samdporter

> `GradientOperator` with C backend expects contiguous data as in a numpy ndarray or a float* C array. I think it gets passed a SIRF object with which it doesn't...

`grad = operators.GradientOperator(fdg, method='backward')` This actually runs because method = 'backward' uses the Numpy backend. I've update the notebook with a c-backend grad operator that does fail (apologies) I agree...

Hi, update: Setting up and running two SPECT reconstructions as in https://github.com/samdporter/SPECTissues/blob/main/OSEM.py results in error: ``` terminate called after throwing an instance of 'std::__cxx11::basic_string' Aborted (core dumped) ``` I'm having...

Hey @paskino, Thanks for looking at this This did have some unintended consequences, unfortunately, so I stopped using it in my code and instead added a `as_array()` Function for a...

I'm doing some synergistic deconvolution. I'm working with SIRF which means I need a `BlockDataContainer `for multiple channels. The problem comes when I define a operator like this: `k =...

`initial_estimate = BlockDataContainer(image0, image1)` but `spdhg_f1 = KullbackLeibler(b=data0)` needs a `DataContainer`. This is required so that `image0`, `image1` can have separate data fidelity functions but share the same prior.