Natural Language Tool to Extract Features from Intents Dataset.
Project description
Introduction
This package will help the user to extract ngrams, nouns, verbs, adverbs and more features from an Intent Dataset in Spanish. This will help the user to balance the dataset by comparing number of ocurrences of each feature in each intent.
The entry format is an excel (.xlsx). Each column would be an intent and in the first row it has to be informed the intent name, then in the next rows the sentences for each intent training.
The output format is a json. It will have one dictionary per intent including all the features extracted.
How to use it?
Installation
pip install nluanalyzer
Usage
Usage
nluanalyzer -h|--help
nluanalyzer -v|--version
nluanalyzer <excel_path>
Options
-h --help Show help screen.
-v --version Show version.
Example
$ nlu_analyzer intents.xlsx
'Successfully saved intents analyzed file in /home/project/location/Formatted_Metrics.xlsx'
Package Requirements
nltk==3.4.5
spacy==2.2.4
docopt==0.6.2
numpy==1.17.4
docopt==0.6.2
pandas==0.25.3
textacy==0.10.0
scikit_learn==0.23.1
History
1.0.0 (2020-05-26)
Spanish Analysis Implemented.
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 nluanalizer-1.0.2-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9a979b9f136a31d3ddc107e4743fcbd7cc370e735e6f52a1556ebcd42f5d9afd |
|
MD5 | 05db026741243f819c2e17d331b68419 |
|
BLAKE2b-256 | ff957ab96064f523b332eafe93cdbe07ab22c6a9529d9b04beeec89e5beb1167 |