Changelogs » Nyoka

Nyoka

3.3.0

Minor update
* PMML representaion updated
* AnomalyDetectionModel is used for IsolationForest instead of MiningModel
* stddev as a build-in function
* Removed threshold from OutputField

Bug fix
* Changed operator in SimplePredicate for LightGBM
* Changed default wordSeparatorCharacterRE in TextIndex from `(?u)\b\w\w+\b` to `\s+`
* Corrected invalid constantTerm, TimeValue in ARIMA.
* Added RMSE, Residuals to ARIMA

3.2.1

Bug fix
* Output dataType is updated to have same data type as the classes have.
* batchNormalizationCenter is added to LayerParameters
* Removed unused PMML elements generated in case of LabelBinarizer and OneHotEncoder

3.2.0

Minor Update
* Added pre-processing Transformer `Lag`
Bug Fix
* OneHotEncoder has not attribute `_active_features_`
* Parameters for `ArimaToPmml` and `ExponentialSmoothingToPmml` have made optional. Now users need to pass only `result_objs` to the exporter. No need to pass `time_series_data` and `model_obj`

3.1.0

Change in PMML representation
* Logistic Regression
* Ridge Classifier
* SGD Classifier
* Linear SVC
* Linear Discriminant Analysis
Bug fix
* Linear SVC does not support probability
* KMeans Comparison Measure
* PCA, Binarizer, Gradient Boosting Classifier, Gradient Boosting Regressor change in precision
* CountVectorizer generates unsupported bytes.
* Removed duplicate OutputField

3.0.10

Bug Fix
* Parameters change in KerasToPmml __init__()

3.0.9

Bug Fix
* Bug fix for Keras Exporter

3.0.8

Bug Fix
* Bug fix for PCA. Changed variable `n_components` to `n_components_`

3.0.7

Bug Fix
* Removed dependencies from statsmodels exporter due to which `import nyoka` failed.

3.0.6

Bug Fix
* Bug fix for Extension class

3.0.3

Bug Fix
* Aadded missing required attribute `type_` in Exponential Smoothing Exporter

3.0.2

Bug fix
* Bug fix for Extension tag in nyoka library

3.0.1

Major Update
* Added SARIMAX Exporter
Bug Fix
* Bug fix in ARIMA exporter
* Added missing required attribute in sklearn exporter

2.0.0

Major update
* Added support for Sequential model in Keras Exporter
* PMML43Ext.xsd is updated with new attributes for LayerParameter

1.5.4

Bug fix
* Bug fix for RegressionModel

1.5.3

Bug fix
* Bug fix for LinearSVC and LinearSVR

1.5.2

Minor bug fix
* Bug fix for `OutputField` in PMML specifically for XGBoost and LightGBM

1.5.1

Minor bug fix
* Bug fix to the `OutputField` of PMML.
* Added `modelName` to all the models.

1.5.0

Minor Update
* LinearSVC
* LinearSVR

1.4.2

Minor bug fix
* Bug fix for Anomaly Detection Model

1.4.1

Minor bug fix
* Extra derived field generated by Label Binarizer
* `lxml` was not in the dependency list

1.4.0

New Feature : Anomaly Detection models to scikit-learn exporters
* Isolation Forests
* OneClassSVM