PlibData¶
- class PlibData(data=None, data_sources=None, path_to_data_sources=None)[source]¶
Bases:
Dataset[OutT],Generic[OutT]Data structure for hosting perturb-lib data.
- Parameters:
data – data to initialize the class with.
data_sources (
UnionType[str,list[str],None]) – if data is None, data_sources can be used to specify the names of the data sources.
- apply_transform(transform)[source]¶
Apply a transformation to the data.
- Return type:
PlibData[TypeVar(NewOutT)]
- abstract property columns: list[str]¶
The list of column names.
- abstract property dtypes: dict¶
Dictionary of data types.
- abstract get_data_loader(batch_size, num_workers=0, pin_memory=False, shuffle=False)[source]¶
Fetch a torch-style data loader for batch sampling.
- Parameters:
batch_size (
Optional[int]) – The size of a batch to fetch in each iteration.num_workers (
int) – Number of pytorch workers.pin_memory (
bool) – If true, Copy Tensors into device pinned memory before returning them.shuffle (
bool) – If false, samples will be sampled sequentially to form batches. If true, samples will be shuffled.
- Return type:
DataLoader[TypeVar(OutT)]- Returns:
an instance of
torch.utils.data.DataLoader