例如,给定矩阵a = [[1,2,3], [4,5,6], [7,8,9]]
将矩阵a中不为0的元素挑选出来组合成为一维向量:
c = [1, 2, 3, 4, 5, 6, 7, 8, 9]
一般思路就是两层for循环挨个添加,有更加简洁的方式:
a = torch.from_numpy(np.array([[1,2,3], [4,5,6], [7,8,9]]))
b = a.nonzero()
c= a[b[:,0], b[:,1]]
np.array(list):将list转换为numpy矩阵
torch.from_numpy(nparray):将numpy矩阵转换为tensor数组
tensor.nonzero():返回tensor矩阵中不为0元素的索引,对于2维矩阵来说,索引就是n*2的维度;
一些结果:
a = torch.from_numpy(np.array([[1,2,3], [4,5,6], [7,8,9]]))
print(a)
tensor([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]], dtype=torch.int32)
a = torch.from_numpy(np.array([[1,2,3], [4,5,6], [7,8,9]]))
b = a.nonzero()
print(b)
tensor([[0, 0],
[0, 1],
[0, 2],
[1, 0],
[1, 1],
[1, 2],
[2, 0],
[2, 1],
[2, 2]])
a = torch.from_numpy(np.array([[1,2,3], [4,5,6], [7,8,9]]))
b = a.nonzero()
print(b[:,0])
print(b[:,1])
tensor([0, 0, 0, 1, 1, 1, 2, 2, 2])
tensor([0, 1, 2, 0, 1, 2, 0, 1, 2])
a = torch.from_numpy(np.array([[1,2,3], [4,5,6], [7,8,9]]))
b = a.nonzero()
c= a[b[:,0], b[:,1]]
print(c)
tensor([1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=torch.int32)