-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathNLP_korean_RNN.html
executable file
·1067 lines (718 loc) · 76.2 KB
/
NLP_korean_RNN.html
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
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html>
<head>
<!-- Document Settings -->
<meta charset="utf-8" />
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
<!-- On Post front-matter YAML, set "use_math: true" to use LaTex -->
<!-- Base Meta -->
<!-- dynamically fixing the title for tag/author pages -->
<title>NLP - Korean Language Text Analysis with RNN</title>
<meta name="HandheldFriendly" content="True" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<!-- Styles'n'Scripts -->
<link rel="stylesheet" type="text/css" href="/assets/built/screen.css" />
<link rel="stylesheet" type="text/css" href="/assets/built/screen.edited.css" />
<link rel="stylesheet" type="text/css" href="/assets/built/syntax.css" />
<!-- syntax.css -->
<link rel="stylesheet" type="text/css" href="/assets/built/syntax.css" />
<!-- highlight.js -->
<link rel="stylesheet" href="//cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/styles/default.min.css">
<style>.hljs { background: none; }</style>
<!--[if IE]>
<style>
p, ol, ul{
width: 100%;
}
blockquote{
width: 100%;
}
</style>
<![endif]-->
<!-- This tag outputs SEO meta+structured data and other important settings -->
<meta name="description" content="" />
<link rel="shortcut icon" href="http://0.0.0.0:4000/assets/built/images/favicon.jpg" type="image/png" />
<link rel="canonical" href="http://0.0.0.0:4000/NLP_korean_RNN" />
<meta name="referrer" content="no-referrer-when-downgrade" />
<!--title below is coming from _includes/dynamic_title-->
<meta property="og:site_name" content="Darron's Devlog" />
<meta property="og:type" content="website" />
<meta property="og:title" content="NLP - Korean Language Text Analysis with RNN" />
<meta property="og:description" content="1. Data Load 2. Preprocessing 2.1. Remove duplicates 2.2. Regexp on Korean Language 3. Tokenizing with konlpy-Okt 4. Train-test Data 5. LSTM Model # 한국어 자료 import sys import os import numpy as np import nltk import konlpy import pandas as pd import re import random import itertools import warnings" />
<meta property="og:url" content="http://0.0.0.0:4000/NLP_korean_RNN" />
<meta property="og:image" content="http://0.0.0.0:4000/assets/built/images/blog-cover1.png" />
<meta property="article:publisher" content="https://www.facebook.com/" />
<meta property="article:author" content="https://www.facebook.com/" />
<meta property="article:published_time" content="2020-06-18T15:00:00+00:00" />
<meta property="article:modified_time" content="2020-06-18T15:00:00+00:00" />
<meta property="article:tag" content="projects" />
<meta name="twitter:card" content="summary_large_image" />
<meta name="twitter:title" content="NLP - Korean Language Text Analysis with RNN" />
<meta name="twitter:description" content="1. Data Load 2. Preprocessing 2.1. Remove duplicates 2.2. Regexp on Korean Language 3. Tokenizing with konlpy-Okt 4. Train-test Data 5. LSTM Model # 한국어 자료 import sys import os import numpy as np import nltk import konlpy import pandas as pd import re import random import itertools import warnings" />
<meta name="twitter:url" content="http://0.0.0.0:4000/" />
<meta name="twitter:image" content="http://0.0.0.0:4000/assets/built/images/blog-cover1.png" />
<meta name="twitter:label1" content="Written by" />
<meta name="twitter:data1" content="Darron's Devlog" />
<meta name="twitter:label2" content="Filed under" />
<meta name="twitter:data2" content="projects" />
<meta name="twitter:site" content="@" />
<meta name="twitter:creator" content="@" />
<meta property="og:image:width" content="1400" />
<meta property="og:image:height" content="933" />
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Website",
"publisher": {
"@type": "Organization",
"name": "Darron's Devlog",
"logo": "http://0.0.0.0:4000/"
},
"url": "http://0.0.0.0:4000/NLP_korean_RNN",
"image": {
"@type": "ImageObject",
"url": "http://0.0.0.0:4000/assets/built/images/blog-cover1.png",
"width": 2000,
"height": 666
},
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "http://0.0.0.0:4000/NLP_korean_RNN"
},
"description": "1. Data Load 2. Preprocessing 2.1. Remove duplicates 2.2. Regexp on Korean Language 3. Tokenizing with konlpy-Okt 4. Train-test Data 5. LSTM Model # 한국어 자료 import sys import os import numpy as np import nltk import konlpy import pandas as pd import re import random import itertools import warnings"
}
</script>
<!-- <script type="text/javascript" src="https://demo.ghost.io/public/ghost-sdk.min.js?v=724281a32e"></script>
<script type="text/javascript">
ghost.init({
clientId: "ghost-frontend",
clientSecret: "f84a07a72b17"
});
</script> -->
<meta name="generator" content="Jekyll 3.6.2" />
<link rel="alternate" type="application/rss+xml" title="NLP - Korean Language Text Analysis with RNN" href="/feed.xml" />
</head>
<body class="post-template">
<div class="site-wrapper">
<!-- All the main content gets inserted here, index.hbs, post.hbs, etc -->
<!-- default -->
<!-- The tag above means: insert everything in this file
into the {body} of the default.hbs template -->
<header class="site-header outer">
<div class="inner">
<nav class="site-nav">
<div class="site-nav-left">
<a class="site-nav-logo" href="/">Darron's Devlog</a>
<ul class="nav" role="menu">
<li class="nav-home" role="menuitem"><a href="/">Home</a></li>
<li class="nav-about" role="menuitem"><a href="/about/">About</a></li>
<li class="nav-projects" role="menuitem"><a href="/tag/projects/">Projects</a></li>
<li class="nav-studies" role="menuitem"><a href="/tag/studies/">Studies</a></li>
<li class="nav-blog" role="menuitem"><a href="/tag/blog/">Blog</a></li>
<li class="nav-archive" role="menuitem">
<a href="/archive.html">All Posts</a>
</li>
</ul>
</div>
<div class="site-nav-right">
<div class="social-links">
</div>
<a class="subscribe-button" href="#subscribe">Search</a>
</div>
</nav>
</div>
</header>
<!-- Everything inside the #post tags pulls data from the post -->
<!-- #post -->
<main id="site-main" class="site-main outer" role="main">
<div class="inner">
<article class="post-full tag-projects no-image">
<header class="post-full-header">
<section class="post-full-meta">
<time class="post-full-meta-date" datetime="18 June 2020">18 June 2020</time>
<span class="date-divider">/</span>
<a href='/tag/projects/'>PROJECTS</a>
</section>
<h1 class="post-full-title">NLP - Korean Language Text Analysis with RNN</h1>
</header>
<!--
-->
<section class="post-full-content">
<div class="kg-card-markdown">
<ul id="markdown-toc">
<li><a href="#1-data-load" id="markdown-toc-1-data-load">1. Data Load</a></li>
<li><a href="#2-preprocessing" id="markdown-toc-2-preprocessing">2. Preprocessing</a> <ul>
<li><a href="#21-remove-duplicates" id="markdown-toc-21-remove-duplicates">2.1. Remove duplicates</a></li>
<li><a href="#22-regexp-on-korean-language" id="markdown-toc-22-regexp-on-korean-language">2.2. Regexp on Korean Language</a></li>
</ul>
</li>
<li><a href="#3-tokenizing-with-konlpy-okt" id="markdown-toc-3-tokenizing-with-konlpy-okt">3. Tokenizing with konlpy-Okt</a></li>
<li><a href="#4-train-test-data" id="markdown-toc-4-train-test-data">4. Train-test Data</a></li>
<li><a href="#5-lstm-model" id="markdown-toc-5-lstm-model">5. LSTM Model</a></li>
</ul>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="c1"># 한국어 자료
</span><span class="kn">import</span> <span class="nn">sys</span>
<span class="kn">import</span> <span class="nn">os</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="n">np</span>
<span class="kn">import</span> <span class="nn">nltk</span>
<span class="kn">import</span> <span class="nn">konlpy</span>
<span class="kn">import</span> <span class="nn">pandas</span> <span class="k">as</span> <span class="n">pd</span>
<span class="kn">import</span> <span class="nn">re</span>
<span class="kn">import</span> <span class="nn">random</span>
<span class="kn">import</span> <span class="nn">itertools</span>
<span class="kn">import</span> <span class="nn">warnings</span>
<span class="n">warnings</span><span class="p">.</span><span class="n">filterwarnings</span><span class="p">(</span><span class="n">action</span><span class="o">=</span><span class="s">'ignore'</span><span class="p">)</span>
<span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <span class="n">classification_report</span><span class="p">,</span><span class="n">f1_score</span><span class="p">,</span><span class="n">precision_score</span><span class="p">,</span><span class="n">recall_score</span>
<span class="n">random</span><span class="p">.</span><span class="n">seed</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
</code></pre></div></div>
<h1 id="1-data-load">1. Data Load</h1>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">train_data</span><span class="o">=</span><span class="n">pd</span><span class="p">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s">'final_train_data.csv'</span><span class="p">)</span>
<span class="n">test_data</span><span class="o">=</span><span class="n">pd</span><span class="p">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s">'final_test_data.csv'</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="c1"># 384 duplicates in content, 240 in title
</span><span class="k">print</span><span class="p">(</span><span class="n">train_data</span><span class="p">.</span><span class="n">describe</span><span class="p">())</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code> category content title
count 10686 10686 10686
unique 7 10302 10124
top 정치개혁 개인회생 36개월 단축소급 전국 적용을 위해 춘천지방법원의 법원에 바란다에 글을 올... 경남제약
freq 3094 16 21
</code></pre></div></div>
<h1 id="2-preprocessing">2. Preprocessing</h1>
<h2 id="21-remove-duplicates">2.1. Remove duplicates</h2>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">train_data</span> <span class="o">=</span> <span class="n">train_data</span><span class="p">.</span><span class="n">drop_duplicates</span><span class="p">([</span><span class="s">'content'</span><span class="p">],</span><span class="n">keep</span><span class="o">=</span><span class="s">'first'</span><span class="p">)</span>
<span class="n">train_data</span><span class="p">.</span><span class="n">duplicated</span><span class="p">().</span><span class="nb">sum</span><span class="p">()</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>0
all duplicates removed
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">train_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]</span><span class="o">=</span><span class="n">train_data</span><span class="p">.</span><span class="n">iloc</span><span class="p">[:,</span><span class="mi">1</span><span class="p">]</span><span class="o">+</span><span class="n">train_data</span><span class="p">.</span><span class="n">iloc</span><span class="p">[:,</span><span class="mi">2</span><span class="p">]</span>
<span class="n">train_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]</span>
<span class="n">test_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]</span><span class="o">=</span><span class="n">test_data</span><span class="p">.</span><span class="n">iloc</span><span class="p">[:,</span><span class="mi">1</span><span class="p">]</span><span class="o">+</span><span class="n">test_data</span><span class="p">.</span><span class="n">iloc</span><span class="p">[:,</span><span class="mi">2</span><span class="p">]</span>
</code></pre></div></div>
<h2 id="22-regexp-on-korean-language">2.2. Regexp on Korean Language</h2>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">train_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]</span> <span class="o">=</span> <span class="n">train_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">].</span><span class="nb">str</span><span class="p">.</span><span class="n">replace</span><span class="p">(</span><span class="s">"[^ㄱ-ㅎㅏ-ㅣ가-힣 ]"</span><span class="p">,</span><span class="s">""</span><span class="p">)</span>
<span class="n">test_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]</span> <span class="o">=</span> <span class="n">test_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">].</span><span class="nb">str</span><span class="p">.</span><span class="n">replace</span><span class="p">(</span><span class="s">"[^ㄱ-ㅎㅏ-ㅣ가-힣 ]"</span><span class="p">,</span><span class="s">""</span><span class="p">)</span>
</code></pre></div></div>
<h1 id="3-tokenizing-with-konlpy-okt">3. Tokenizing with konlpy-Okt</h1>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="c1"># tokenize
</span><span class="kn">from</span> <span class="nn">konlpy.tag</span> <span class="kn">import</span> <span class="n">Okt</span>
<span class="n">okt</span> <span class="o">=</span> <span class="n">Okt</span><span class="p">()</span> <span class="c1">#형태소 분석기
</span><span class="n">tokenized_data</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">stopwords</span> <span class="o">=</span> <span class="p">[</span><span class="s">'의'</span><span class="p">,</span><span class="s">'가'</span><span class="p">,</span><span class="s">'이'</span><span class="p">,</span><span class="s">'은'</span><span class="p">,</span><span class="s">'들'</span><span class="p">,</span><span class="s">'는'</span><span class="p">,</span><span class="s">'좀'</span><span class="p">,</span><span class="s">'잘'</span><span class="p">,</span><span class="s">'걍'</span><span class="p">,</span><span class="s">'과'</span><span class="p">,</span><span class="s">'도'</span><span class="p">,</span><span class="s">'를'</span><span class="p">,</span><span class="s">'으로'</span><span class="p">,</span><span class="s">'자'</span><span class="p">,</span><span class="s">'에'</span><span class="p">,</span><span class="s">'와'</span><span class="p">,</span><span class="s">'한'</span><span class="p">,</span><span class="s">'하다'</span><span class="p">]</span>
<span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">train_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]:</span>
<span class="n">temp_X</span> <span class="o">=</span> <span class="n">okt</span><span class="p">.</span><span class="n">morphs</span><span class="p">(</span><span class="n">sentence</span><span class="p">,</span> <span class="n">norm</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">stem</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span> <span class="c1"># 형태소 추출
</span> <span class="n">temp_X</span> <span class="o">=</span> <span class="p">[</span><span class="n">word</span> <span class="k">for</span> <span class="n">word</span> <span class="ow">in</span> <span class="n">temp_X</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">word</span> <span class="ow">in</span> <span class="n">stopwords</span><span class="p">]</span> <span class="c1"># 불용어 제거
</span> <span class="n">tokenized_data</span><span class="p">.</span><span class="n">append</span><span class="p">(</span><span class="n">temp_X</span><span class="p">)</span>
<span class="n">x_train</span><span class="o">=</span><span class="n">tokenized_data</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">okt</span> <span class="o">=</span> <span class="n">Okt</span><span class="p">()</span> <span class="c1">#형태소 분석기
</span><span class="n">tokenized_data</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">stopwords</span> <span class="o">=</span> <span class="p">[</span><span class="s">'의'</span><span class="p">,</span><span class="s">'가'</span><span class="p">,</span><span class="s">'이'</span><span class="p">,</span><span class="s">'은'</span><span class="p">,</span><span class="s">'들'</span><span class="p">,</span><span class="s">'는'</span><span class="p">,</span><span class="s">'좀'</span><span class="p">,</span><span class="s">'잘'</span><span class="p">,</span><span class="s">'걍'</span><span class="p">,</span><span class="s">'과'</span><span class="p">,</span><span class="s">'도'</span><span class="p">,</span><span class="s">'를'</span><span class="p">,</span><span class="s">'으로'</span><span class="p">,</span><span class="s">'자'</span><span class="p">,</span><span class="s">'에'</span><span class="p">,</span><span class="s">'와'</span><span class="p">,</span><span class="s">'한'</span><span class="p">,</span><span class="s">'하다'</span><span class="p">]</span>
<span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">test_data</span><span class="p">[</span><span class="s">'document'</span><span class="p">]:</span>
<span class="n">temp_X</span> <span class="o">=</span> <span class="n">okt</span><span class="p">.</span><span class="n">morphs</span><span class="p">(</span><span class="n">sentence</span><span class="p">,</span> <span class="n">norm</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">stem</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span> <span class="c1"># 형태소 추출 - 토큰화 #norm=True : 근사어
</span> <span class="n">temp_X</span> <span class="o">=</span> <span class="p">[</span><span class="n">word</span> <span class="k">for</span> <span class="n">word</span> <span class="ow">in</span> <span class="n">temp_X</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">word</span> <span class="ow">in</span> <span class="n">stopwords</span><span class="p">]</span> <span class="c1"># 불용어 제거 #https://www.ranks.nl/stopwords/korean
</span> <span class="n">tokenized_data</span><span class="p">.</span><span class="n">append</span><span class="p">(</span><span class="n">temp_X</span><span class="p">)</span>
<span class="n">x_test</span><span class="o">=</span><span class="n">tokenized_data</span>
</code></pre></div></div>
<ul>
<li>save and re-import processed data</li>
</ul>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">import</span> <span class="nn">pickle</span>
<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s">'nlp_final_x_tr.data'</span><span class="p">,</span> <span class="s">'wb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
<span class="n">pickle</span><span class="p">.</span><span class="n">dump</span><span class="p">(</span><span class="n">x_train</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span>
<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s">'nlp_final_x_te.data'</span><span class="p">,</span> <span class="s">'wb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
<span class="n">pickle</span><span class="p">.</span><span class="n">dump</span><span class="p">(</span><span class="n">x_test</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">import</span> <span class="nn">pickle</span>
<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s">'nlp_final_x_tr.data'</span><span class="p">,</span> <span class="s">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
<span class="n">x_train</span> <span class="o">=</span> <span class="n">pickle</span><span class="p">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s">'nlp_final_x_te.data'</span><span class="p">,</span> <span class="s">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
<span class="n">x_test</span> <span class="o">=</span> <span class="n">pickle</span><span class="p">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">from</span> <span class="nn">tensorflow.keras.preprocessing.text</span> <span class="kn">import</span> <span class="n">Tokenizer</span>
<span class="n">tokenizer</span> <span class="o">=</span> <span class="n">Tokenizer</span><span class="p">()</span>
<span class="n">tokenizer</span><span class="p">.</span><span class="n">fit_on_texts</span><span class="p">(</span><span class="n">x_train</span><span class="p">)</span>
<span class="n">threshold</span> <span class="o">=</span> <span class="mi">3</span>
<span class="n">total_cnt</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">.</span><span class="n">word_index</span><span class="p">)</span> <span class="c1"># 단어의 수
</span><span class="n">rare_cnt</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># 등장 빈도수가 threshold보다 작은 단어의 개수를 카운트
</span><span class="n">total_freq</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># 훈련 데이터의 전체 단어 빈도수 총 합
</span><span class="n">rare_freq</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># 등장 빈도수가 threshold보다 작은 단어의 등장 빈도수의 총 합
</span>
<span class="c1"># 단어와 빈도수의 쌍(pair)을 key와 value로 받는다.
</span><span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">tokenizer</span><span class="p">.</span><span class="n">word_counts</span><span class="p">.</span><span class="n">items</span><span class="p">():</span>
<span class="n">total_freq</span> <span class="o">=</span> <span class="n">total_freq</span> <span class="o">+</span> <span class="n">value</span>
<span class="c1"># 단어의 등장 빈도수가 threshold보다 작으면
</span> <span class="k">if</span><span class="p">(</span><span class="n">value</span> <span class="o"><</span> <span class="n">threshold</span><span class="p">):</span>
<span class="n">rare_cnt</span> <span class="o">=</span> <span class="n">rare_cnt</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">rare_freq</span> <span class="o">=</span> <span class="n">rare_freq</span> <span class="o">+</span> <span class="n">value</span>
<span class="k">print</span><span class="p">(</span><span class="s">'단어 집합(vocabulary)의 크기 :'</span><span class="p">,</span><span class="n">total_cnt</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="s">'등장 빈도가 %s번 이하인 희귀 단어의 수: %s'</span><span class="o">%</span><span class="p">(</span><span class="n">threshold</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">rare_cnt</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="s">"단어 집합에서 희귀 단어의 비율:"</span><span class="p">,</span> <span class="p">(</span><span class="n">rare_cnt</span> <span class="o">/</span> <span class="n">total_cnt</span><span class="p">)</span><span class="o">*</span><span class="mi">100</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="s">"전체 등장 빈도에서 희귀 단어 등장 빈도 비율:"</span><span class="p">,</span> <span class="p">(</span><span class="n">rare_freq</span> <span class="o">/</span> <span class="n">total_freq</span><span class="p">)</span><span class="o">*</span><span class="mi">100</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>단어 집합(vocabulary)의 크기 : 34537
등장 빈도가 2번 이하인 희귀 단어의 수: 15483
단어 집합에서 희귀 단어의 비율: 44.830182123519705
전체 등장 빈도에서 희귀 단어 등장 빈도 비율: 1.1132793250941202
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">vocab_size</span> <span class="o">=</span> <span class="n">total_cnt</span> <span class="o">-</span> <span class="n">rare_cnt</span> <span class="o">+</span> <span class="mi">1</span> <span class="c1"># 전체 단어 개수 중 빈도수 2이하인 단어 개수는 제거. 0번 패딩 토큰을 고려하여 +1
</span><span class="k">print</span><span class="p">(</span><span class="s">'단어 집합의 크기 :'</span><span class="p">,</span><span class="n">vocab_size</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>단어 집합의 크기 : 19055
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">Tokenizer</span><span class="p">(</span><span class="n">vocab_size</span><span class="p">)</span>
<span class="n">tokenizer</span><span class="p">.</span><span class="n">fit_on_texts</span><span class="p">(</span><span class="n">x_train</span><span class="p">)</span>
<span class="n">X_train</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">.</span><span class="n">texts_to_sequences</span><span class="p">(</span><span class="n">x_train</span><span class="p">)</span>
<span class="n">X_test</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">.</span><span class="n">texts_to_sequences</span><span class="p">(</span><span class="n">x_test</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">y_train</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">train_data</span><span class="p">.</span><span class="n">category</span><span class="p">)</span>
<span class="n">y_test</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">test_data</span><span class="p">.</span><span class="n">category</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">drop_train</span> <span class="o">=</span> <span class="p">[</span><span class="n">index</span> <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sentence</span><span class="p">)</span> <span class="o"><</span> <span class="mi">1</span><span class="p">]</span>
<span class="n">drop_test</span> <span class="o">=</span> <span class="p">[</span><span class="n">index</span> <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sentence</span><span class="p">)</span> <span class="o"><</span> <span class="mi">1</span><span class="p">]</span>
<span class="n">X_train</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">drop_train</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">y_train</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">y_train</span><span class="p">,</span> <span class="n">drop_train</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">X_train</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_train</span><span class="p">))</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>10301
10301
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">X_test</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">X_test</span><span class="p">,</span> <span class="n">drop_test</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">y_test</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">drop_test</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">X_train</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">X_test</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_train</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_test</span><span class="p">))</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>10301
1158
10301
1158
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="n">plt</span>
<span class="k">print</span><span class="p">(</span><span class="s">'리뷰의 최대 길이 :'</span><span class="p">,</span><span class="nb">max</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">l</span><span class="p">)</span> <span class="k">for</span> <span class="n">l</span> <span class="ow">in</span> <span class="n">X_train</span><span class="p">))</span>
<span class="k">print</span><span class="p">(</span><span class="s">'리뷰의 평균 길이 :'</span><span class="p">,</span><span class="nb">sum</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="nb">len</span><span class="p">,</span> <span class="n">X_train</span><span class="p">))</span><span class="o">/</span><span class="nb">len</span><span class="p">(</span><span class="n">X_train</span><span class="p">))</span>
<span class="n">plt</span><span class="p">.</span><span class="n">hist</span><span class="p">([</span><span class="nb">len</span><span class="p">(</span><span class="n">s</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">X_train</span><span class="p">],</span> <span class="n">bins</span><span class="o">=</span><span class="mi">50</span><span class="p">)</span>
<span class="n">plt</span><span class="p">.</span><span class="n">xlabel</span><span class="p">(</span><span class="s">'length of samples'</span><span class="p">)</span>
<span class="n">plt</span><span class="p">.</span><span class="n">ylabel</span><span class="p">(</span><span class="s">'number of samples'</span><span class="p">)</span>
<span class="n">plt</span><span class="p">.</span><span class="n">show</span><span class="p">()</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>리뷰의 최대 길이 : 9032
리뷰의 평균 길이 : 170.45791670711583
</code></pre></div></div>
<p><img src="/assets/images/output_17_1.png" alt="png" /></p>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="k">def</span> <span class="nf">below_threshold_len</span><span class="p">(</span><span class="n">max_len</span><span class="p">,</span> <span class="n">nested_list</span><span class="p">):</span>
<span class="n">cnt</span> <span class="o">=</span> <span class="mi">0</span>
<span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">nested_list</span><span class="p">:</span>
<span class="k">if</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">s</span><span class="p">)</span> <span class="o"><=</span> <span class="n">max_len</span><span class="p">):</span>
<span class="n">cnt</span> <span class="o">=</span> <span class="n">cnt</span> <span class="o">+</span> <span class="mi">1</span>
<span class="k">print</span><span class="p">(</span><span class="s">'전체 샘플 중 길이가 %s 이하인 샘플의 비율: %s'</span><span class="o">%</span><span class="p">(</span><span class="n">max_len</span><span class="p">,</span> <span class="p">(</span><span class="n">cnt</span> <span class="o">/</span> <span class="nb">len</span><span class="p">(</span><span class="n">nested_list</span><span class="p">))</span><span class="o">*</span><span class="mi">100</span><span class="p">))</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">max_len</span> <span class="o">=</span> <span class="mi">600</span>
<span class="n">below_threshold_len</span><span class="p">(</span><span class="n">max_len</span><span class="p">,</span> <span class="n">X_train</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>전체 샘플 중 길이가 600 이하인 샘플의 비율: 96.02951169789341
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">from</span> <span class="nn">tensorflow.keras.preprocessing.sequence</span> <span class="kn">import</span> <span class="n">pad_sequences</span>
<span class="n">X_train</span> <span class="o">=</span> <span class="n">pad_sequences</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">maxlen</span> <span class="o">=</span> <span class="n">max_len</span><span class="p">)</span>
<span class="n">X_test</span> <span class="o">=</span> <span class="n">pad_sequences</span><span class="p">(</span><span class="n">X_test</span><span class="p">,</span> <span class="n">maxlen</span> <span class="o">=</span> <span class="n">max_len</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">y_train</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">train_data</span><span class="p">.</span><span class="n">category</span><span class="p">)</span>
<span class="n">y_test</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">test_data</span><span class="p">.</span><span class="n">category</span><span class="p">)</span>
<span class="n">y_train</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">y_train</span><span class="p">,</span> <span class="n">drop_train</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">y_test</span> <span class="o">=</span> <span class="n">np</span><span class="p">.</span><span class="n">delete</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">drop_test</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_train</span><span class="p">),</span><span class="nb">len</span><span class="p">(</span><span class="n">y_test</span><span class="p">))</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>10301 1158
</code></pre></div></div>
<h1 id="4-train-test-data">4. Train-test Data</h1>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">t</span><span class="o">=</span><span class="p">{</span><span class="s">'경제민주화'</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span> <span class="s">'교통/건축/국토'</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span> <span class="s">'보건복지'</span><span class="p">:</span> <span class="mi">3</span><span class="p">,</span> <span class="s">'육아/교육'</span><span class="p">:</span> <span class="mi">4</span><span class="p">,</span> <span class="s">'인권/성평등'</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span> <span class="s">'일자리'</span><span class="p">:</span> <span class="mi">6</span><span class="p">,</span> <span class="s">'정치개혁'</span><span class="p">:</span> <span class="mi">7</span><span class="p">}</span>
<span class="k">print</span><span class="p">(</span><span class="n">t</span><span class="p">[</span><span class="s">'보건복지'</span><span class="p">])</span>
<span class="n">index1</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">zeros</span><span class="p">([</span><span class="mi">10301</span><span class="p">,</span><span class="mi">7</span><span class="p">])</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_train</span><span class="p">)):</span>
<span class="n">index1</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="n">t</span><span class="p">[</span><span class="n">y_train</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">=</span><span class="mi">1</span>
<span class="n">y_train</span><span class="o">=</span><span class="n">index1</span>
<span class="n">index1</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">zeros</span><span class="p">([</span><span class="mi">1158</span><span class="p">,</span><span class="mi">7</span><span class="p">])</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">y_test</span><span class="p">)):</span>
<span class="n">index1</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="n">t</span><span class="p">[</span><span class="n">y_test</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">=</span><span class="mi">1</span>
<span class="n">y_test</span><span class="o">=</span><span class="n">index1</span>
<span class="k">print</span><span class="p">(</span><span class="n">y_train</span><span class="p">.</span><span class="n">shape</span><span class="p">,</span><span class="n">y_test</span><span class="p">.</span><span class="n">shape</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>3
(10301, 7) (1158, 7)
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">X_train</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="n">X_test</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="n">y_train</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">y_train</span><span class="p">)</span>
<span class="n">y_test</span><span class="o">=</span><span class="n">np</span><span class="p">.</span><span class="n">array</span><span class="p">(</span><span class="n">y_test</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">X_train</span><span class="p">.</span><span class="n">shape</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">X_test</span><span class="p">.</span><span class="n">shape</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">y_train</span><span class="p">.</span><span class="n">shape</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">y_test</span><span class="p">.</span><span class="n">shape</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">vocab_size</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>(10301, 600)
(1158, 600)
(10301, 7)
(1158, 7)
19055
</code></pre></div></div>
<h1 id="5-lstm-model">5. LSTM Model</h1>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">from</span> <span class="nn">tensorflow.keras.layers</span> <span class="kn">import</span> <span class="n">Embedding</span><span class="p">,</span> <span class="n">Dense</span><span class="p">,</span> <span class="n">LSTM</span>
<span class="kn">from</span> <span class="nn">tensorflow.keras.models</span> <span class="kn">import</span> <span class="n">Sequential</span>
<span class="kn">from</span> <span class="nn">tensorflow.keras.models</span> <span class="kn">import</span> <span class="n">load_model</span>
<span class="kn">from</span> <span class="nn">tensorflow.keras.callbacks</span> <span class="kn">import</span> <span class="n">EarlyStopping</span><span class="p">,</span> <span class="n">ModelCheckpoint</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">import</span> <span class="nn">tensorflow_addons</span> <span class="k">as</span> <span class="n">tfa</span>
<span class="n">f1</span> <span class="o">=</span> <span class="n">tfa</span><span class="p">.</span><span class="n">metrics</span><span class="p">.</span><span class="n">F1Score</span><span class="p">(</span><span class="n">num_classes</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span><span class="n">threshold</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">from</span> <span class="nn">tensorflow</span> <span class="kn">import</span> <span class="n">keras</span>
<span class="n">keras</span><span class="p">.</span><span class="n">__version__</span>
<span class="c1"># '2.6.0'
</span></code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">model</span> <span class="o">=</span> <span class="n">Sequential</span><span class="p">()</span>
<span class="n">model</span><span class="p">.</span><span class="n">add</span><span class="p">(</span><span class="n">Embedding</span><span class="p">(</span><span class="n">vocab_size</span><span class="p">,</span> <span class="mi">128</span><span class="p">))</span>
<span class="n">model</span><span class="p">.</span><span class="n">add</span><span class="p">(</span><span class="n">LSTM</span><span class="p">(</span><span class="mi">128</span><span class="p">))</span>
<span class="n">model</span><span class="p">.</span><span class="n">add</span><span class="p">(</span><span class="n">Dense</span><span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="n">activation</span><span class="o">=</span><span class="s">'sigmoid'</span><span class="p">))</span>
<span class="n">es</span> <span class="o">=</span> <span class="n">EarlyStopping</span><span class="p">(</span><span class="n">monitor</span><span class="o">=</span><span class="s">'val_loss'</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s">'min'</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">patience</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
<span class="n">mc</span> <span class="o">=</span> <span class="n">ModelCheckpoint</span><span class="p">(</span><span class="s">'best_model.h5'</span><span class="p">,</span> <span class="n">monitor</span><span class="o">=</span><span class="n">f1</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s">'max'</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">save_best_only</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
<span class="n">model</span><span class="p">.</span><span class="nb">compile</span><span class="p">(</span><span class="n">optimizer</span><span class="o">=</span><span class="s">'rmsprop'</span><span class="p">,</span> <span class="n">loss</span><span class="o">=</span><span class="s">'categorical_crossentropy'</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="p">[</span><span class="s">'acc'</span><span class="p">,</span><span class="n">f1</span><span class="p">])</span>
<span class="n">model</span><span class="p">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">epochs</span><span class="o">=</span><span class="mi">15</span><span class="p">,</span> <span class="n">callbacks</span><span class="o">=</span><span class="p">[</span><span class="n">es</span><span class="p">,</span> <span class="n">mc</span><span class="p">],</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span> <span class="n">validation_split</span><span class="o">=</span><span class="mf">0.2</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>Epoch 1/15
17/17 [==============================] - 39s 2s/step - loss: 1.8842 - acc: 0.2830 - f1_score: 0.2583 - val_loss: 1.7326 - val_acc: 0.3081 - val_f1_score: 0.2726
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 2/15
17/17 [==============================] - 38s 2s/step - loss: 1.6177 - acc: 0.4615 - f1_score: 0.3981 - val_loss: 1.5555 - val_acc: 0.4639 - val_f1_score: 0.3776
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 3/15
17/17 [==============================] - 38s 2s/step - loss: 1.3590 - acc: 0.5495 - f1_score: 0.4018 - val_loss: 1.3879 - val_acc: 0.5075 - val_f1_score: 0.4232
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 4/15
17/17 [==============================] - 38s 2s/step - loss: 1.1364 - acc: 0.6211 - f1_score: 0.4351 - val_loss: 1.3140 - val_acc: 0.5303 - val_f1_score: 0.4173
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 5/15
17/17 [==============================] - 38s 2s/step - loss: 0.9084 - acc: 0.7212 - f1_score: 0.4893 - val_loss: 1.1852 - val_acc: 0.6487 - val_f1_score: 0.4386
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 6/15
17/17 [==============================] - 38s 2s/step - loss: 0.7336 - acc: 0.7920 - f1_score: 0.5183 - val_loss: 1.1071 - val_acc: 0.6473 - val_f1_score: 0.4536
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 7/15
17/17 [==============================] - 39s 2s/step - loss: 0.5796 - acc: 0.8381 - f1_score: 0.5395 - val_loss: 1.0900 - val_acc: 0.6458 - val_f1_score: 0.4837
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 8/15
17/17 [==============================] - 38s 2s/step - loss: 0.4641 - acc: 0.8757 - f1_score: 0.5575 - val_loss: 0.9854 - val_acc: 0.6982 - val_f1_score: 0.4781
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 9/15
17/17 [==============================] - 38s 2s/step - loss: 0.3588 - acc: 0.9039 - f1_score: 0.5759 - val_loss: 1.2205 - val_acc: 0.6458 - val_f1_score: 0.4993
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 10/15
17/17 [==============================] - 38s 2s/step - loss: 0.3082 - acc: 0.9181 - f1_score: 0.5908 - val_loss: 1.0990 - val_acc: 0.6870 - val_f1_score: 0.5031
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 11/15
17/17 [==============================] - 39s 2s/step - loss: 0.2331 - acc: 0.9408 - f1_score: 0.6076 - val_loss: 1.8883 - val_acc: 0.5793 - val_f1_score: 0.4935
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 12/15
17/17 [==============================] - 39s 2s/step - loss: 0.2625 - acc: 0.9380 - f1_score: 0.6111 - val_loss: 1.3714 - val_acc: 0.6642 - val_f1_score: 0.5215
WARNING:tensorflow:Can save best model only with <tensorflow_addons.metrics.f_scores.F1Score object at 0x7f1bd3f444f0> available, skipping.
Epoch 00012: early stopping
<keras.callbacks.History at 0x7f1bf1402a60>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="c1"># evaluating function: report f1_macro
</span><span class="k">def</span> <span class="nf">evaluate</span><span class="p">(</span><span class="n">test_x</span><span class="p">,</span><span class="n">test_y</span><span class="p">,</span><span class="n">model</span><span class="p">):</span>
<span class="n">predictions</span><span class="o">=</span><span class="n">model</span><span class="p">.</span><span class="n">predict</span><span class="p">(</span><span class="n">test_x</span><span class="p">)</span>
<span class="n">y_pred</span><span class="o">=</span><span class="nb">max</span><span class="p">(</span><span class="n">predictions</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="n">classification_report</span><span class="p">(</span><span class="n">test_y</span><span class="p">,</span><span class="n">y_pred</span><span class="p">))</span>
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="n">model</span><span class="p">.</span><span class="n">evaluate</span><span class="p">(</span><span class="n">X_test</span><span class="p">,</span> <span class="n">y_test</span><span class="p">)</span>
</code></pre></div></div>
<div class="language-plaintext highlighter-rouge"><div class="highlight"><pre class="highlight"><code>37/37 [==============================] - 6s 149ms/step - loss: 1.1846 - acc: 0.6986 - f1_score: 0.5419
[1.1846439838409424,
0.6986182928085327,
array([0.5124555 , 0.59907836, 0.48712873, 0.45633796, 0.5854922 ,
0.4855967 , 0.6674057 ], dtype=float32)]
</code></pre></div></div>
<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="c1"># from saved best model
</span><span class="n">loaded_model</span> <span class="o">=</span> <span class="n">load_model</span><span class="p">(</span><span class="s">'best_model.h5'</span><span class="p">)</span>
<span class="n">loaded_model</span><span class="p">.</span><span class="n">evaluate</span><span class="p">(</span><span class="n">X_test</span><span class="p">,</span> <span class="n">y_test</span><span class="p">)</span>
</code></pre></div></div>
</div>
</section>
<!-- Email subscribe form at the bottom of the page -->
<!--
<section class="subscribe-form">
<h3 class="subscribe-form-title">Subscribe to Darron's Devlog</h3>
<p>Get the latest posts delivered right to your inbox</p>
<span id="searchform" method="post" action="/subscribe/" class="">
<input class="confirm" type="hidden" name="confirm" />
<input class="location" type="hidden" name="location" />
<input class="referrer" type="hidden" name="referrer" />
<div class="form-group">
<input class="subscribe-email" onkeyup="myFunc()"
id="searchtext" type="text" name="searchtext"
placeholder="Search..." />
</div>
<script type="text/javascript">
function myFunc() {
if(event.keyCode == 13) {
var url = encodeURIComponent($("#searchtext").val());
location.href = "/search.html?query=" + url;
}
}
</script>
</span>
</section>
-->
<footer class="post-full-footer">
<!-- Everything inside the #author tags pulls data from the author -->
<!-- #author-->
<!-- /author -->
</footer>
<!-- If you use Disqus comments, just uncomment this block.
The only thing you need to change is "test-apkdzgmqhj" - which
should be replaced with your own Disqus site-id. -->
<section class="post-full-comments">
<div id="disqus_thread"></div>
<script>
var disqus_config = function () {
var this_page_url = 'http://0.0.0.0:4000/NLP_korean_RNN';
var this_page_identifier = '/NLP_korean_RNN';
var this_page_title = 'NLP - Korean Language Text Analysis with RNN';
};
(function() {
var d = document, s = d.createElement('script');
s.src = 'https://.disqus.com/embed.js';
s.setAttribute('data-timestamp', +new Date());
(d.head || d.body).appendChild(s);
})();
</script>
</section>
</article>
</div>
</main>
<!-- Links to Previous/Next posts -->
<aside class="read-next outer">
<div class="inner">
<div class="read-next-feed">
<article class="read-next-card"
style="background-image: url(/assets/built/images/blog-cover1.png)"
>
<header class="read-next-card-header">
<small class="read-next-card-header-sitetitle">— Darron's Devlog —</small>
<h3 class="read-next-card-header-title"><a href="/tag/projects/">Projects</a></h3>
</header>
<div class="read-next-divider"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M13 14.5s2 3 5 3 5.5-2.463 5.5-5.5S21 6.5 18 6.5c-5 0-7 11-12 11C2.962 17.5.5 15.037.5 12S3 6.5 6 6.5s4.5 3.5 4.5 3.5"/></svg>
</div>
<div class="read-next-card-content">
<ul>
<li><a href="/starfish_detection">Starfish detection w/ TF Object Detection API</a></li>
<li><a href="/fashion_GNN">GNN-based Fashion Coordinator</a></li>
<li><a href="/stock_portfolio">Stock Market Portfolio Modeling with R</a></li>
</ul>
</div>
<footer class="read-next-card-footer">
<a href="/tag/projects/">
See all 7 posts →
</a>
</footer>
</article>
<!-- If there's a next post, display it using the same markup included from - partials/post-card.hbs -->
<article class="post-card post-template no-image">
<div class="post-card-content">
<a class="post-card-content-link" href="/R_air_polution">
<header class="post-card-header">
<span class="post-card-tags">Projects</span>
<h2 class="post-card-title">R - Air Pollution Data Analysis</h2>
</header>
<section class="post-card-excerpt">
<p>
</p>
</section>
</a>
<footer class="post-card-meta">
</footer>
</div>
</article>
<!-- If there's a previous post, display it using the same markup included from - partials/post-card.hbs -->
<article class="post-card post-template no-image">
<div class="post-card-content">
<a class="post-card-content-link" href="/islr_ch10">
<header class="post-card-header">
<span class="post-card-tags">Islr</span>
<h2 class="post-card-title">ISLR - Chapter 10. Deep Learning</h2>
</header>
<section class="post-card-excerpt">
<p>Chapter 10. Deep Learning 10.1. Single Layer Neural Networks 10.2. Multilayer Neural Networks 10.3. Convolutional Neural Networks 10.3.1. Convolution Layers 10.3.2. Pooling Layers 10.3.3. Architecture of a Convolutional Neural Network 10.3.4. Data Augmentation</p>
</section>
</a>
<footer class="post-card-meta">
</footer>
</div>
</article>
</div>
</div>
</aside>
<!-- Floating header which appears on-scroll, included from includes/floating-header.hbs -->
<div class="floating-header">
<div class="floating-header-logo">
<a href="/">
<span>Darron's Devlog</span>
</a>
</div>
<span class="floating-header-divider">—</span>
<div class="floating-header-title">NLP - Korean Language Text Analysis with RNN</div>
<div class="floating-header-share">
<div class="floating-header-share-label">Share this <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24">
<path d="M7.5 15.5V4a1.5 1.5 0 1 1 3 0v4.5h2a1 1 0 0 1 1 1h2a1 1 0 0 1 1 1H18a1.5 1.5 0 0 1 1.5 1.5v3.099c0 .929-.13 1.854-.385 2.748L17.5 23.5h-9c-1.5-2-5.417-8.673-5.417-8.673a1.2 1.2 0 0 1 1.76-1.605L7.5 15.5zm6-6v2m-3-3.5v3.5m6-1v2"/>
</svg>
</div>
<a class="floating-header-share-tw" href="https://twitter.com/share?text=NLP+-+Korean+Language+Text+Analysis+with+RNN&url=https://12kdh43.github.io/NLP_korean_RNN"
onclick="window.open(this.href, 'share-twitter', 'width=550,height=235');return false;">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 32 32"><path d="M30.063 7.313c-.813 1.125-1.75 2.125-2.875 2.938v.75c0 1.563-.188 3.125-.688 4.625a15.088 15.088 0 0 1-2.063 4.438c-.875 1.438-2 2.688-3.25 3.813a15.015 15.015 0 0 1-4.625 2.563c-1.813.688-3.75 1-5.75 1-3.25 0-6.188-.875-8.875-2.625.438.063.875.125 1.375.125 2.688 0 5.063-.875 7.188-2.5-1.25 0-2.375-.375-3.375-1.125s-1.688-1.688-2.063-2.875c.438.063.813.125 1.125.125.5 0 1-.063 1.5-.25-1.313-.25-2.438-.938-3.313-1.938a5.673 5.673 0 0 1-1.313-3.688v-.063c.813.438 1.688.688 2.625.688a5.228 5.228 0 0 1-1.875-2c-.5-.875-.688-1.813-.688-2.75 0-1.063.25-2.063.75-2.938 1.438 1.75 3.188 3.188 5.25 4.25s4.313 1.688 6.688 1.813a5.579 5.579 0 0 1 1.5-5.438c1.125-1.125 2.5-1.688 4.125-1.688s3.063.625 4.188 1.813a11.48 11.48 0 0 0 3.688-1.375c-.438 1.375-1.313 2.438-2.563 3.188 1.125-.125 2.188-.438 3.313-.875z"/></svg>
</a>
<a class="floating-header-share-fb" href="https://www.facebook.com/sharer/sharer.php?u=https://12kdh43.github.io/NLP_korean_RNN"
onclick="window.open(this.href, 'share-facebook','width=580,height=296');return false;">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 32 32"><path d="M19 6h5V0h-5c-3.86 0-7 3.14-7 7v3H8v6h4v16h6V16h5l1-6h-6V7c0-.542.458-1 1-1z"/></svg>
</a>
</div>
<progress class="progress" value="0">
<div class="progress-container">
<span class="progress-bar"></span>
</div>
</progress>
</div>
<!-- /post -->
<!-- The #contentFor helper here will send everything inside it up to the matching #block helper found in default.hbs -->
<!-- Previous/next page links - displayed on every page -->
<!-- The footer at the very bottom of the screen -->
<footer class="site-footer outer">
<div class="site-footer-content inner">
<section class="copyright"><a href="/">Darron's Devlog</a> © 2022</section>
<!--
<section class="poweredby">Proudly published with <a href="https://jekyllrb.com/">Jekyll</a> &
<a href="https://pages.github.com/" target="_blank" rel="noopener">GitHub Pages</a> using
<a href="https://github.com/jekyllt/jasper2" target="_blank" rel="noopener">Jasper2</a></section>
-->
<nav class="site-footer-nav">
<a href="/">Latest Posts</a>
</nav>
</div>
</footer>
</div>
<!-- The big email subscribe modal content -->
<div id="subscribe" class="subscribe-overlay">
<a class="subscribe-overlay-close" href="#"></a>
<div class="subscribe-overlay-content">
<h1 class="subscribe-overlay-title">Search Darron's Devlog</h1>
<p class="subscribe-overlay-description">
</p>
<span id="searchform" method="post" action="/subscribe/" class="">
<input class="confirm" type="hidden" name="confirm" />
<input class="location" type="hidden" name="location" />
<input class="referrer" type="hidden" name="referrer" />
<div class="form-group">
<input class="subscribe-email" onkeyup="myFunc()"
id="searchtext" type="text" name="searchtext"
placeholder="Search..." />
</div>
<script type="text/javascript">
function myFunc() {
if(event.keyCode == 13) {
var url = encodeURIComponent($("#searchtext").val());
location.href = "/search.html?query=" + url;
}
}
</script>
</span>
</div>
</div>
<!-- highlight.js -->
<script src="https://cdnjs.cloudflare.com/ajax/libs/prism/1.10.0/components/prism-abap.min.js"></script>
<script>$(document).ready(function() {
$('pre code').each(function(i, block) {
hljs.highlightBlock(block);
});
});</script>
<!-- jQuery + Fitvids, which makes all video embeds responsive -->
<script
src="https://code.jquery.com/jquery-3.2.1.min.js"
integrity="sha256-hwg4gsxgFZhOsEEamdOYGBf13FyQuiTwlAQgxVSNgt4="
crossorigin="anonymous">
</script>
<script type="text/javascript" src="/assets/js/jquery.fitvids.js"></script>
<script type="text/javascript" src="https://demo.ghost.io/assets/js/jquery.fitvids.js?v=724281a32e"></script>
<!-- Paginator increased to "infinit" in _config.yml -->
<!-- if paginator.posts -->
<!-- <script>
var maxPages = parseInt('');
</script>
<script src="/assets/js/infinitescroll.js"></script> -->
<!-- /endif -->
<!-- Add Google Analytics -->
<!-- Google Analytics Tracking code -->
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');
ga('create', '', 'auto');
ga('send', 'pageview');
</script>
<!-- The #block helper will pull in data from the #contentFor other template files. In this case, there's some JavaScript which we only want to use in post.hbs, but it needs to be included down here, after jQuery has already loaded. -->
<script>
// NOTE: Scroll performance is poor in Safari
// - this appears to be due to the events firing much more slowly in Safari.