原文地址:https://stackoverflow.com/questions/23359886/selecting-rows-in-numpy-ndarray-based-on-the-value-of-two-columns?rq=1
And here is the sample code. For simplicity I changed the size3600000 to 5.
a = np.array(xrange(5*3))a.shape =(5,3)
Now array a looklike this:
array([[0,1,2],[3,4,5],[6,7,8],[9,10,11],[12,13,14]])
If you want row with HUE=9, dolike this:
a[np.where(a[:,0]==9)]#array([[ 9, 10, 11]])
If you want row with VALUE=4, dolike this:
a[np.where(a[:,1]==4)]#array([[3, 4, 5]])
If you want rowwith HUE=0 and VALUE=1,do like this:
a[np.where((a[:,0]==0)*(a[:,1]==1))]#array([[0,1, 2]])
自己写一个小例子:
a=np.array(range(5*3))
print(a)
a.shape=(5,3) print(a)#使用print之后和单纯用变量结果显示并不完全一样 index=np.where((a[:,0]==9)) print(index) d=a[index] print(d) [out1]:
[out2]:[ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14]
[[ 0 1 2] [ 3 4 5] [ 6 7 8] [ 9 10 11] [12 13 14]][out3]: [[ 9 10 11]]