jotterbach / dstk Goto Github PK
View Code? Open in Web Editor NEWData Science Toolkit
License: MIT License
Data Science Toolkit
License: MIT License
Introduce BinnerBase
and a proper abstraction for bins and bin collections. The latter looks very similar to a ShapeFunction.
If the same binner is used to fit two different sets of values
binner = ConditionalInferenceBinner('test')
binner.fit(values_v1, target)
binner.fit(values_V2, target)
The second fit is not clean, and plotting the resulting binner results in extra bins
SGDBolasso is sensitive to the choice of the penalization parameter alpha
. We need to implement cross-validation for automating the choice of a good alpha.
We can use a generalization of MDLP or the ConditionalInferenceBinner to limit the sensible depth of a RegressionTree and hence can avoid overfitting of a GAM
The sklearn.DecisionTreeRegressor
contains arrays tree_.n_node_samples
and tree_.weighted_n_node_samples
that can be used to re-weight the leaf value to be the average rather than the weighted average. The latter leads to large spikes in the generated shape if only few samples are in the final leaf.
Instead of implementing the bagging in the training function we can just leverage the sklearn.ensemble.BaggingRegressor
to do that calculation
The sklearn.DecisionTreeRegressor
contains arrays tree_.n_node_samples
and tree_.weighted_n_node_samples
that can be used to reweight the leaf value to be the average rather than the weighted average. The latter leads to large spikes in the generated shape if only few samples are in the final leaf.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.