Skip to content

Added more options like min_samples_split,min_samples_leaf,max_features and max_nodes in Decision Tree scikit implementation - #43

Merged
zoq merged 3 commits into
mlpack:masterfrom
Iron-Stark:dectree
Apr 6, 2017
Merged

Added more options like min_samples_split,min_samples_leaf,max_features and max_nodes in Decision Tree scikit implementation #43
zoq merged 3 commits into
mlpack:masterfrom
Iron-Stark:dectree

Conversation

@Iron-Stark

Copy link
Copy Markdown
Contributor

@rcurtin

I have added this PR in place of PR(#39) after updating my fork. Please Review.

@mlpack-jenkins

Copy link
Copy Markdown
Collaborator

Can one of the admins verify this patch?

Comment thread methods/scikit/dtc.py Outdated
random_state=self.seed)
random_state=self.seed,
splitter = self.splitter,
min_samples_split = self.min_samples_split, min_weight_fraction_leaf=self.min_weight_fraction_leaf,

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you use one line for each parameter, to improve the readability?

@Iron-Stark

Copy link
Copy Markdown
Contributor Author

@zoq

Done. Please Review.

@zoq

zoq commented Apr 5, 2017

Copy link
Copy Markdown
Member

@mlpack-jenkins test this

@Iron-Stark

Iron-Stark commented Apr 5, 2017

Copy link
Copy Markdown
Contributor Author

@zoq

Not getting why the benchmark_ann file fails to compile here. Last time it was checked it compiled just fine.

@Iron-Stark

Iron-Stark commented Apr 5, 2017

Copy link
Copy Markdown
Contributor Author

It compiles without any warnings on my local system while working on the same branch.

@zoq

zoq commented Apr 5, 2017

Copy link
Copy Markdown
Member

@Iron-Stark Turns out the test sometimes fails when used with a small dataset, I'm not why. Let's see if testing on a bigger dataset helps.

@zoq

zoq commented Apr 5, 2017

Copy link
Copy Markdown
Member

@mlpack-jenkins test this

@zoq
zoq merged commit 09148bf into mlpack:master Apr 6, 2017
@zoq

zoq commented Apr 6, 2017

Copy link
Copy Markdown
Member

Another great contribution, thanks!

@Iron-Stark

Iron-Stark commented Apr 6, 2017

Copy link
Copy Markdown
Contributor Author

@zoq

Thanks. :) I am planning to implement approximate nearest neighbours using the NearPy library for the benchmarks . Will make a PR by tonight.

Iron-Stark pushed a commit to Iron-Stark/benchmarks that referenced this pull request Apr 6, 2017
Added more options like min_samples_split,min_samples_leaf,max_features and max_nodes in Decision Tree scikit implementation.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants