deepmd.utils.tabulate

Contents

deepmd.utils.tabulate#

Attributes#

Classes#

A base class for pt and tf tabulation.

Module Contents#

deepmd.utils.tabulate.log[source]#

class deepmd.utils.tabulate.BaseTabulate(descrpt, neuron, type_one_side, exclude_types, is_pt)[source]#

Bases: abc.ABC

A base class for pt and tf tabulation.

descrpt[source]#

neuron[source]#

type_one_side[source]#

exclude_types[source]#

is_pt[source]#: Need to be initialized in the subclass.

descrpt_type = 'Base'[source]#

sel_a = [][source]#

rcut = 0.0[source]#

rcut_smth = 0.0[source]#

davg[source]#

dstd[source]#

ntypes = 0[source]#

layer_size = 0[source]#

table_size = 0[source]#

bias[source]#

matrix[source]#

data_type = None[source]#

last_layer_size = 0[source]#: Save the tabulation result.

data[source]#

upper[source]#

lower[source]#

build(min_nbor_dist: float, extrapolate: float, stride0: float, stride1: float) → tuple[dict[str, int], dict[str, int]][source]#

Build the tables for model compression.

Parameters:

min_nbor_dist: The nearest distance between neighbor atoms
extrapolate: The scale of model extrapolation
stride0: The uniform stride of the first table
stride1: The uniform stride of the second table

Returns:

lowerdict[str, int]: The lower boundary of environment matrix by net
upperdict[str, int]: The upper boundary of environment matrix by net

_build_lower(net, xx, idx, upper, lower, stride0, stride1, extrapolate, nspline) → None[source]#

abstract _make_data(xx, idx) → tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray][source]#

Generate tabulation data for the given input.

Parameters:

xxnp.ndarray: Input values to tabulate
idxint: Index for accessing the correct network parameters

Returns:

tuple[np.ndarray, np.ndarray, np.ndarray]: Values, first derivatives, and second derivatives

_all_excluded(ii: int) → bool[source]#

Check if type ii excluds all types.

Parameters:

iiint: type index

Returns:

bool: if type ii excluds all types

abstract _get_descrpt_type()[source]#: Get the descrpt type.

abstract _get_layer_size()[source]#: Get the number of embedding layer.

_get_table_size()[source]#

_get_data_type()[source]#

_get_last_layer_size()[source]#

abstract _get_bias()[source]#: Get bias of embedding net.

abstract _get_matrix()[source]#: Get weight matrx of embedding net.

abstract _convert_numpy_to_tensor()[source]#: Convert self.data from np.ndarray to torch.Tensor.

_convert_numpy_float_to_int() → None[source]#: Convert self.lower and self.upper from np.float32 or np.float64 to int.

_get_env_mat_range(min_nbor_dist)[source]#: Change the embedding net range to sw / min_nbor_dist.

_spline5_switch(xx, rmin, rmax)[source]#