find_reps

Methods for the find_reps module in hetgpy

docstring for find_reps

find_reps(X, Z, return_Zlist=True, rescale=False, normalize=False, inputBounds=None, use_torch=False)[source]

Prepare data for use with mleHomGP and mleHetGP

In particular to find replicated observations

Parameters:

  • X (array_like) – matrix of design locations, one point per row

  • Z (array_like) – vector of observations at X

  • return_Zlist (bool) – to return Zlist, see below

  • rescale (bool) – if True, the inputs are rescaled to the unit hypercube

  • normalize (bool) – if True, the outputs are centered and normalized

  • inputBounds (array_like) – optional matrix of known boundaries in original input space, of size 2 times X.shape[1]. If not provided, and rescale == True, it is estimated from the data.

Returns:

  • dict – dictionary of outputs

  • type – explain types

  • out –

    dictionary of

    • X0 matrix with unique designs locations, one point per row

    • Z0 vector of averaged observations at X0

    • mult number of replicates at X0

    • Z vector with all observations, sorted according to X0

    • Zlist optional list, each element corresponds to observations at a design in X0

    • inputBounds optional matrix, to rescale back to the original input space

    • outputStats optional vector, with mean and variance of the original outputs.

References

Binois et. al (2018)

Examples

>>> from hetgpy.data import mcycle
>>> from hetgpy.hetGP import hetGP
>>> X = mcycle['times']
>>> Z = mcycle['accel']

>>> model = hetGP()
>>> out = model.find_reps(X, Z)
>>> print(out)