PyTorch torch.as_tensor Function
Pytorch torch Reference Manual
torch.as_tensoris a function in PyTorch used to convert data into tensors. Unliketorch.tensorit shares data memory as much as possible.
Function Definition
torch.as_tensor(data, dtype=None, device=None)
Usage Example
Example
import torch
import numpy as np
# Create from a NumPy array (shared memory)
numpy_array = np.array([1, 2, 3])
x = torch.as_tensor(numpy_array)
# Modify the NumPy array
numpy_array[0] = 100
print("Tensor:", x) # The tensor will also change
import numpy as np
# Create from a NumPy array (shared memory)
numpy_array = np.array([1, 2, 3])
x = torch.as_tensor(numpy_array)
# Modify the NumPy array
numpy_array[0] = 100
print("Tensor:", x) # The tensor will also change
The output result is:
张量: tensor([100, 2, 3])
Other Extensions