espaloma.data.dataset.Dataset
- class espaloma.data.dataset.Dataset(graphs=None)[source]
Bases:
abc.ABC,torch.utils.data.dataset.DatasetThe base class of map-style dataset.
- Parameters
graphs (List) – objects in the dataset
- apply(fn, in_place=True)[source]
Apply a function to every graph in the dataset. If in_place=True, modify the graph in-place.
Note
This also supports iterative-style dataset by deleting __getitem__ and __len__ function.
- Variables
transforms (an iterable of callables that transforms the input.) – the __getiem__ method applies these transforms later.
Examples
>>> data = Dataset([esp.Graph("C")])
Methods
__init__([graphs])apply(fn[, in_place])Apply functions to the elements of the dataset.
load(path)Load path to dataset.
regenerate_impropers([improper_def])Regenerate the improper nodes for all graphs.
save(path)Save dataset to path.
shuffle([seed])split(partition)Split the dataset according to some partition.
subsample(ratio[, seed])Subsample the dataset according to some ratio.
- apply(fn, in_place=False)[source]
Apply functions to the elements of the dataset.
- Parameters
fn (callable)
Note
If in_place is False, fn is added to the transforms else it is applied to elements and modifies them.
- regenerate_impropers(improper_def='smirnoff')[source]
Regenerate the improper nodes for all graphs.
- Parameters
improper_def (str) – Which convention to use for permuting impropers.