3-21Pytorch与张量的索引与数据筛选

import torch
#torch.where

a = torch.rand(4,4)
b = torch.rand(4,4)
out = torch.where(a>0.5,a,b)

print(a)
print(b)
print(out)

tensor([[0.9875, 0.0399, 0.4628, 0.0218],
        [0.6645, 0.6531, 0.6494, 0.4461],
        [0.0189, 0.6904, 0.1290, 0.1731],
        [0.0277, 0.1196, 0.8279, 0.6042]])
tensor([[0.0552, 0.7499, 0.3638, 0.6014],
        [0.0851, 0.3886, 0.7147, 0.1416],
        [0.6719, 0.7295, 0.0700, 0.0164],
        [0.7201, 0.5614, 0.7305, 0.7452]])
tensor([[0.9875, 0.7499, 0.3638, 0.6014],
        [0.6645, 0.6531, 0.6494, 0.1416],
        [0.6719, 0.6904, 0.0700, 0.0164],
        [0.7201, 0.5614, 0.8279, 0.6042]])

#torch.index_select()


a = torch.rand(4,4)
#b = torch.rand(4,4)
print(a)
print(torch.index_select(a,dim = 0,index=torch.tensor([0,3,2])))
print(torch.index_select(a,dim = 1,index=torch.tensor([0,3,2])))

tensor([[0.2547, 0.1544, 0.5950, 0.1760],
        [0.3791, 0.4782, 0.5972, 0.6868],
        [0.4553, 0.2166, 0.5993, 0.2549],
        [0.4715, 0.8413, 0.7631, 0.6736]])
tensor([[0.2547, 0.1544, 0.5950, 0.1760],
        [0.4715, 0.8413, 0.7631, 0.6736],
        [0.4553, 0.2166, 0.5993, 0.2549]])
tensor([[0.2547, 0.1760, 0.5950],
        [0.3791, 0.6868, 0.5972],
        [0.4553, 0.2549, 0.5993],
        [0.4715, 0.6736, 0.7631]])

#torch.gather
a = torch.linspace(1,16,16).view(4,4)

print(a)

out = torch.gather(a, dim = 0,
             index = torch.tensor([[0,1,1,1],
                                   [0,1,2,2],
                                   [0,1,3,3]]))
print(out)
print(out.shape)
#dim = 0,out[i,j,k] = input[index[i,j,k],j,k]
#dim = 1,out[i,j,k] = input[i,index[i,j,k],k]
#dim = 0,out[i,j,k] = input[i,j,index[i,j,k],j,k]

tensor([[ 1.,  2.,  3.,  4.],
        [ 5.,  6.,  7.,  8.],
        [ 9., 10., 11., 12.],
        [13., 14., 15., 16.]])
tensor([[ 1.,  6.,  7.,  8.],
        [ 1.,  6., 11., 12.],
        [ 1.,  6., 15., 16.]])
torch.Size([3, 4])

a = torch.linspace(1,16,16).view(4,4)
mask = torch.gt(a,8)
print(a)
print(mask)
out = torch.masked_select(a, mask)
print(out)

tensor([[ 1.,  2.,  3.,  4.],
        [ 5.,  6.,  7.,  8.],
        [ 9., 10., 11., 12.],
        [13., 14., 15., 16.]])
tensor([[False, False, False, False],
        [False, False, False, False],
        [ True,  True,  True,  True],
        [ True,  True,  True,  True]])
tensor([ 9., 10., 11., 12., 13., 14., 15., 16.])

#torch.take
a = torch.linspace(1,16,16).view(4,4)
b = torch.take(a,index = torch.tensor([0,15,13,10]))
print(b)

tensor([ 1., 16., 14., 11.])

#torch.nonzero
a = torch.tensor([[0,1,2,0],[2,3,0,1]])
out = torch.nonzero(a)
print(a)
print(out)
#稀疏表示

tensor([[0, 1, 2, 0],
        [2, 3, 0, 1]])
tensor([[0, 1],
        [0, 2],
        [1, 0],
        [1, 1],
        [1, 3]])