关于.squeeze()方法
输入:
a = torch.rand([4,1,3,2,1,5])
print(a.shape)
b = a.squeeze()
print(b.shape)
a1 = torch.rand([4,1,3,2,1,5])
print(a1.shape)
c = a1[0].squeeze()
print(c.shape)
d = a1[1].squeeze()
print(d.shape)
e = a1[2].squeeze()
print(e.shape)
输出:
torch.Size([4, 1, 3, 2, 1, 5])
torch.Size([4, 3, 2, 5])
torch.Size([4, 1, 3, 2, 1, 5])
torch.Size([3, 2, 5])
torch.Size([3, 2, 5])
torch.Size([3, 2, 5])
Process finished with exit code 0
.squeeze(a,axis = None)
axis的取值可为None 或 int 或 tuple of ints, 可选。若axis为空,则删除所有单维度的条目;