All Versions
98
Latest Version
Avg Release Cycle
15 days
Latest Release
-
Changelog History
Page 7
Changelog History
Page 7
-
v0.9.1 Changes
- โ Add class for Extra-Trees classifier.
- โ Add class for Extra-Trees regressor.
- ๐ Refactor extension modules of decision tree estimators for improving performance.
-
v0.9.0 Changes
๐ฅ Breaking changes
- ๐ Decide to introduce Ruby extensions for improving performance.
- ๐ Fix to find split point on decision tree estimators using extension modules.
-
v0.8.4 Changes
- โ Remove unused parameter on Nadam.
- ๐ Fix condition to stop growing tree about decision tree.
-
v0.8.3 Changes
- โ Add optimizer class for AdaGrad.
- โ Add evaluator class for ROC AUC.
- โ Add class for scaling with maximum absolute value.
-
v0.8.2 Changes
- โ Add class for Adam optimizer.
- โ Add data splitter classes for random permutation cross validation.
- โ Add accessor method for number of splits to K-fold splitter classes.
- โ Add execution result of example script on README (#3).
-
v0.8.1 Changes
- โ Add some evaluator classes.
- MeanSquaredLogError
- MedianAbsoluteError
- ExplainedVarianceScore
- AdjustedRandScore
- MutualInformation
- ๐จ Refactor normalized mutual infomation evaluator.
- ๐ Fix typo on document (#2).
- โ Add some evaluator classes.
-
v0.8.0 Changes
๐ฅ Breaking changes
- ๐ Rename SVMKit to Rumale.
- ๐ Rename SGDLienareEstimator class to BaseLienarModel class.
- Add data type option to load_libsvm_file method. By default, the method represents the feature with Numo::DFloat.
๐จ Refactoring
- ๐จ Refactor factorization machine estimators.
- ๐จ Refactor decision tree estimators.
-
v0.7.3 Changes
- โ Add class for grid search performing hyperparameter optimization.
- โ Add argument validations to Pipeline.
-
v0.7.2 Changes
- โ Add class for Pipeline that constructs chain of transformers and estimators.
- ๐ Fix some typos on document (#1).
-
v0.7.1 Changes
- ๐ Fix to use CSV class in parsing libsvm format file.
- ๐จ Refactor ensemble estimators.