NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
Machine learning algorithms, Machine learning models performance evaluation functionality
Last release 2 months ago
26 Jul 2026
Release timing varies
gaps range from 2 weeks to 13 months
Nearly every release is documented
notes for 60 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
8 years old
170 releases · first in 2018
Unit tests: iterable2dAlmostEqualTo and iterableAlmostEqualTo matchers used from ml_tech
ml_linalg 11.0.0 supportedUnit tests: iterable2dAlmostEqualTo and iterableAlmostEqualTo matchers used from ml_tech- Decision tree classifier added
ScoreToProbMapperFactory removed
One column per quarter.
ScoreToProbMapperFactory removedScoreToProbMapperType enum removedScoreToProbMapper: the entity renamed to LinkFunction- Cost function factory removed - Cost function type removed
Breaking change: GradientType enum removed
Predictor: fit method removed, fitting is happening while a model is being createdPredictor: interface replaced with Assessable, redundant properties removedLinearClassifier reorganizedInterceptPreprocessor replaced with a helper function addInterceptIfUnit tests for cross validator added
Added immutable state to all the predictor subclasses
NoNParametricRegressor.nearestNeighbour: added possibility to specify the kernel function
NoNParametricRegressor.nearestNeighbour: added possibility to specify the kernel function- test coverage restored
NoNParametricRegressor class added
NoNParametricRegressor class addedKNNRegressor class addedml_linalg v9.0.0 supported- ml_linalg v7.0.0 support
ml_linalg v7.0.0 supportData preprocessing: all the entities moved to separate repo - ml_preprocessing
Data preprocessing: All categorical values are now converted to String type
Examples for Linear regression and Logistic regression updated (vector's normalize method used)
normalize method used)CategoricalDataEncoderType: one-hot encoding documentation correctedSoftmax regression example added to README
- README corrected
LinearClassifier.logisticRegressor: numerical stability improved
LinearClassifier.logisticRegressor: numerical stability improvedLinearClassifier.logisticRegressor: probabilityThreshold parameter addedDataFrame.fromCsv: parameter fieldDelimiter addedDataFrame: labelName parameter added
DataFrame: labelName parameter addedClassifier: type of weightsByClasses changed from Map to Matrix
ml_linalg v6.0.2 supportedClassifier: type of weightsByClasses changed from Map to MatrixSoftmaxRegressor: more detailed unit tests for softmax regression addedDataFrame introduced (former MLData)LinearClassifier.softmaxRegressor implemented
LinearClassifier.softmaxRegressor implementedMetric interface refactored (getError renamed to getScore)SoftmaxMapper added (aka Softmax activation function)
SoftmaxMapper added (aka Softmax activation function)ConvergenceDetector added (this entity stops the optimizer when it is needed)
ConvergenceDetector added (this entity stops the optimizer when it is needed)All the exports packed into ml_algo entry
ml_algo entryCoefficients in optimizers now are a matrix
LinkFunction renamed to ScoreToProbMapper
LinkFunction renamed to ScoreToProbMapperScoreToProbMapper accepts vector and returns vector instead of a scalarPedantic package integration added
- Coveralls integration added - dartfm check task added
Documentation for linear regression corrected
MLData correctedDocumentation for logistic regression corrected
Tests corrected: removed import test_api.dart
test_api.dartFloat32x4CsvMlData significantly extended
Float32x4CsvMlData significantly extendedReal-life example added (black friday dataset)
rows parameter added to Float32x4CsvMlDataOne hot encoder integrated into CSV ML data
Performance test for one hot encoder added
- One hot encoder implemented
enum for categorical data encoding added
- Cross validator factory added - README updated
- csv-parser added
ml_linalg removed from export file
ml_linalg removed from export filedatasets directory created- ml_linal ^4.0.0 supported
ml_linal ^4.0.0 supportedbuild_runner dependency updated
dartfmt tool applied to all necessary files
dartfmt tool applied to all necessary filesVectorized cost functions applied
- ml_linalg 2.0.0 supported
ml_linalg 2.0.0 supportedMatrix-based gradient calculation added for log likelihood cost function
Matrix-based gradient calculation added for squared cost function
- Description corrected
- dartfm tool applied
dartfm tool appliedGet rid of MLVector's deprecated methods
- Library public release
Your coding agent can read these notes before it upgrades. Set up the MCP server →