Created
December 20, 2016 13:26
-
-
Save clausd/1566cb47a03a0517975d7058a17dcd85 to your computer and use it in GitHub Desktop.
How to do LDA in python (for Morten)
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import pandas as pd | |
import lda | |
import lda.datasets | |
from sklearn.feature_extraction.text import CountVectorizer | |
def load_questions(): | |
sheet = pd.read_excel('android_watch.xlsx') | |
Qs = sheet['Question'] | |
return sheet, list(Qs) | |
def featurize(questions, stop_words = None): | |
cv = CountVectorizer(stop_words = stop_words) | |
X = cv.fit_transform(questions) | |
return X, list(cv.vocabulary_) | |
def find_topics(features): | |
model = lda.LDA(n_topics=20, n_iter=500, random_state=1) | |
model.fit(features) | |
return model | |
def report(): | |
s,q = load_questions() | |
feats, vocab = featurize(q) | |
model = find_topics(feats) | |
# list topics | |
n_top_words = 8 | |
topic_word = model.topic_word_ | |
for i, topic_dist in enumerate(topic_word): | |
topic_words = np.array(vocab)[np.argsort(topic_dist)][:-n_top_words:-1] | |
print('Topic {}: {}'.format(i, ' '.join(topic_words))) | |
# list primary topic for each q | |
doc_topic = model.doc_topic_ | |
for i in range(10): | |
print("{} (top topic: {})".format(q[i], doc_topic[i].argmax())) | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Koden scoret fra https://ariddell.org/lda.html (og ikke helt sikker på om alt er som det skal være)