search

inducer / pyopencl

OpenCL integration for Python, plus shiny features
http://mathema.tician.de/software/pyopencl
Other
1.06k stars 241 forks source link

Array.concatenate fails for axis!= 0 #506

Open panchoop opened 3 years ago

panchoop commented 3 years ago

The function only works for axis = 0.

To replicate:

import pyopencl as cl
platform = cl.get_platforms()[0]
device = platform.get_devices()[0]
context = cl.create_some_context([device])
queue = cl.CommandQueue(context)

a = np.random.rand(2,2)
b = np.random.rand(2,2)

a_cl = clarray.to_device(queue, np.require(a, np.float64, 'C'))
b_cl = clarray.to_device(queue, np.require(b, np.float64, 'C'))

c_cl = clarray.concatenate( (a_cl, b_cl), axis=1)

Error: "cannot assign between arrays of differing strides"

The error appears to be originated when creating the buffer results, that is expected to contain the whole data. Since it is initialized with shape (2, 2+2), its stride is different from both a, and b. Then Array.setitem fails as it is no yet implemented to set items with different stride.

I am unsure if it is a bug. If it is not yet implemented, maybe it could be handy to raise an error for axis != 0 inside concatenate.

inducer commented 3 years ago

That's a limitation of the current implementation. There are not currently plans to address this. I'd be open to considering a PR that adds a better error message.

cc @kaushikcfd