Model in RIVET run fails to match the data

[mmikhasenko]

Here is a picture from Ilya's run

The problem seems with Flatte fo L1405.
That is the one that we've just fixed in json file.

Story of failing tests in ThreeBodyDecaysIO.jl

@IlyaSegal used an updated model from ThreeBodyDecaysIO.jl.
The code is updated. All tests are passing on my machine and in CI.

• The PR that modifies the json https://github.com/mmikhasenko/ThreeBodyDecaysIO.jl/pull/12/files
• PR for amplitude-serialization

The core modification is MultichannelBW in HadronicLineshapes.jl.
The change has happen long time ago. I think @H178561, your implementation was after the change, should be correct.

[mmikhasenko]

mentioned isegal by mistake in the comments, it should be IlyaSegal.

[H178561]

I don’t really get it. So is the error in my implementation or is the json file wrong?

[IlyaSegal]

I used your up-to-date version of ThreeBodyDecays, but an updated version of the .json file with the model from amplitude-serialization.

I think the first check would be to cross-check Julia and your implementation with the updated .json file.

[H178561]

Should work now, I changed the Flatte to match the updated Julia Code. I will create a Test this week but I just haven't found the time yet.

[mmikhasenko]

Hi Henri, could you please explain

• what you have changed,
• which Julia code changes you refer to,
• and why you think that the problem is solved?

[H178561]

I wrote my comment wrong. I had an error in my Flatte Lineshape function,
e1ddda9

Now I checked the lineshapes and amplitudes from the updated json file in Julia and c++ and they match.