nd.array.where

http://mxnet.apache.org/api/python/ndarray/ndarray.html#mxnet.ndarray.where

Return the elements, either from x or y, depending on the condition.

Given three ndarrays, condition, x, and y, return an ndarray with the elements from x or y, depending on the elements from condition are true or false. x and y must have the same shape. If condition has the same shape as x, each element in the output array is from x if the corresponding element in the condition is true, and from y if false.

If condition does not have the same shape as x, it must be a 1D array whose size is the same as x’s first dimension size. Each row of the output array is from x’s row if the corresponding element from condition is true, and from y’s row if false.

Note that all non-zero values are interpreted as True in condition.

x = [[1, 2], [3, 4]]
y = [[5, 6], [7, 8]]
cond = [[0, 1], [-1, 0]]

where(cond, x, y) = [[5, 2], [3, 8]]

csr_cond = cast_storage(cond, 'csr')

where(csr_cond, x, y) = [[5, 2], [3, 8]]

X，Y必须同样形状，然后条件可以为同样形状，也可以是一维的；条件为0，才为Y，否则正数，负数都为X

from mxnet import nd

ious = nd.array([[0.1,0.3,0.3],[0.9,0.4,0.01],[0,0.55,2],[0.56,0.77,3],[0.9,0.73,4]])
print(ious)

label = [0. ,1., 1.]
print(label)

flag = nd.ones_like(ious)

y = nd.full(ious.shape,0.5)

for i, hard in enumerate(label):
    if hard == 1.0:
        flag[:,i] = ious[:,i] < 0.7
print(flag)

ious = nd.where(flag,ious,y)
print(ious)