I'll cut to the chase, so let's get right to the code~
inputs = Variable((2,2)) inputs.is_cuda # will return false inputs = Variable((2,2).cuda()) inputs.is_cuda # returns true
Judgment:
torch.is_tensor() # return true if it is a pytorch tensor type
torch.is_storage() # if it is the storage type of pytorch return ture
Here's another tip, if you need to determine whether the tensor is empty or not, you can do it as follows
>>> a=() >>> len(a) 0 >>> len(a) is 0 True
Settings: Through some built-in functions, you can set the precision, type and print parameters of the tensor.
torch.set_default_dtype(d) # Set the default floating point type for () torch.set_default_tensor_type() # Ibid, set default tensor type for () >>> ([1.2, 3]).dtype # initial default for floating point is torch.float32 torch.float32 >>> torch.set_default_dtype(torch.float64) >>> ([1.2, 3]).dtype # a new floating point tensor torch.float64 >>> torch.set_default_tensor_type() >>> ([1.2, 3]).dtype # a new floating point tensor torch.float64 torch.get_default_dtype() # Get the current default floating point type torch.set_printoptions(precision=None, threshold=None, edgeitems=None, linewidth=None, profile=None）#) ## set upprintingThe print parameters of the
Determining the type of a variable: either of the following methods will work
if isinstance(downsample, ):
# if (downsample) != :
Additional knowledge:pytorch: test if GPU is available
Without further ado, let's look at the code.
import torch
flag = .is_available()
print(flag)
ngpu= 1
# Decide which device we want to run on
device = ("cuda:0" if (.is_available() and ngpu > 0) else "cpu")
print(device)
print(.get_device_name(0))
print((3,3).cuda())
True cuda:0 GeForce GTX 1080 tensor([[0.9530, 0.4746, 0.9819], [0.7192, 0.9427, 0.6768], [0.8594, 0.9490, 0.6551]], device='cuda:0')
Above this pytorch judgment whether cuda Judgment variable type way is all I share with you, I hope to give you a reference, and I hope you support me more.