Thank you. Sorry, kindergarten error. I should have read the numpy doc more carefully.
> > On ,Fri Oct 04 2024 18:02:46 GMT+1000 (Australian Eastern Standard Time), > Philipp Niedermayer <p.niederma...@gsi.de> wrote: > > ---------- Original Message ---------- > > > > > Float32 has been deprecated (Numpy 1.20, 1.24). > > > > > That is not true. What they deprecated and removed is np.float, you must > now explicitly use float32 (aka single) or float64 (aka double). And there > are many more float precision available. > > > > > real_signal = np. fromfile ( fname , dtype = "float32 " ) > > > > > iq_signal = np. fromfile ( fname , dtype = "complex64 " ) > > > > > > > > > > > > > > > > > *Von:* dave_a...@bigpond.com > > > > > *Gesendet:* Donnerstag, 3. Oktober 2024 um 8:19 AM MESZ > > > > > *An:* discuss-gnuradio@gnu.org > > > > > *Betreff:* Reading File Sink output in Python > > > > >> I have been trying to read, with Python/Numpy, a file created by a File >> Sink. That file is for a stream of floats. >> >> Numpy is 1.21.5. GNU Radio is 3.10.11. >> >> The File Sink wiki page, in the hints on reading from Python, gives >> float32 as a data type. Float32 has been deprecated (Numpy 1.20, 1.24). The >> default (only?) float data type in Numpy now is float64. Is this a known >> issue, please? >> >> The C code on the wiki page works fine. >> >> Perhaps there's another simple workaround? >> >> >> >> Dave Abel >> >> >> >> >> >> >> >> > >
