A collection of Natural Language Processing tools to make your life easier.
Project description
NLP Engine
nlpengine is a collection of Natural Language Processing functions. Its designed to make a NLP engineer's life easier by bundling some of the everyday tasks like data cleaning, feature extraction, text classification, etc. This module is built on top of other popular open source python libraries.
Installation
pip install nlpengine
Features
The key featues in the initial release of nlpengine includes:
- Easy to use text classifiers
- Vectorization of texts
- Active Learning (in development)
- Several utility functions
Text Classifier
A wrapper on top of Facebook's FastText text classifier to build a text classifier with just few lines of codes.
texts = ["sample sentence one", "just another sentence!", "is this a sentence?"]
labels = ["not question", "not question", "question"]
from nlpengine.classifiers import FastTextClassifier
clf = FastTextClassifier()
model = clf.fit(text, labels)
Convert texts to vectors
This module helps convert a corpus of texts to a vector matrix easily. This extracted matrix could be used for further downstream tasks such as text similiarity, vector decomposition & visualization, etc.
from nlpengine.feature_extraction import def get_glove_embeddings_from_sentences
texts = ["a great sentence", "and a meaningful one"]
vectors = get_glove_embeddings_from_sentences(texts, download_model=True)
=======
History
=======
0.0.1 (2021-05-22)
------------------
* First release on PyPI.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for nlpengine-0.0.2-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | c507755ed7ae67f8eda32f2ae3d63a859a833ac6ba5709646be5026b1b3785d7 |
|
MD5 | fd8aba299bb62f79d8af81fb371fb653 |
|
BLAKE2b-256 | acd31beb0225b9fddc066805934b8f45a3f91f34af536b03a822ee67944f2888 |