Alexander Skorikov

Since SciPy 1.9.0 sparse matrices with `object`datatype have been deprecated completely (https://github.com/scipy/scipy/pull/15828) and now this issue breaks `BroadcastOperator` and many examples/use cases that require it.

@sem-geologist Although it might not be very useful to HyperSpy, could you still share your ideas on how live/dead time can be calculated from the information contained in .bcf file?...

Yes, also see: https://stackoverflow.com/a/61054845/11968043

For testing, images should be put in the `test` folder. See also: https://github.com/OniroAI/MonoDepth-PyTorch#dataloader

Hi! It will work in a sense that tomosipo will accept torch.float16 inputs, so it can be used inside the `torch.autocast()` context. However, inputs will be internally converted to torch.float32,...

@sem-geologist Although it might not be very useful to HyperSpy, could you still share your ideas on how live/dead time can be calculated from the information contained in .bcf file?...

This shouldn't affect anything for now, but FYI the deprecation was introduced in Slicer 5.6 (see https://github.com/Slicer/Slicer/commit/1cc3b2c62bf5d3836802d3aab4ce37feee6500f0) and originates from https://github.com/fedesemeraro/SlicerTomoSAM/blob/4d968fe6c942c16949e7ff54967bd65821513603/TomoSAM/tomosam.py#L375-L383 The problem is to fix that without breaking compatibility...

Hi! This was actually an oversight on Astra part, which we are planning to revert in Astra 2.3.1. I will try to release that asap, after which this fix should...

Hi @romainvo! Interesting. I would expect iterative solvers to be close or better than FBP, but they can explode (because of overfitting to noise) or produce artifact patterns (because of...

Right, I see. I do suspect that subjectively FBP quality may be better than iterative solver results in many cases. For a general guess though (in `A_dagger`), it feels like...