adler-j / learned_gradient_tomography Goto Github PK
View Code? Open in Web Editor NEWSolving ill-posed inverse problems using iterative deep neural networks
Solving ill-posed inverse problems using iterative deep neural networks
Bonjour,
Is there any cpu implementation of the code ?
I read in your paper that you have used the ’astra gpu’ backend.
Regards
Hi, I'm currently using your learned gradient tomography in a 3D case.
I would like to use a poisson noise instead of a white noise
So in your generate_data function I replace:
data = operator(phantom)
noisy_data = data + odl.phantom.white_noise(operator.range) * np.mean(np.abs(data)) * 0.05
fbp = pseudoinverse(noisy_data)
By:
data = operator(phantom)
noise = np.random.poisson(0.05,size=operator.range.shape)
noisy_data = data + noise
fbp = pseudoinverse(noisy_data)
But I don't get a good result, the image is only noisy in some places
After some researches I noticed that it was due to the following code:
# Ensure operator has fixed operator norm for scale invariance
opnorm = odl.power_method_opnorm(operator)
operator = (1 / opnorm) * operator
pseudoinverse = pseudoinverse * opnorm
But if I remove this part of code the network does not learn anymore.
I would like to know if there is a clean way to apply a good poisson noise on data.
Thank you in advance for your assistance with this.
Hi, I am learning your paper and code.
I installed ODL correctly according to the commands :
'$ conda install -c odlgroup odl matplotlib pytest scikit-image spyder’,
but when I run these program, I have this error message:
space = odl.uniform_discr([-1, -1], [1, 1], [100, 100])
AttributeError: module 'odl' has no attribute 'uniform_discr'.
I'm sure that I have install discr. How can I solve it?
I run these codes in pycharm.
This is the specific information of some packages I installed.
python 3.6.13
odl 0.7.0
numpy 1.19.5
By the way, the URL https://github.com/adler-j/odl/archive/tensorflow_support.zip is no longer accessible.
According to odlgroup/odl#1092
Hey Jonas,
I was trying to run code: "Partially_learned_gradient_descent.py" in this repository but I have encountered a dilemma as follow.
In order to utilize GPU for the computation of forward and backward projection in ODL, it is mandatory for us to install "Astra-toolbox". However, installing "Astra-toolbox" will downgrade package "numpy" to version 1.12. If we import tensorflow (version 1.13.1) to python in this situation, the following error message will occur:
ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
ImportError: numpy.core.multiarray failed to importThe above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "", line 968, in _find_and_load
SystemError: <class '_frozen_importlib._ModuleLockManager'> returned a result with an error set
ImportError: numpy.core._multiarray_umath failed to import
ImportError: numpy.core.umath failed to import
I have tried to upgrade "numpy" (for instance, to version 1.16) after installing "astra-toolbox", but then ODL will complain there is no available back-end for Ray-Transform.
Have you met this problem before? If so, could you please tell me how could I solve it? Thanks for your attention, I am looking forward to hear back from you.
Cheers,
Bo
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.