reference/generated/tensordict.TensorDictBase.is_memmap

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.is_memmap

TensorDictBase.is_memmap() → bool

Whether the tensordict is memory-mapped.

True for the tensordict of memmap_(), and for the results of memmap() and memmap_like() written to a directory or to anonymous memory (not to an archive).

A memory-mapped tensordict is locked (see Locking and temporary changes), and unlocking it makes is_memmap return False. A tensordict built from memory-mapped tensors, or returned by load_memmap(), is not memory-mapped in this sense: a new entry would not be.

Examples

>>> import torch
>>> from tensordict import TensorDict
>>> td = TensorDict(a=torch.zeros(3))
>>> td.is_memmap(), td.memmap().is_memmap()
(False, True)

Every tensordict class and every tensorclass has this method. For the other methods, see Methods by task.