reference/generated/tensordict.TensorDictBase.pin_memory_

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.pin_memory

TensorDictBase.pin_memory_(num_threads: int = 0) → Self

Pins the tensor leaves of the tensordict and returns self.

Deprecated since version 0.15: Use pin_memory(inplace=True) instead. Pinning allocates new storage, so it does not fit the trailing-underscore convention (see In-place operations). pin_memory_ will be removed in TensorDict 0.17.

Parameters:

num_threads (int, optional) – the number of threads that pin the leaves, as in pin_memory(). Defaults to 0, which pins them on the calling thread.

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