mlinsights 0.2.288

Extends scikit-learn with a couple of new models, transformers, metrics, plotting.

mlinsights - extensions to scikit-learn

mlinsights extends scikit-learn with a couple of new models, transformers, metrics, plotting. It provides new trainers such as QuantileLinearRegression which trains a linear regression with L1 norm non-linear correlation based on decision trees, or QuantileMLPRegressor a modification of scikit-learn’s MLPRegressor which trains a multi-layer perceptron with L1 norm. It also explores PredictableTSNE which trains a supervized model to replicate t-SNE results or a PiecewiseRegression which partitions the data before fitting a model on each bucket.

• GitHub/mlinsights

• documentation

• Blog

History

current - 2019-05-23 - 0.00Mb

• 55: Explore caching for gridsearchCV (2019-05-22)

• 56: Fixes #55, explore caching for scikit-learn pipeline (2019-05-22)

0.2.269 - 2019-05-20 - 0.41Mb

• 53: implements a function to extract intermediate model outputs within a pipeline (2019-05-07)

• 51: Implements a tfidfvectorizer which keeps more information about n-grams (2019-04-26)

• 46: implements a way to determine close leaves in a decision tree (2019-04-01)

• 44: implements a model which produces confidence intervals based on bootstrapping (2019-03-29)

• 40: implements a custom criterion for a decision tree optimizing for a linear regression (2019-03-28)

• 39: implements a custom criterion for decision tree (2019-03-26)

• 41: implements a direct call to a lapack function from cython (2019-03-25)

• 38: better implementation of a regression criterion (2019-03-25)

0.1.199 - 2019-03-05 - 0.05Mb

• 37: implements interaction_only for polynomial features (2019-02-26)

• 36: add parameter include_bias to extended features (2019-02-25)

• 34: rename PiecewiseLinearRegression into PiecewiseRegression (2019-02-23)

• 33: implement the piecewise classifier (2019-02-23)

• 31: uses joblib for piecewise linear regression (2019-02-23)

• 30: explore transpose matrix before computing the polynomial features (2019-02-17)

• 29: explore different implementation of polynomialfeatures (2019-02-15)

• 28: implement PiecewiseLinearRegression (2019-02-10)

• 27: implement TransferTransformer (2019-02-04)

• 26: add function to convert a scikit-learn pipeline into a graph (2019-02-01)

• 25: implements kind of trainable t-SNE (2019-01-31)

• 6: use keras and pytorch (2019-01-03)

• 22: modifies plot gallery to impose coordinates (2018-11-10)

• 20: implements a QuantileMLPRegressor (quantile regression with MLP) (2018-10-22)

• 19: fix issues introduced with changes in keras 2.2.4 (2018-10-06)

• 18: remove warning from scikit-learn about cloning (2018-09-16)

• 16: move CI to python 3.7 (2018-08-21)

• 17: replace as_matrix by values (pandas deprecated warning) (2018-07-29)

• 14: add transform to convert a learner into a transform (sometimes called a featurizer) (2018-06-19)

• 13: add transform to do model stacking (2018-06-19)

• 8: move items from papierstat (2018-06-19)

• 12: fix bug in quantile regression: wrong weight for linear regression (2018-06-16)

• 11: specifying quantile (2018-06-16)

• 4: add function to compute non linear correlations (2018-06-16)

• 10: implements combination between logistic regression and k-means (2018-05-27)

• 9: move items from ensae_teaching_cs (2018-05-08)

• 7: add quantile regression (2018-05-07)

• 5: replace flake8 by code style (2018-04-14)

• 1: change background for cells in notebooks converted into rst then in html, highlight-ipython3 (2018-01-05)

• 2: save features and metadatas for the search engine and retrieves them (2017-12-03)