Notebooks about experimentations.
- Decision Tree and Logistic Regression
- Faster Polynomial Features
- KMeans with norm L1
- LogisticRegression and Clustering
- Piecewise classification with scikit-learn predictors
- Piecewise linear regression with scikit-learn predictors
- Predictable t-SNE
- Quantile MLPRegressor
- Quantile Regression
- Regression with confidence interval
- Traceable n-grams with tf-idf
- Transformed Target
- Visualize a scikit-learn pipeline
Experiments with scikit-learn and cython. The first experiment implements a criterion for a sklearn.tree.DecisionTreeRegressor. This code is based on the API in Criterion which changed in version 0.21.
The notebooks explore trees, mostly trees from scikit-learn, and compute unusual results from the structure.