Torch.unsqueeze

Pytorch squeeze

x = torch.zeros(2, 1, 2, 1, 2)
x.size()
>>> torch.Size([2, 1, 2, 1, 2])

y = torch.squeeze(x) # remove 1
y.size()
>>> torch.Size([2, 2, 2])

y = torch.squeeze(x, 0)
y.size()
>>> torch.Size([2, 1, 2, 1, 2])

y = torch.squeeze(x, 1)
y.size()
>>> torch.Size([2, 2, 1, 2])

Source: pytorch.org

Torch.unsqueeze

a = torch.randn(4, 4, 4)
>>> torch.unsqueeze(a, 0).size()
torch.Size([1, 4, 4, 4])

>>> torch.unsqueeze(a, 1).size()
torch.Size([4, 1, 4, 4])

>>> torch.unsqueeze(a, 2).size()
torch.Size([4, 4, 1, 4])

>>> torch.unsqueeze(a, 3).size()
torch.Size([4, 4, 4, 1])

Source: linuxpip.org

Similar codes

30 votes180 views2 answers
Python clear screen windows and linux
Python
1 vote253 views1 answer
Python remove all except numbers
Python
0 vote224 views2 answers
Data frame list value change to string
Python
28 votes417 views3 answers
Django abstractuser
Python Django
-1 vote195 views1 answer
Groupby where only
Python
1 vote198 views1 answer
Django how to set a navbar active
Python
0 vote193 views1 answer
Get maximum value index after groupby
Python
1 vote176 views1 answer
Pandas dataframe lists as columns
Python
0 vote204 views1 answer
Pandas dataframe to excel hyperlink length limit
Python
1 vote207 views1 answer
Python stock prediction code
Python