reference/generated/tensordict.TensorDictBase.is_shared

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.is_shared

TensorDictBase.is_shared() → bool

Checks if tensordict is in shared memory.

If a TensorDict instance is in shared memory, it is locked (entries cannot be renamed, removed or added). If a TensorDict is created with tensors that are all in shared memory, this does __not__ mean that is_shared will return True (as a new tensor may or may not be in shared memory). Only if one calls tensordict.share_memory_() or places the tensordict on a device where the content is shared by default (eg, "cuda") will the tensordict be considered in shared memory.

This is always True for tensordicts on a CUDA device.

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