Proposed solution for the second Part of the issue #186

[PaoloBiolghini]

Proposed a solution for the second Part of the issue #186

As proposed in the comments, i have used m.center.T.. to avoid the possible issues given by the difference between the centers of the maps in the sequence. xshift_arcseconds[i] = new_coordinate.Tx - m.center.Tx yshift_arcseconds[i] = new_coordinate.Ty - m.center.Ty

In order to make the rotation vectors relative to the reference index, before returnig these vectores i subtract the value of the recerence map to all the elements; in this way the rotation of the reference map is 0. xshift_arcseconds=[x-xshift_ref for x in xshift_arcseconds] yshift_arcseconds=[y-yshift_ref for y in yshift_arcseconds]

I have also removed the variable rotate_to_this_layer and created a variable observer_coordinate_ref_layer that stores the observer coordinate because we no longer use rotate_to_this_layer.center.T...

[nabobalis]

Is there a before and after this change example or output you can show us?

[PaoloBiolghini]

Is there a before and after this change example or output you can show us?

I can try to create one if it is helpful

[nabobalis]

I would like to know if this works and what it changes to the output, so I think it would be helpful.

[PaoloBiolghini]

I would like to know if this works and what it changes to the output, so I think it would be helpful.

Hi, I have create a python notebook that show the difference with the previous method https://paolobiolghini.github.io/Sunkit_Image_Examples/showExample.html

[nabobalis]

I would like to know if this works and what it changes to the output, so I think it would be helpful.

Hi, I have create a python notebook that show the difference with the previous method paolobiolghini.github.io/Sunkit_Image_Examples/showExample.html

Thanks for the notebook, but I wasn't clear so I caused confusion.

This change did not affect any of the unit tests and that means we do not have a way to validate this change within our test suite. We can't merge a change like this without some sort of test to ensure that it was broken before, its fixed now and will be fixed in the future.

Could you add a unit test for this?

[PaoloBiolghini]

I would like to know if this works and what it changes to the output, so I think it would be helpful.

Hi, I have create a python notebook that show the difference with the previous method paolobiolghini.github.io/Sunkit_Image_Examples/showExample.html

Thanks for the notebook, but I wasn't clear so I caused confusion.

This change did not affect any of the unit tests and that means we do not have a way to validate this change within our test suite. We can't merge a change like this without some sort of test to ensure that it was broken before, its fixed now and will be fixed in the future.

Could you add a unit test for this?

yes i can try in the next few days

[PaoloBiolghini]

is it possible to use my custom map inside the tests or should i use the one that are already used in the tests?

[nabobalis]

I would prefer if we can use existing ones however, if the existing ones are not good enough to use. Then we can consider adding some resized test data that works

[PaoloBiolghini]

is actually requiring more time than i thought because i have some troubles with pytest

[nabobalis]

That's ok, there is no deadline or rush.

Anything I can try to help with?

[wtbarnes]

Thanks for your efforts here @PaoloBiolghini. Given the complete refactoring of the API that's ongoing in #207, I'm goong to close this since we've now removed this particular alignment method anyway. Thanks again for your efforts and apologies this was stuck in limbo for so long.