v0.1.6

Release v0.1.6

• fixed token frequency support (add token frequency support #85)
• fixed threshold selection for binary classification (add threshold selection for binary classification #86)

v0.1.5

Release v0.1.5 🥳🎉🍾

• added pymorphy2 lemmatization (#81)
• added token frequency support (#85)
• added threshold selection for binary classification (#86)
• added arbitrary save folder name (#83)

---

pymorphy2 lemmatization (config.yaml)

# preprocessing
# (included in resulting model pipeline, so preserved for inference)
preprocessing:
  lemmatization: pymorphy2

token frequency support

• text_clf.token_frequency.get_token_frequency(path_to_config) -
  get token frequency of train dataset according to the config file parameters

threshold selection for binary classification

• text_clf.pr_roc_curve.get_precision_recall_curve(path_to_model_folder) -
  get precision and recall metrics for precision-recall curve
• text_clf.pr_roc_curve.get_roc_curve(path_to_model_folder) -
  get false positive rate (fpr) and true positive rate (tpr) metrics for roc curve
• text_clf.pr_roc_curve.plot_precision_recall_curve(precision, recall) -
  plot precision-recall curve
• text_clf.pr_roc_curve.plot_roc_curve(fpr, tpr) -
  plot roc curve
• text_clf.pr_roc_curve.plot_precision_recall_f1_curves_for_thresholds(precision, recall, thresholds) -
  plot precision, recall, f1-score curves for probability thresholds

arbitrary save folder name (config.yaml)

experiment_name: model

v0.1.4

• fixed load_20newsgroups.py (#65 #71)
• added Makefile (#71)
• added logging confusion matrix (#72)
• replaced all "valid" occurrences with "test" (#74)
• updated docstrings (#77)
• changed python interface - train function returns model and target_names_mapping (#78)

v0.1.3

• added hyper-parameters tuning (#58)

v0.1.2

• fixed bug with multiple logging (#55)

v0.1.1

• added logging (#43)
• added unittests (#49)
• added CI with linter, tests, codecov (#46 #49)
• added docker (#48)

v0.1.0

First release.