Jakob Vanhoefer

Use Literal consistently to annotate type hint, where string arguments come from a predefined input range.

1

**Feature description** Since python 3.8 one can add a type hint of the form `argument: Literal['a', 'b', 'c']`, if a string argument comes from a predefined range of inputs. We...

Refactor sample result

2

The sample result was never used, while the `McmcPtResult` did not ensure a stable interface to the outside world. Fixed that. Might not be the final solution to the problem,...

## Problem/Bug description The scipy optimizer calls `fun` and `grad` via `object` via `objective.get_fun` and `objective.get_grad`, see https://github.com/ICB-DCM/pyPESTO/blob/4c30abfca56ba714c302141cd44a9dd366bff4bb/pypesto/optimize/optimizer.py#L420 and for the function/gradient definition see (and line below) https://github.com/ICB-DCM/pyPESTO/blob/4c30abfca56ba714c302141cd44a9dd366bff4bb/pypesto/optimize/optimizer.py#L399 The `get_grad`...

`__repr__` method for relevant classes

1

**Feature description** `print(problem/result/...)` gives no meaningful output. **Motivation/Application** Nicer syntax, everyone likes that...

Feature separate likelihood and prior

1

- [x] add `get_negloglikelihood` and `get_neglogposterior` in all the different objectives. - [x] Go through samplers, and adapt them, if null & prior are needed separately - [x] remove `problem.x_priors`...

Consolidate different gradient checks

8

**Feature description** Currently we have several different (rather redundant) gradient checking routines, which only have slightly different functionalities, which in my opinion do not all justify being there in parallel....

the negative log prior should be `inf` outside the parameter bounds

If e.g. you perform `visualize.profiles` without computing profiles before, you get a non-informative Error message. Would be good to test for similar instances in order to make it easier for...

Automatic generation of PEtab visualization tables from fitting results

2

I think, it would be great, to have a function, that takes the result of a parameter optimization run from a _PEtab-problem_ and creates a _PEtab visualization table_ (+necessarily a...

I would imagine a function, that can be called like `profile.get_profile_ci(result, alpha, x_ids=None)`, that returns the confidence intervals from a profile run... This function should then also be used in...