Repository files navigation

scikit-ntk is implementation of the neural tangent kernel (NTK) for thescikit-learn machine learning library as part of "An Empirical Analysis of the Laplace and Neural Tangent Kernels" master's thesis (found athttp://hdl.handle.net/20.500.12680/d504rr81v andhttps://arxiv.org/abs/2208.03761). This library is meant to directly integrate withsklearn.gaussian_process module. This implementation of the NTK can be used in combination with other kernels to train and predict with Gaussian process regressors and classifiers.

scikit-ntk requires:

• Python (>=3.8)
• scikit-learn (>=1.0.1)

In terminal usingpip run:

pip install scikit-ntk

Usage is described inexamples/usage.py; however, to get started simply import theNeuralTangentKernel class:

fromskntkimportNeuralTangentKernelasNTKkernel_ntk=NTK(D=3,bias=0.01,bias_bounds=(1e-6,1e6))

Once declared, usage is the same as otherscikit-learn kernels.

Python Poetry (>=1.2) is required if you wish to buildscikit-ntk from source. In order to build follow these steps:

1. Clone the repository

git clone git@github.com:392781/scikit-ntk.git

2. Enable a Poetry virtual environment

poetry shell

3. Build and install

poetry buildpoetry install --with dev

If you use scikit-ntk in your scientific work, please use the following citation alongside the scikit-learn citations found athttps://scikit-learn.org/stable/about.html#citing-scikit-learn:

@mastersthesis{lencevicius2022laplacentk,  author  = "Ronaldas Paulius Lencevicius",  title   = "An Empirical Analysis of the Laplace and Neural Tangent Kernels",  school  = "California State Polytechnic University, Pomona",  year    = "2022",  month   = "August",  note    = {\url{http://hdl.handle.net/20.500.12680/d504rr81v}}}