scikit benchmark for decision_tree, unit tests

mlpack / benchmarks

Machine Learning Benchmark Scripts

101 stars 49 forks source link

scikit benchmark for decision_tree, unit tests #25

Closed Iron-Stark closed 7 years ago

Iron-Stark commented 7 years ago

Options include c (criterion), d (max_depth), s (splitter), mss (min_samples_split), msl (min_samples_leaf), mf (max_features), mln (max_leaf_nodes).
Also added a simple RunTiming function and unit tests for constructor, RunTiming.

To use the unit tests, modify tests/tests.py to include benchmark_decision_tree in modules list.

@rcurtin @zoq Please provide feedback, this is my first PR. Thanks!

ThyrixYang commented 7 years ago

HI, @Iron-Stark , there is already a benchmark script of scikit-learn at "dtc.py" (decision tree classifier), and in block DTC, I write dtc benchmark for mlpack in block DTC too. Did you see that?

mlpack-jenkins commented 7 years ago

Can one of the admins verify this patch?