create

Create tensors from other array-like data, and change device and type.

class AsTensor(dtype=None, device=None)[source]

Partial of the top-level PyTorch function as_tensor.

Parameters:

dtype (dtype, optional) – Torch dtype of the tensor to create. Defaults to None
device (device, optional) – Torch device to create the tensor on. Defaults to None

__call__(data)[source]

Create a PyTorch tensor with the cached (keyword) arguments.

Parameters:: data – Any type of data a PyTorch tensor can be created from.
Returns:: A PyTorch tensor sharing the memory with data.
Return type:: Tensor

class Create(dtype=None, device=None, requires_grad=False, pin_memory=False)[source]

Partial of the top-level PyTorch function tensor.

Parameters:

dtype (dtype, optional) – Torch dtype of the tensor to create. Defaults to None
device (device, optional) – Torch device to create the tensor on. Defaults to None
requires_grad (bool, optional) – If autograd should record operations on the returned tensor. Defaults to False.
pin_memory (bool, optional) – If set, returned tensor would be allocated in the pinned memory. Works only for CPU tensors. Defaults to False.

__call__(data)[source]

Create a PyTorch tensor with the cached (keyword) arguments.

Parameters:: data – Any type of data a PyTorch tensor can be created from.
Returns:: A PyTorch tensor.
Return type:: Tensor

class To(target, *args, **kwargs)[source]

Move or change a tensor or module to a different device or dtype.

Refer to the PyTorch documentation for more information.

Parameters:

target (device or dtype) – The device or dtype to move the tensor or module to.
*args – Additional argument to pass to the to method.
**kwargs – Additional keyword arguments to pass to the to method.

__call__(inp)[source]

Move or change a tensor or module to the specified target.

Parameters:: inp (Tensor or Module) – The input tensor or module to move or change to the target.
Returns:: The inp moved or changed to the cached target.
Return type:: Tensor or Module

from_dataframe(df)[source]

Convert a pandas dataframe to a PyTorch tensor.

This is simply wrapper around the top-level PyTorch function from_numpy function called on the dataframe’s values attribute.