Estimate Vapnik-Chervonenkis Dimension and Sample Complexity
Utility function to generate accuracy metrics, for use with `estimate_...
Estimate sample complexity bounds for a binary classification algorith...
Simulate data with appropriate structure to be used in estimating samp...
Recalculate achieved sample complexity bounds given different paramete...
Utility function to define the least-squares loss function to be optim...
Represent simulated sample complexity bounds graphically
Utility function to generate data points for estimation of the VC Dime...
Calculate sample complexity bounds for a classifier given target accur...
Estimate the Vapnik-Chervonenkis (VC) dimension of an arbitrary binary...
We provide a suite of tools for estimating the sample complexity of a chosen model through theoretical bounds and simulation. The package incorporates methods for estimating the Vapnik-Chervonenkis dimension (VCD) of a chosen algorithm, which can be used to estimate its sample complexity. Alternatively, we provide simulation methods to estimate sample complexity directly. For more details, see Carter, P & Choi, D (2024). "Learning from Noise: Applying Sample Complexity for Political Science Research" <doi:10.31219/osf.io/evrcj>.