Vispy Code (6.6) Error

[Ogi Moore]

In the 6th python cell:

program['a_position'] = np.c_[
        np.linspace(-1.0, +1.0, 1000),
        np.random.uniform(-0.5, +0.5, 1000)]

results in:

TypeError: data must be 32-bit not float64

I may mess with this later, but I know next to nothing about vispy so I'm not sure how well I can troubleshoot the issue.

[Cyrille Rossant]

try to append .astype(np.float32) at the end of the array
float64 doesn't work well on GPUs so VisPy requires explicit cast to float32

[Ogi Moore]

That was going to be my first try, perhaps I should have done that first instead of opening an issue... still learning thing 'git' thing and associated etiquette. Thanks for the fast response!

[Cyrille Rossant]

closed by #23

[giumas]

It looks like this fix did not propagate to http://ipython-books.github.io/featured-06/

[Eddy Hartley]

Specifically

program['a_position'] = np.c_[
        np.linspace(-1.0, +1.0, 1000),
        np.random.uniform(-0.5, +0.5, 1000)].astype(np.float32)

[Cyrille Rossant]

Should be fixed now