Nlp

NLP Knowledge and experiences

Python

import sklearn
from sklearn.feature_extraction.text import CountVectorizer
# instantiate an ngram counter
counts = CountVectorizer(analyzer='word', ngram_range=(n,n))

# create a dictionary of n-grams by calling `.fit`
vocab2int = counts.fit([a_text, s_text]).vocabulary_

Resources

• CS 224N (Don't remember why i need this: 51506490)

• https://www.fast.ai/

• Intro to TensorFlow for Deep Learning | Udacity

• Embedding projector - visualization of high-dimensional data

• LDA visualized using t-SNE and Bokeh | Kaggle

• (PDF) Classification of Twitter Users Who Tweet About E-Cigarettes

• Visualizing Top Tweeps with t-SNE, in Javascript