Nanmin

Nanmin is a numerical computing function in NumPy that computes the minimum value of an array or along a specified axis while ignoring NaN (not-a-number) values. It is part of NumPy’s family of “nan” functions, which provide reductions that skip invalid or missing entries in the data.

The function accepts an input array and optional parameters that control the reduction. The axis parameter specifies the dimension to reduce over; if axis is None, the operation is applied to the flattened array. Other options include out for an alternative output array, keepdims to retain reduced dimensions, and initial or where to influence the computation. When there are NaNs in the data, nanmin disregards them in the calculation. If all elements along the reduced axis are NaN, the result is NaN and a runtime warning such as “All-NaN slice encountered” may be issued.

Nanmin is essentially equivalent to applying the standard minimum reduction to the data after removing or masking NaN values. It is often used in data cleaning and analysis workflows where missing values must be ignored rather than propagated.

Examples illustrate its behavior. For instance, given a = [1.0, NaN, 3.0], nanmin(a) yields 1.0. For a 2D array b = [[1, NaN], [3, 4]], nanmin(b, axis=0) returns [1.0, 4.0], since NaN is ignored in each column. If all values are NaN, such as [NaN, NaN], nanmin returns NaN.

Related functions include nanmax, nanmean, and other NumPy nan- reductions, which provide similar behavior for maximum, average, and other statistics with NaN handling.