a
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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
import gradio as gr
import fasttext

from transformers import AutoModelForSequenceClassification
from transformers import AutoTokenizer

import numpy as np
import pandas as pd
import torch


id2label = {0: "NEGATIVE", 1: "POSITIVE"}
label2id = {"NEGATIVE": 0, "POSITIVE": 1}


title = "Movie Review Score Discriminator"
description = "It is a program that classifies whether it is positive or negative by entering movie reviews.  \
                You can choose between the Korean version and the English version.  \
                It also provides a version called ""Default"", which determines whether it is Korean or English and predicts it."


class LanguageIdentification:
    def __init__(self):
        pretrained_lang_model = "./lid.176.ftz"
        self.model = fasttext.load_model(pretrained_lang_model)

    def predict_lang(self, text):
        predictions = self.model.predict(text, k=200) # returns top 200 matching languages
        return predictions

LANGUAGE = LanguageIdentification()



def tokenized_data(tokenizer, inputs):
    return tokenizer.batch_encode_plus(
        [inputs],
        return_tensors="pt",
        padding="max_length",
        max_length=64,
        truncation=True)



examples = []
df = pd.read_csv('examples.csv', sep='\t', index_col='Unnamed: 0')
np.random.seed(100)

idx = np.random.choice(50, size=5, replace=False)
eng_examples = [ ['Eng', df.iloc[i, 0]] for i in idx ]
kor_examples = [ ['Kor', df.iloc[i, 1]] for i in idx ]
examples = eng_examples + kor_examples



eng_model_name = "roberta-base"
eng_step = 1900
eng_tokenizer = AutoTokenizer.from_pretrained(eng_model_name)
eng_file_name = "{}-{}.pt".format(eng_model_name, eng_step)
eng_state_dict = torch.load(eng_file_name)
eng_model = AutoModelForSequenceClassification.from_pretrained(
    eng_model_name, num_labels=2, id2label=id2label, label2id=label2id,
    state_dict=eng_state_dict
)


kor_model_name = "klue/roberta-small"
kor_step = 2400
kor_tokenizer = AutoTokenizer.from_pretrained(kor_model_name)
kor_file_name = "{}-{}.pt".format(kor_model_name.replace('/', '_'), kor_step)
kor_state_dict = torch.load(kor_file_name)
kor_model = AutoModelForSequenceClassification.from_pretrained(
    kor_model_name, num_labels=2, id2label=id2label, label2id=label2id,
    state_dict=kor_state_dict
)


def builder(Lang, Text):
    percent_kor, percent_eng = 0, 0
    text_list = Text.split(' ')


    # [ output_1 ]
    if Lang == '์–ธ์–ด๊ฐ์ง€ ๊ธฐ๋Šฅ ์‚ฌ์šฉ':
        pred = LANGUAGE.predict_lang(Text)
        if '__label__en' in pred[0]:
            Lang = 'Eng'
            idx = pred[0].index('__label__en')
            p_eng = pred[1][idx]
        if '__label__ko' in pred[0]:
            Lang = 'Kor'
            idx = pred[0].index('__label__ko')
            p_kor = pred[1][idx]
        # Normalize Percentage
        percent_kor = p_kor / (p_kor+p_eng)
        percent_eng = p_eng / (p_kor+p_eng)

    if Lang == 'Eng':
        model = eng_model
        tokenizer = eng_tokenizer
        if percent_eng==0: percent_eng=1

    if Lang == 'Kor':
        model = kor_model
        tokenizer = kor_tokenizer
        if percent_kor==0: percent_kor=1
        

    # [ output_2 ]
    inputs = tokenized_data(tokenizer, Text)
    model.eval()
    with torch.no_grad():
        logits = model(input_ids=inputs['input_ids'], 
            attention_mask=inputs['attention_mask']).logits
    
    m = torch.nn.Softmax(dim=1)
    output = m(logits)
    # print(logits, output)


    # [ output_3 ]
    output_analysis = []
    for word in text_list:
        tokenized_word = tokenized_data(tokenizer, word)
        with torch.no_grad():
            logit = model(input_ids=tokenized_word['input_ids'], 
                attention_mask=tokenized_word['attention_mask']).logits
        word_output = m(logit)
        if word_output[0][1] > 0.99:
            output_analysis.append( (word, '+++') )
        elif word_output[0][1] > 0.9:
            output_analysis.append( (word, '++') )
        elif word_output[0][1] > 0.8:
            output_analysis.append( (word, '+') )
        elif word_output[0][1] < 0.01:
            output_analysis.append( (word, '---') )
        elif word_output[0][1] < 0.1:
            output_analysis.append( (word, '--') )
        elif word_output[0][1] < 0.2:
            output_analysis.append( (word, '-') )
        else:
            output_analysis.append( (word, None) )
    

    return [ {'Kor': percent_kor, 'Eng': percent_eng}, 
            {id2label[1]: output[0][1].item(), id2label[0]: output[0][0].item()}, 
            output_analysis ]
            
    # prediction = torch.argmax(logits, axis=1)
    return id2label[prediction.item()]


# demo3 = gr.Interface.load("models/mdj1412/movie_review_score_discriminator_eng", inputs="text", outputs="text", 
#                          title=title, theme="peach",
#                          allow_flagging="auto",
#                          description=description, examples=examples)



# demo = gr.Interface(builder, inputs=[gr.inputs.Dropdown(['Default', 'Eng', 'Kor']), gr.Textbox(placeholder="๋ฆฌ๋ทฐ๋ฅผ ์ž…๋ ฅํ•˜์‹œ์˜ค.")], 
#                     outputs=[ gr.Label(num_top_classes=3, label='Lang'), 
#                             gr.Label(num_top_classes=2, label='Result'),
#                             gr.HighlightedText(label="Analysis", combine_adjacent=False)
#                             .style(color_map={"+++": "#CF0000", "++": "#FF3232", "+": "#FFD4D4", "---": "#0004FE", "--": "#4C47FF", "-": "#BEBDFF"}) ],
#                     # outputs='label',
#                     title=title, description=description, examples=examples)



with gr.Blocks() as demo1:
    gr.Markdown(
    """
    <h1 align="center">
    Movie Review Score Discriminator
    </h1>
    """)

    gr.Markdown(
    """
    ์˜ํ™” ๋ฆฌ๋ทฐ๋ฅผ ์ž…๋ ฅํ•˜๋ฉด, ๋ฆฌ๋ทฐ๊ฐ€ ๊ธ์ •์ธ์ง€ ๋ถ€์ •์ธ์ง€ ํŒ๋ณ„ํ•ด์ฃผ๋Š” ๋ชจ๋ธ์ด๋‹ค. \
    ์˜์–ด์™€ ํ•œ๊ธ€์„ ์ง€์›ํ•˜๋ฉฐ, ์–ธ์–ด๋ฅผ ์ง์ ‘ ์„ ํƒํ• ์ˆ˜๋„, ํ˜น์€ ๋ชจ๋ธ์ด ์–ธ์–ด๊ฐ์ง€๋ฅผ ์ง์ ‘ ํ•˜๋„๋ก ํ•  ์ˆ˜ ์žˆ๋‹ค.  
    ๋ฆฌ๋ทฐ๋ฅผ ์ž…๋ ฅํ•˜๋ฉด, (1) ๊ฐ์ง€๋œ ์–ธ์–ด, (2) ๊ธ์ • ๋ฆฌ๋ทฐ์ผ ํ™•๋ฅ ๊ณผ ๋ถ€์ • ๋ฆฌ๋ทฐ์ผ ํ™•๋ฅ , (3) ์ž…๋ ฅ๋œ ๋ฆฌ๋ทฐ์˜ ์–ด๋Š ๋‹จ์–ด๊ฐ€ ๊ธ์ •/๋ถ€์ • ๊ฒฐ์ •์— ์˜ํ–ฅ์„ ์ฃผ์—ˆ๋Š”์ง€ \
    (๊ธ์ •์ผ ๊ฒฝ์šฐ ๋นจ๊ฐ•์ƒ‰, ๋ถ€์ •์ผ ๊ฒฝ์šฐ ํŒŒ๋ž€์ƒ‰)๋ฅผ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค.
    """)

    with gr.Accordion(label="๋ชจ๋ธ์— ๋Œ€ํ•œ ์„ค๋ช… ( ์—ฌ๊ธฐ๋ฅผ ํด๋ฆญ ํ•˜์‹œ์˜ค. )", open=False):
        gr.Markdown(
        """
        ์˜์–ด ๋ชจ๋ธ์€ bert-base-uncased ๊ธฐ๋ฐ˜์œผ๋กœ, ์˜์–ด ์˜ํ™” ๋ฆฌ๋ทฐ ๋ถ„์„ ๋ฐ์ดํ„ฐ์…‹์ธ SST-2๋กœ ํ•™์Šต ๋ฐ ํ‰๊ฐ€๋˜์—ˆ๋‹ค.  
        ํ•œ๊ธ€ ๋ชจ๋ธ์€ klue/roberta-base ๊ธฐ๋ฐ˜์ด๋‹ค. ๊ธฐ์กด ํ•œ๊ธ€ ์˜ํ™” ๋ฆฌ๋ทฐ ๋ถ„์„ ๋ฐ์ดํ„ฐ์…‹์ด ์กด์žฌํ•˜์ง€ ์•Š์•„, ๋„ค์ด๋ฒ„ ์˜ํ™”์˜ ๋ฆฌ๋ทฐ๋ฅผ ํฌ๋กค๋งํ•ด์„œ ์˜ํ™” ๋ฆฌ๋ทฐ ๋ถ„์„ ๋ฐ์ดํ„ฐ์…‹์„ ์ œ์ž‘ํ•˜๊ณ , ์ด๋ฅผ ์ด์šฉํ•˜์—ฌ ๋ชจ๋ธ์„ ํ•™์Šต ๋ฐ ํ‰๊ฐ€ํ•˜์˜€๋‹ค.  
        ์˜์–ด ๋ชจ๋ธ์€ SST-2์—์„œ 92.8%, ํ•œ๊ธ€ ๋ชจ๋ธ์€ ๋„ค์ด๋ฒ„ ์˜ํ™” ๋ฆฌ๋ทฐ ๋ฐ์ดํ„ฐ์…‹์—์„œ 94%์˜ ์ •ํ™•๋„๋ฅผ ๊ฐ€์ง„๋‹ค (test set ๊ธฐ์ค€).  
        ์–ธ์–ด๊ฐ์ง€๋Š” fasttext์˜ language detector๋ฅผ ์‚ฌ์šฉํ•˜์˜€๋‹ค. ๋ฆฌ๋ทฐ์˜ ๋‹จ์–ด๋ณ„ ์˜ํ–ฅ๋ ฅ์€, ๋‹จ์–ด ๊ฐ๊ฐ์„ ๋ชจ๋ธ์— ๋„ฃ์—ˆ์„ ๋•Œ ๊ฒฐ๊ณผ๊ฐ€ ๊ธ์ •์œผ๋กœ ๋‚˜์˜ค๋Š”์ง€ ๋ถ€์ •์œผ๋กœ ๋‚˜์˜ค๋Š”์ง€๋ฅผ ๋ฐ”ํƒ•์œผ๋กœ ์ธก์ •ํ•˜์˜€๋‹ค.
        """)

    with gr.Row():
        with gr.Column():
            inputs_1 = gr.Dropdown(choices=['์–ธ์–ด๊ฐ์ง€ ๊ธฐ๋Šฅ ์‚ฌ์šฉ', 'Eng', 'Kor'], value='์–ธ์–ด๊ฐ์ง€ ๊ธฐ๋Šฅ ์‚ฌ์šฉ', label='Lang')
            inputs_2 = gr.Textbox(placeholder="๋ฆฌ๋ทฐ๋ฅผ ์ž…๋ ฅํ•˜์‹œ์˜ค.", label='Text')
            with gr.Row():
                # btn2 = gr.Button("ํด๋ฆฌ์–ด")
                btn = gr.Button("์ œ์ถœํ•˜๊ธฐ")
        with gr.Column():
            output_1 = gr.Label(num_top_classes=3, label='Lang')
            output_2 = gr.Label(num_top_classes=2, label='Result')
            output_3 = gr.HighlightedText(label="Analysis", combine_adjacent=False) \
                .style(color_map={"+++": "#CF0000", "++": "#FF3232", "+": "#FFD4D4", "---": "#0004FE", "--": "#4C47FF", "-": "#BEBDFF"})
    
    # btn2.click(fn=fn2, inputs=[None, None], output=[output_1, output_2, output_3])
    btn.click(fn=builder, inputs=[inputs_1, inputs_2], outputs=[output_1, output_2, output_3])
    gr.Examples(examples, inputs=[inputs_1, inputs_2])
    


if __name__ == "__main__":
    # print(examples)
    # demo.launch()
    demo1.launch()