-
안녕하세요. pretrained model과 관련된 질문이 있어 글 남깁니다.
1. 사전교육 때 pretrained model은 사용하지 못한다고 하셨는데, 이미지 복원(inpainting) 등의 문제에서 loss를 계산할 때 feature map 간의 loss 계산을 위해 pretrained vgg network 등을 사용하는 경우가 있습니다. 대…
-
Hello,
Thank you for making konlpy available to us.
I am having a strange issue when using Komoran to tag strings coming from a file. It works fine when I pass the string directly. I tried the s…
-
Traceback (most recent call last):
File "run_squad.py", line 78, in
(),
File "run_squad.py", line 77, in
(tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, Ro…
-
jdk 설치, 환경변수 설정 다 맞췄고, python 버전, jpype 버전, 비트 수 모두 확인했는데... 오류가 사라지질 않네요.
jdk 16.0.2
python --version 3.9.1
jpype1 1.3.0
실행코드는 아래와 같았습니다.
```
from konlpy.tag import Kkma
kkma=Kkma()
pri…
dseop updated
2 years ago
-
운영체제 : 윈도우 10 64비트
파이썬 : 3.7
konlpy: 0.5.2
터미널 안에서는
>> jpype.getDefaultJVMPath()
'C:\\Program Files\\Java\\jdk-14.0.1\\bin\\server\\jvm.dll'
라고 뜹니다.
쥬피터 안에서는
jpype.getDefaul…
-
hello~ :)
i am wondering if there is a method to assign eoj index in addition to (morph, tag) from Komoran tagger.
``` python
from konlpy.tag import Komoran
komoran = Komoran()
...
…
-
[프로그램]
from konlpy.tag import Okt
okt = Okt()
data = ('노잼노씨 잼없음 신조어등을 나열하고 새로운 문자어를 등록합니다. ')
print('Okt 명사 추출 Results\n', okt.nouns(data), '\n')
[출력결과]
Okt 명사 추출 Results
['노잼', '노씨', '잼'…
-
## 문제 설명
flask 에서 엔드포인트를 만들어서 아래와 같이 객체 생성했습니다.
``` python
ko_test = Komoran('EXP')
```
그 후
``` python
ko_test.get_nouns(searchWord)
```
엔드포인트에 요청을 주면 이부분에서 항상 어떤 문장이나 어떤 단어들을 줘도 [] 빈 리스트만 …
-
먼저 제 사용환경은 아래와 같습니다.
OS : 윈도우 10
JAVA : java 10.0.1 2018-04-17
Java(TM) SE Runtime Environment 18.3 (build 10.0.1+10)
Python : Python 3.7.2
Jpype1 : (0.6.3)
Konlpy : (0.5.1)
…
-
JVMNotSupportedException: JVM mismatch, python is 32 bit and JVM is 64 bit