np.argsort() throw IndexOutOfRangeException

[rainyl]

Description

When applying np.argsort() to a ndarray with dtype of np.Float32, the IndexOutOfRangeException will be thrown, with dtype=np.Float64, it worked as expected.

Minimum sample to reproduce

using NumpyDotNet;

namespace Test
{
    public class TestClass
    {
        public static void Main()
        {
            var rng = new np.random();
            var arr = rng.normal(4000, 1, new shape(3000, 2)).astype(np.Float32);
            var x = arr.T[0, ":"] as ndarray;
            var y = arr.T[1, ":"] as ndarray;
            var idx = np.argsort(x);
       }
    }
}

and the error message:

Unhandled exception. System.IndexOutOfRangeException: Index was outside the bounds of the array.
   at NumpyLib.CopyHelper`1.strided_byte_copy(Int64 dest_offset, Int64 src_offset, Int64 N)
   at NumpyLib.numpyinternal._new_argsort(NpyArray op, Int32 axis, NPY_SORTKIND kind)
   at NumpyLib.numpyinternal.NpyArray_ArgSort(NpyArray op, Int32 axis, NPY_SORTKIND kind)
   at NumpyLib.numpyAPI.NpyArray_ArgSort(NpyArray op, Int32 axis, NPY_SORTKIND which)
   at NumpyDotNet.NpyCoreApi.ArgSort(ndarray arr, Int32 axis, NPY_SORTKIND sortkind)
   at NumpyDotNet.np.argsort(ndarray a, Nullable`1 axis, NPY_SORTKIND kind, IEnumerable`1 order)
   at Test.TestClass.Main() 

I am not familiar enough with this project for now, so, if there are something wrong with my sample, please fix me :-)

[KevinBaselinesw]

I just pushed a new version 0.9.83.7 that will fix this problem.

thank you!

[rainyl]

I just pushed a new version 0.9.83.7 that will fix this problem.

thank you!

Great! thanks for your works!