stateinterpreter.metastable.approximate_FES#

class stateinterpreter.metastable.approximate_FES(colvar, bandwidth, selected_cvs=None, kBT=2.5, logweights=None)[source]#

Bases:

Approximate Free Energy Surface (FES) in the space of selected_cvs through Gaussian Kernel Density Estimation

Args:: bandwidth (scalar, vector or matrix): selected_cvs (numpy.ndarray or pd.Dataframe): List of sampled collective variables with dimensions [num_timesteps, num_CVs] kBT (scalar): Temperature logweights (arraylike log weights, optional): Logarithm of the weights. Defaults to None (uniform weights).
Returns:: function: Function approximating the free Energy Surface