NumPy | nanargmin method
Start your free 7-days trial now!
Numpy's nanargmin(~) method ignores all missing values (i.e. NaNs) returns the index that corresponds to the smallest element in the array.
Parameters
1. a | array_like
The input array.
2. axis | int | optional
The axis along which to compute the method. For 2D arrays, if axis=0, then the method is performed column-wise, and if axis=1 then row-wise. If no axis is provided, then Numpy will deem your array as a flattened array.
Return value
If no axis is provided, then a scalar is returned. Otherwise, a Numpy array is returned.
Examples
One-dimensional arrays
x = np.array([5,np.NaN,1,3])np.nanargmin(x)
2
Here, 2 is returned because the smallest value (i.e. 1) is located at index 2. In contrast, the np.argmin(x) method would return 1 since it considers NaN to be the smallest.
Two-dimensional arrays
Suppose we have the following 2D Numpy array:
x = np.array([[5,np.NaN],[1,3]])x
array([[ 5., nan], [ 1., 3.]])
Min index of entire array
To obtain the index of the minimum value in the entire array, leave out the axis parameter:
np.nanargmin(x)
2
Min indices of every column
To obtain the index of the minimum values column-wise, set axis=0:
np.nanargmin(x, axis=0)
array([1, 1])
Here, we're going over each column of the matrix and computing the index of its smallest value, while ignoring any missing values.
Min indices of every row
To obtain the index of the minimum values row-wise, set axis=1:
np.nanargmin(x, axis=1)
array([0, 0])
Here, we're going over each row of the matrix and computing the index of its smallest value, while ignoring any missing values.
