reference/generated/tensordict.TensorDictBase.irecv

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.irecv

TensorDictBase.irecv(src: int | None = None, *, group: 'torch.distributed.ProcessGroup' | None = None, group_src: int | None = None, return_premature: bool = False, init_tag: int = 0, pseudo_rand: bool = False, consolidated: bool = False) → tuple[int, list[torch.Future]] | list[torch.Future] | None

Receives the leaves that isend() or send() sends, one asynchronous transfer per leaf.

The tensordict must have the keys, shapes and dtypes of the tensordict that the sender sends; the values are written into its tensors in place. By default, the call waits until all the transfers are done. With return_premature=True, it returns at once: call wait() on each returned Work object before you read the tensordict. See isend() for an example, and Distributed tensordicts.

Parameters:

src (int, optional) – the global rank of the sender. Exclusive with group_src: pass exactly one of them. Defaults to None.

Keyword Arguments:

• group (torch.distributed.ProcessGroup, optional) – the process group. Defaults to None, the default process group.

• group_src (int, optional) – the rank of the sender in group, which must be given. Exclusive with src. As in recv(), group can then be a standalone process group. Defaults to None.

• return_premature (bool, optional) – if True, returns the list of Work objects of the transfers without waiting for them. Defaults to False.

• init_tag (int, optional) – the init_tag that the sender passed. Defaults to 0.

• pseudo_rand (bool, optional) – the pseudo_rand that the sender passed. Defaults to False.

• consolidated (bool, optional) – if True, receives one message into the consolidated storage of the tensordict, as recv(consolidated=True) does. The tensordict must be consolidated already. init_tag and pseudo_rand are ignored. Defaults to False.

Returns:

With return_premature=True, a list of Work objects. Otherwise, None.

Raises:

ValueError – if both or none of src and group_src are given, or if group_src is given without group.

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