Phase Retrieval for Coherent Diffractive Imaging: Theory and Algorithm

[@enlu6002]

It explains the algorithm clearly. I'm new in Ptychography. It is very useful for me to get started.Many thanks.

[@sahilsharma9385]

thank you for explaining these algorithms. I understood them very clearly!

[@raihan0606175]

What a wonderful presentation. Thank you so much.

[@HelloWorlds__JTS]

Mostly a note to my-future-self here, since I again got stuck here pondering this: Parseval's theorem at 27:38 is not meant to imply pointwise identification, but just that the total energy is the same in either representation. Also, the ||...|| notation used here implies a norm of some sort, but this is irrelevant because these are functions.

[@jeremiesefo9487]

Incredibly insightful. Thanks a lot. I just got worried if for the sake of the existence of the projection, we must make sure that the constraint domains onto which we project stay closed and convex. Or is there a way around to handle non convex supports for instance? Do they cause issues in practice? Thanks again

[@jingtaodong7972]

So does the combination of HIO and ER mean that you implement 200 iterations of HIO at first place and use its output as the input of the next 200 interations of ER, and do like this for several loops?

[@frodo279]

Can we get the codes please?

[@shwetajoglekar6631]

Getting a key error for supp after running the same code