Carter Francis

This is an interesting bug... The way that the `get_integrated_intensity` function works is that it just calls the roi on the data and then sums the data. You can repeat...

As to the function working differently with different file types that suggests that maybe your axes managers are not equal for the `MIB` and `.zspy` files? You could also try...

@garrofederico Did you ever figure out what the issue was here?

@ericpre I didn't add in `ColumnAxis` because I think it should be a separate class and not an axis. Maybe ColumnLabel is a better name for it. Even though it...

+1 for this idea as well. This would also allow for plotting signal and navigation axes in two columns rather in two rows which can be frustrating when using the...

> Should it be a workflow/data management thing? Are you saying this is outside of the scope of hyperspy? Possibly, this might be overkill and could introduce some bad habits....

> When I looked into this, not all functions/libraries behaved nicely like you describe for `np.fft.fft`. For example, scipy functions explicitly expect numpy arrays and will fail with cupy arrays;...

@ericpre There are a couple of libraries that would benefit from hyperspy/hyperspy#2876. I'd like for the `find_peaks` method to just return a vector signal. As written there shouldn't be many...

It might also be a good idea to formalize how Hyperspy handles parallelization. At the very least I think that this involves more documentation about how to use `dask` as...

> Major releases force people to update their codes and scripts, so we should minimize their number. Therefore, I think that it would be good to split the TEM features...