torch_adata._tools._split

Module Contents

Functions

uniform_group_sizes(n_cells[, n_groups])	Split groups based on number of cells and groups.
proportioned_group_sizes(dataset, percentages, ...)	Split dataset length with specified proportions/number of groups
calculate_split_lengths(dataset, n_groups, ...)	Split the length of the dataset into proportioned subsets
split(dataset, n_groups, percentages, , ...)	Split dataset using torch.utils.data.random_split.

Attributes

__module_name__
__doc__
__author__
__email__

torch_adata._tools._split.__module_name__ = _split.py

torch_adata._tools._split.__doc__ = Helper function(s) for splitting the dataset.

torch_adata._tools._split.__author__

torch_adata._tools._split.__email__

torch_adata._tools._split.uniform_group_sizes(n_cells: int, n_groups: int = 2)

Split groups based on number of cells and groups.

torch_adata._tools._split.proportioned_group_sizes(dataset: torch.utils.data.Dataset, percentages: list([float, '...', float]), remainder_idx: int = - 1)

Split dataset length with specified proportions/number of groups

torch_adata._tools._split.calculate_split_lengths(dataset: torch.utils.data.Dataset, n_groups: int = 2, percentages: list([float, '...', float]) = None)

Split the length of the dataset into proportioned subsets

torch_adata._tools._split.split(dataset: torch.utils.data.Dataset, n_groups: int = 2, percentages: list([float, '...', float]) = None)

Split dataset using torch.utils.data.random_split.

dataset

type: torch.utils.data.Dataset

n_groups

type: int default: 2

percentages

type: list([float, …, float]) default: None

list([torch.utils.data.Dataset, …, torch.utils.data.Dataset])

1. Uses the torch.utils.data.random_split function to actually do the split.