layg.emulator.torch_emulator.TorchEmulator

class layg.emulator.torch_emulator.TorchEmulator

Class that uses pytorch to do emulation

The Universal Approximation Theorem says that any Lebesgue integrable function can be approximated by a feed-forward network with sufficient layers of sufficient width. It doesn’t guarantee that we can train the network though.

Methods

add_data(self, x_train, y_train)	Add data to the training set on the fly
set_emul_error_func(self, x_cv, y_cv_err)	Fit a quadratic to the residuals and mean distance to nearby points

set_emul_func

__init__(self)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(self)	Initialize self.
add_data(self, x_train, y_train)	Add data to the training set on the fly
set_emul_error_func(self, x_cv, y_cv_err)	Fit a quadratic to the residuals and mean distance to nearby points
set_emul_func(self, x_train, y_train)