목록tomoto (3)

무회blog

python: 200609-python_part1_topics_추출

In [1]: # for_005_topikTs_tomoto # %pip install tomotopy # %pip install nltk # 한국어 전처리 # %pip install --upgrade kiwipiepy # %pip install KoNLP # import nltk # nltk.download() ############################################# import tomotopy as tp import pandas as pd import numpy as np import nltk.stem, nltk.corpus, nltk.tokenize, re ,os from kiwipiepy import Kiwi kiwi = Kiwi() kiwi.prepare() #######..
Python 2020. 6. 9. 18:21
python: 200608-python, LDA 토픽추출 테스트 001_success, tomoto

point k_cnt = 5 # 토픽의 개수 , 행 , 1 ~ 32767 사이의 정수 top_n_cnt = 7 # 토픽의 갯수 , 열 min_cf_cnt = 10 # 단어 최소 출현 빈도 , 0 일시 모든 단어를 동일하게 봄 alpha_cnt = 0.1 # 문헌‐토픽 빈도 eta_cnt = 0.01 # 토픽‐단어 빈도 tran_cnt = 500 # 자동학습 빈도 model = tp.LDAModel(k=k_cnt, alpha=alpha_cnt,eta = eta_cnt) model_PMI = tp.LDAModel(k=k_cnt, alpha=alpha_cnt,eta = eta_cnt, min_cf=min_cf_cnt,tw=tp.TermWeight.PMI) model_IDF = tp.LDAModel(k=k_cn..
Python 2020. 6. 8. 16:12
python: 200607- python 토픽추출 , nltk, tomoto, test->for_topics-004

for_topics-004 # %pip install tomotopy # %pip install nltk # nltk.download() # %pip install --upgrade kiwipiepy # 한국어 전처리 # %pip install KoNLP import tomotopy as tp import nltk.stem, nltk.corpus, nltk.tokenize, re from kiwipiepy import Kiwi # model = tp.LDAModel(k=20) model = tp.LDAModel(k=20, alpha=0.1, eta=0.01, min_cf=5) # k 토픽 개수 for i, line in enumerate(open('./corpos/newsSogBo_test003_001...
Python 2020. 6. 7. 04:37
Prev 1 Next