Data Options

data_options:

type: dict

argument path: data_options

The options for dataset settings in training.

r_max:

type: str | float | int, optional, default: 5.0

argument path: data_options/r_max

r_max

oer_max:

type: str | float | int, optional, default: 5.0

argument path: data_options/oer_max

oer_max

er_max:

type: str | float | int, optional, default: 5.0

argument path: data_options/er_max

er_max

train:

type: dict

argument path: data_options/train

The dataset settings for training.

type:

type: str, optional, default: DefaultDataset

argument path: data_options/train/type

The type of dataset.

root:

type: str

argument path: data_options/train/root

This is where the dataset stores data files.

prefix:

type: str | NoneType, optional, default: None

argument path: data_options/train/prefix

The prefix of the folders under root, which will be loaded in dataset.

separator:

type: str, optional, default: .

argument path: data_options/train/separator

the sepatator used to separate the prefix and suffix in the dataset directory. Default: ‘.’

get_Hamiltonian:

type: bool, optional, default: False

argument path: data_options/train/get_Hamiltonian

Choose whether the Hamiltonian blocks (and overlap blocks, if provided) are loaded when building dataset.

get_overlap:

type: bool, optional, default: False

argument path: data_options/train/get_overlap

Choose whether the overlap blocks are loaded when building dataset.

get_DM:

type: bool, optional, default: False

argument path: data_options/train/get_DM

Choose whether the density matrix is loaded when building dataset.

get_eigenvalues:

type: bool, optional, default: False

argument path: data_options/train/get_eigenvalues

Choose whether the eigenvalues and k-points are loaded when building dataset.

validation:

type: dict, optional

argument path: data_options/validation

The dataset settings for validation.

type:

type: str, optional, default: DefaultDataset

argument path: data_options/validation/type

The type of dataset.

root:

type: str

argument path: data_options/validation/root

This is where the dataset stores data files.

prefix:

type: str | NoneType, optional, default: None

argument path: data_options/validation/prefix

The prefix of the folders under root, which will be loaded in dataset.

separator:

type: str, optional, default: .

argument path: data_options/validation/separator

the sepatator used to separate the prefix and suffix in the dataset directory. Default: ‘.’

get_Hamiltonian:

type: bool, optional, default: False

argument path: data_options/validation/get_Hamiltonian

Choose whether the Hamiltonian blocks (and overlap blocks, if provided) are loaded when building dataset.

get_overlap:

type: bool, optional, default: False

argument path: data_options/validation/get_overlap

Choose whether the overlap blocks are loaded when building dataset.

get_DM:

type: bool, optional, default: False

argument path: data_options/validation/get_DM

Choose whether the density matrix is loaded when building dataset.

get_eigenvalues:

type: bool, optional, default: False

argument path: data_options/validation/get_eigenvalues

Choose whether the eigenvalues and k-points are loaded when building dataset.

reference:

type: dict, optional

argument path: data_options/reference

The dataset settings for reference.

type:

type: str, optional, default: DefaultDataset

argument path: data_options/reference/type

The type of dataset.

root:

type: str

argument path: data_options/reference/root

This is where the dataset stores data files.

prefix:

type: str | NoneType, optional, default: None

argument path: data_options/reference/prefix

The prefix of the folders under root, which will be loaded in dataset.

separator:

type: str, optional, default: .

argument path: data_options/reference/separator

the sepatator used to separate the prefix and suffix in the dataset directory. Default: ‘.’

get_Hamiltonian:

type: bool, optional, default: False

argument path: data_options/reference/get_Hamiltonian

Choose whether the Hamiltonian blocks (and overlap blocks, if provided) are loaded when building dataset.

get_overlap:

type: bool, optional, default: False

argument path: data_options/reference/get_overlap

Choose whether the overlap blocks are loaded when building dataset.

get_DM:

type: bool, optional, default: False

argument path: data_options/reference/get_DM

Choose whether the density matrix is loaded when building dataset.

get_eigenvalues:

type: bool, optional, default: False

argument path: data_options/reference/get_eigenvalues

Choose whether the eigenvalues and k-points are loaded when building dataset.