Skip to main content

Build and test a variety of text multi-class classification models.

Project description

lazy-text-classifiers

Build Status Documentation

Build and test a variety of text multi-class classification models.


Installation

Stable Release: pip install lazy-text-classifiers
Development Head: pip install git+https://github.com/evamaxfield/lazy-text-classifiers.git

Quickstart

from lazy_text_classifiers import LazyTextClassifiers
from sklearn.datasets import fetch_20newsgroups
from sklearn.model_selection import train_test_split

# Example data from sklearn
# `x` should be an iterable of strings
# `y` should be an iterable of string labels
data = fetch_20newsgroups(subset="all", remove=("header", "footers", "quotes"))
x = data.data[:1000]
y = data.target[:1000]
y = [data.target_names[id_] for id_ in y]

# Split the data into train and test
x_train, x_test, y_train, y_test = train_test_split(
    x,
    y,
    test_size=0.4,
    random_state=12,
)

# Init and fit all models
ltc = LazyTextClassifiers(random_state=12)
results = ltc.fit(x_train, x_test, y_train, y_test)

# Results is a dataframe
# | model                  |   accuracy |   balanced_accuracy |   precision |   recall |       f1 |    time |
# |:-----------------------|-----------:|--------------------:|------------:|---------:|---------:|--------:|
# | semantic-logit         |    0.73    |            0.725162 |    0.734887 |  0.73    | 0.728247 |  13.742 |
# | tfidf-logit            |    0.70625 |            0.700126 |    0.709781 |  0.70625 | 0.702073 | 187.217 |
# | fine-tuned-transformer |    0.11125 |            0.1118   |    0.10998  |  0.11125 | 0.109288 | 220.105 |

# Get a specific model
semantic_logit = ltc.fit_models["semantic-logit"]
# either an scikit-learn Pipeline or a custom Transformer wrapper class

# All models have a `save` function which will store into the normal format
# * pickle for scikit-learn pipelines
# * torch model directory for Transformers

Documentation

For full package documentation please visit evamaxfield.github.io/lazy-text-classifiers.

Acknowledgements

This package was heavily inspired by lazypredict.

Development

See CONTRIBUTING.md for information related to developing the code.

MIT License

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lazy-text-classifiers-0.1.0.tar.gz (21.4 kB view details)

Uploaded Source

Built Distribution

lazy_text_classifiers-0.1.0-py3-none-any.whl (17.9 kB view details)

Uploaded Python 3

File details

Details for the file lazy-text-classifiers-0.1.0.tar.gz.

File metadata

  • Download URL: lazy-text-classifiers-0.1.0.tar.gz
  • Upload date:
  • Size: 21.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.11.1

File hashes

Hashes for lazy-text-classifiers-0.1.0.tar.gz
Algorithm Hash digest
SHA256 95388c7b7d9b6dba55939ce31d2e74678b3bc5fab990ff78c1c5639bb520a495
MD5 e81e402d223cdbc2939ab0ed9bc6b3ce
BLAKE2b-256 3474ae5e5bc3776e635ff71a1ccafbcf713ebbba6d10ab0f13c92f00dfdae085

See more details on using hashes here.

File details

Details for the file lazy_text_classifiers-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for lazy_text_classifiers-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 afc2327105e2a008bc0552cab4dd3630817d3d19721916364f33c02e4b4a5b5e
MD5 41764e69d49e02a3355f485d72f99329
BLAKE2b-256 cfa4f5483621060970cf0ab50cf2166a4b0f51b098dc716667a0745adaffb8a9

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page