Remove the experimental in Temp target view

[avouspierre]

Is your feature request related to a problem? Please describe.
The temp target view allows to activate a experimental slider to "simulate" a target in regard of the % basal insulin (HBT) . Now, the implementation is done with Override profil and not useful to stay here

Describe the solution you'd like
remove this part in the view and all code associated in HomeView and others...

Describe alternatives you've considered
no object

Additional context

[bjornoleh]

I agree that the slider is not needed in the Temp target setting if profiles can be used instead. For caregivers, there is still a slight barrier to adopting profiles though, as active profiles/overrides are not displayed in Nightscout. Perhaps we could get #145 sorted first? I don’t know if that is a big or small task.

[avouspierre]

I'm working on #145 with a complete refactoring of the code of override management (a PR is in progress) - not the link with oref but the storage in coredata, respect of MVP pattern... -. But the part of this code (experimental) required to be refactor too... so my idea to remove it.

[bjornoleh]

Ok, that’s great! Knowing a refactor of overrides is in the works, for me at least, feel free to remove that slider!

[mountrcg]

I think, substituting temp targets completely with profile overrides amputates an important oref feature. HBexerciseT is crucial to setting meaningful temp targets.
The current implementation is a little backwards in terms of boundaries of the sliders.
Here is how I had implemented it. With a low TT you can only set lower sensitivity meaning more (>100%) insulin. Boundary is autosensMax. With a highTT you can only set a higher sensitivity meaning less (<100%) insulin.

RPReplay_Final1715693903.mp4

[mountrcg]

There is not just Loop users migrating to Trio, but also AAPS users who are familiar with TT and exercise mode. This gives them a great usability and flexibility for TempTargets.

[bjornoleh]

I don't think it was proposed to remove temp targets as such, only the slider. But there might be a case for keeping it still? Your implementation seems to take target and percent insulin as input, and adjust halfBasal target?

[avouspierre]

I don't want to remove TT but remove the slider and the logic behind. I don't understand the difference between the experimental slider and the override function. For me, it is a duplicate function and override allows to do more options. I know @mountrcg that you don't implement override profil but Trio has the functionality.

[mountrcg]

Yes I am aware that it is not about removing TTs. But adjusting HBT is what makes TTs flexible as it adjusts the sensitivity change impact for any given TT. So the slider lets you do that quite easily. Its valuable for exercise at different intensities. TT only adjusts sensitivity. Overrides adjust a whole lot of things including CR I think.

[mountrcg]

Your implementation seems to take target and percent insulin as input, and adjust halfBasal target?

yes Jons slider adjusted HBT and the target and calculated the sens ratio. That results in unnatural boundaries for HBT as you need that for the slider.
The slider in autoISF calculates hbt with the input of desired target and sens ratio. Both have easily defined boundaries.
What is left to do, is that the calc result replaces the current hbt setting in profile, while at the moment it injects the result into oref overriding the hbt that comes from the profile - not good.

[bjornoleh]

The CR adjustment is optional in profiles. In addition, SMB can be disabled permanently or on schedule for the duration of the profile/override.

[avouspierre]

The difference between temp target with experimental and override is two different algos :
the first will modify HBT before sending to oref. and determinebasal is based on this new HBT profile.

the second did not change hbt but send the "override pourcentage" to oref. oref use the override to recalculate the ISF (and eventually the CR).

So, the result could be different ? Do we have to stay with this 2 algos in Trio ? @MikePlante1 ?

[mountrcg]

Yes indeed 2 different things. TT is short term, especially sports or eating soon stuff where only isf is adjusted, nothing else.
At the moment where you adjust cr, isf, basal etc you are on a not-so-short term adjustment reason. This gets mixed up in the current override implementation because people are used to that from Loop.

[dnzxy]

I‘d vote for fixing the TT slider rather than removing it. There are different types of users and those that use TTs will want to be able to adjust their HBT, the slider is just not a really good implementation to it.

[dnzxy]

The difference between temp target with experimental and override is two different algos

It’s not different algorithms, it’s different inputs that result in different outputs. As long as it’s clearly documented and intuitive to the user what the difference in input and output is (currently I‘d say it’s not), that’s an okay thing to have in the app and would work as expected.

[mountrcg]

refer to mountrcg@7974cc2 as a commit for Trio improving the TT slider.

[github-actions]

hey 👋 - silence for 30 days 🤐 ... anybody? triage is required!

[github-actions]

closed 📴 because silencio 🤫 since an additional 14 days after staleness 📠