CPU implementation of the code.

[aboudjelal]

Bonjour,

Is there any cpu implementation of the code ?
I read in your paper that you have used the ’astra gpu’ backend.

Regards

[adler-j]

The code should be runnable with a non-GPU version of Tensorflow and by selecting impl='astra_cpu' here

learned_gradient_tomography/code/partially_learned_gradient_descent.py

Line 17 in f358cc4

operator = odl.tomo.RayTransform(space, geometry)

Beware that it is probably extremely slow though.

[aboudjelal]

Thank you for your answer,

when I execute the code : python3 partially_learned_gradient_descent_no_gradients.py .
I show this message,

2019-03-18 14:48:17.850394: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2594125000 Hz
2019-03-18 14:48:17.850835: I tensorflow/compiler/xla/service/service.cc:150] XLA service 0x1ce2690 executing computations on platform Host. Devices:
2019-03-18 14:48:17.850875: I tensorflow/compiler/xla/service/service.cc:158] StreamExecutor device (0): ,
Traceback (most recent call last):
File "partially_learned_gradient_descent_no_gradients.py", line 17, in
operator = odl.tomo.RayTransform(space, geometry)
File "/home/abdel/.local/lib/python3.6/site-packages/odl/tomo/operators/ray_trafo.py", line 383, in init
variant='forward', **kwargs)
File "/home/abdel/.local/lib/python3.6/site-packages/odl/tomo/operators/ray_trafo.py", line 118, in init
raise RuntimeError('no ray transform back-end available; '
RuntimeError: no ray transform back-end available; this requires 3rd party packages, please check the install docs

thanks

[adler-j]

Did you install ASTRA?

https://odlgroup.github.io/odl/getting_started/installing_extensions.html#astra-for-x-ray-tomography

[aboudjelal]

no I didn't
i will try

[aboudjelal]

I installed ASTRA but some error

[adler-j]

Could you elaborate a bit more on what error?