CRFsuite (python-crfsuite) wrapper which provides interface simlar to scikit-learn
Project description
sklearn-crfsuite is a thin CRFsuite (python-crfsuite) wrapper which provides interface simlar to scikit-learn. sklearn_crfsuite.CRF is a scikit-learn compatible estimator: you can use e.g. scikit-learn model selection utilities (cross-validation, hyperparameter optimization) with it, or save/load CRF models using joblib.
License is MIT.
Documentation can be found here.
Changes
0.5.0 (2024-06-18)
The CRF.predict() and CRF.predict_marginals() methods now return a numpy array, as expected by newer versions of scikit-learn.
Fixed the parameters of a call to the sklearn.metrics.classification_report() function from the flat_classification_report() function.
sequence_accuracy_score now works with numpy arrays.
0.4.0 (2024-06-18)
Dropped official support for Python 3.7 and lower, and added official support for Python 3.8 and higher.
Added support for scikit-learn 0.24.0 and higher.
Increased minimum versions of dependencies as follows:
python-crfsuite: 0.8.3 → 0.9.7
scikit-learn: 0.24.0
tabulate: 0.4.2
Internal changes: enabled GitHub Actions for CI, added a tox environment for minimum supported versions of dependencies, applied automatic code cleanups.
0.3.6 (2017-06-22)
added sklearn_crfsuite.metrics.flat_recall_score.
0.3.5 (2017-03-21)
Properly close file descriptor in FileResource.cleanup;
declare Python 3.6 support, stop testing on Python 3.3.
0.3.4 (2016-11-17)
Small formatting fixes.
0.3.3 (2016-03-15)
scikit-learn dependency is now optional for sklearn_crfsuite; it is required only when you use metrics and scorers;
added metrics.flat_precision_score.
0.3.2 (2015-12-18)
Ignore more errors in FileResource.__del__.
0.3.1 (2015-12-17)
Ignore errors in FileResource.__del__.
0.3 (2015-12-17)
Added sklearn_crfsuite.metrics.sequence_accuracy_score() function and related sklearn_crfsuite.scorers.sequence_accuracy;
FileResource.__del__ method made more robust.
0.2 (2015-12-11)
backwards-incompatible: crf.tagger attribute is renamed to crf.tagger_; when model is not trained accessing this attribute no longer raises an exception, its value is set to None instead.
new CRF attributes available after training:
classes_
size_
num_attributes_
attributes_
state_features_
transition_features_
Tutorial is added.
0.1 (2015-11-27)
Initial release.
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
File details
Details for the file sklearn_crfsuite-0.5.0.tar.gz
.
File metadata
- Download URL: sklearn_crfsuite-0.5.0.tar.gz
- Upload date:
- Size: 26.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.12.4
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 136cb941ff5f7ce62a2c378988d1123ac1037241e73999ae404509a2135078e9 |
|
MD5 | 638dd38da11846451ce488cb7eb59c6a |
|
BLAKE2b-256 | d7d426211eb22127f8476589e92448273c698f4ac39fce2e300cd0a1f748de98 |
File details
Details for the file sklearn_crfsuite-0.5.0-py2.py3-none-any.whl
.
File metadata
- Download URL: sklearn_crfsuite-0.5.0-py2.py3-none-any.whl
- Upload date:
- Size: 10.8 kB
- Tags: Python 2, Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.12.4
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9cb7a881969b6426d28ada69ada7eff4c789a3c508eafc32d39c0d078ce3bebc |
|
MD5 | e738a2f00a1e4699bbedb4b135ce344c |
|
BLAKE2b-256 | b211a8370dd6fce65f8f4e74a0adffae72be9db5799d8ed8ddbf84415356a764 |