torch.Tensor.copy_

Tensor.copy_(src, non_blocking=False) → Tensor

Copies the elements from src into self tensor and returns self.

The src tensor must be broadcastable with the self tensor. It may be of a different data type or reside on a different device.

See Device-to-host copies on NVLink-C2C systems for guidance on repeated CUDA-to-CPU transfers and reusing CPU buffers.

Parameters:

• src (Tensor) – the source tensor to copy from

• non_blocking (bool, optional) – if True and this copy is between CPU and GPU, the copy may occur asynchronously with respect to the host. For other cases, this argument has no effect. Default: False

Note

When non_blocking is True and the copy is issued on a non-default CUDA stream, the caller is responsible for proper cross-stream synchronization. See CUDA streams for the required pattern.