Jordão Bragantini

Hi Gigon, There are also some effort within the python microscopy community to move towards zarr arrays for acessing large images. It would be great to have a faster interface...

Hey @jakirkham. I agree with @gigony, an implementation that produces results equivalent to skimage's watershed is tricky and might not even be that fast. @alanpeixinho might have the GPU implementation...

@netw0rkf10w were you able to improve the performance to get closer to the TF version?

Hi @sofroniewn , I forgot about this PR. I changed to a nested menu as @tlambert03 suggested and I had to change the `qt_main_window import Window` in a few files...

@sofroniewn, I was following @tlambert03 suggestion. The layer-variable nested menu will require additional changes.

I didn't add any other buttons to the widget to be as minimal as possible, but it might not be super intuitive to new users. It closes when the user...

Thanks for the input. I added the Confirm/Cancel/Exit buttons and added the "..." to the action description. Is there an example of how to make the window more napari-like? Like...

> I'm not sure I quite understand - would there be an event emitted when actions are called? That could be an option, from my understanding, the actions' menus are...

> What about 'dialog': my_function where my_function is an annotated function that we can pass to magicgui to autogenerate the dialog? That would be my dream and maybe we try...

Does the `magicgui` call have to be inside the `_qt` directory to keep the headless napari working? I found t[his usage](https://github.com/napari/napari/blob/eea4117525bed8fb11cd0ba13aba4e9a7c83b848/napari/layers/base/base.py) in another sub-directory. It could be something like this...