Quansight-Labs/numpy.net

np.delete should not flatten its output

Happypig375 opened this issue · 2 comments

Compare

using NumpyDotNet;
System.Console.WriteLine(np.delete(np.array(new[,]{
    {"0", "1", "2"},
    {"1", "0", "2"},
  }), 1, 1))
STRING 
{ 0, 2, 1, 2 }
import numpy as np
print(np.delete(np.array([["0", "1", "@"], ["1", "0", "@"]]), 1, 1))
[['0' '@']
 ['1' '@']]

Are you hitting this string conversion issue in a real project you are doing or are you just testing it out?

issue still present