torchcvnn.transforms.ToTensor

class torchcvnn.transforms.ToTensor(dtype: str)

Converts numpy array or torch tensor to torch tensor of specified dtype. This transform converts input data to a PyTorch tensor with the specified data type. It handles both numpy arrays and existing PyTorch tensors as input.

Parameters:

dtype (str) – Target data type for the output tensor. Should be one of PyTorch’s supported dtype strings (e.g. ‘float32’, ‘float64’, ‘int32’, etc.)

Returns:

The converted tensor with the specified dtype.

Return type:

torch.Tensor

Example

>>> transform = ToTensor(dtype='float32')
>>> x_numpy = np.array([1, 2, 3])
>>> x_tensor = transform(x_numpy)  # converts to torch.FloatTensor
>>> x_existing = torch.tensor([1, 2, 3], dtype=torch.int32)
>>> x_converted = transform(x_existing)  # converts to torch.FloatTensor

__init__(dtype: str) → None

Methods

__init__(dtype)