I look forward to your pull-request. TinyNdArray supports only float array. In the following Python code dtype=float32 is omitted, and in C++ code assuming using namespace tinyndarray; is declared.
Results that may be inaccessible to you are currently showing.
Hide inaccessible results