tf.keras.ops.nanargmin

Returns the indices of the minimum values along an axis, ignoring NaNs.

tf.keras.ops.nanargmin(
    x, axis=None, keepdims=False
)

x Input tensor.
axis By default, the index is into the flattened tensor, otherwise along the specified axis.
keepdims If this is set to True, the axes which are reduced are left in the result as dimensions with size one. Defaults to False.

Tensor of indices. It has the same shape as x, with the dimension along axis removed. NaN values are ignored when computing the minimum.

Examples:

import numpy as np
from keras import ops
x = np.array([[1.0, np.nan, 3.0],
              [np.nan, 2.0, 0.0]])
 
ops.nanargmin(x)
array(5, dtype=int32)
 
ops.nanargmin(x, axis=0)
array([0, 1, 1], dtype=int32)
 
ops.nanargmin(x, axis=1)
array([0, 2], dtype=int32)
 
ops.nanargmin(x, axis=1, keepdims=True)
array([[0],
       [2]], dtype=int32)