parsnip.ParsnipModel.decode

ParsnipModel.decode(encoding, ref_times, color, times, redshifts, band_indices, amplitude=None)

Predict the light curves for a given set of latent variables

Parameters:

encoding (FloatTensor) – Coordinates in the ParSNIP intrinsic latent space for each light curve
ref_times (FloatTensor) – Reference time for each light curve
color (FloatTensor) – Color of each light curve
times (FloatTensor) – Times to predict each light curve at
redshifts (FloatTensor) – Redshift of each light curve
band_indices (LongTensor) – Band indices for each observation
amplitude (FloatTensor, optional) – Amplitude to scale each light curve by, by default no scaling will be applied

Returns: