Dear TI:
We have a deep learning network model and its postProcType is set to be "0" zero and according to TIDL document (md_tidl_sample_test.html), it means "disable post processing on output tensor".
We want to know how to parse the outputs by our self. We run the PC simulation tool (PC_dsp_test_dl_algo.out) and got the output in ./testvecs/output/<xxx>.bin. The size of output bin file is just the same as our expect (94000 x 2 bytes = 188000 bytes) but it's fixed point. How could we parse the output and what's the 2 byes means, could we express it as float point? Thanks.
