Doc: Prettify petab.select docs

pypesto/select/method.py

@@ -2,7 +2,7 @@
 import logging
 from dataclasses import dataclass
 from enum import Enum
-from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+from typing import Any, Callable, Optional, Union
 
 import numpy as np
 import petab_select
@@ -55,7 +55,7 @@ class MethodLogger:
     level:
         The logging level.
     logger:
-        A logger from the `logging` module.
+        A logger from the :mod:`logging` module.
     """
 
     column_width: int = 12
@@ -138,8 +138,7 @@ def get_model_id(model: Model) -> str:
 
             Returns
             -------
-            str
-                The ID.
+            The ID.
             """
             model_subspace_id = model.model_subspace_id or ''
             original_model_id = model.model_id or model.get_hash()
@@ -193,11 +192,11 @@ class MethodCaller:
     petab_select_problem:
         The PEtab Select problem.
     candidate_space:
-        A `petab_select.CandidateSpace`, used to generate candidate models.
+        A :class:`petab_select.CandidateSpace`, used to generate candidate models.
     criterion:
         The criterion by which models will be compared.
     criterion_threshold:
-        The minimum improvement in criterion that a test model must have to
+        The minimum improvement in `criterion` that a test model must have to
         be selected. The comparison is made according to the method. For
         example, in `ForwardSelector`, test models are compared to the
         previously selected model.
@@ -208,7 +207,7 @@ class MethodCaller:
         Limit the number of calibrated models. NB: the number of accepted
         models may (likely) be fewer.
     logger:
-        A `MethodLogger`, used to log results.
+        A :class:`MethodLogger`, used to log results.
     minimize_options:
         A dictionary that will be passed to `pypesto.minimize` as keyword
         arguments for model optimization.
@@ -219,24 +218,24 @@ class MethodCaller:
         objective is initialized, before calibration.
     predecessor_model:
         Specify the predecessor (initial) model for the model selection
-        algorithm. If `None`, then the algorithm will generate an
+        algorithm. If ``None``, then the algorithm will generate an initial
         predecessor model if required.
     select_first_improvement:
-        If `True`, model selection will terminate as soon as a better model
+        If ``True``, model selection will terminate as soon as a better model
         is found. If `False`, all candidate models will be tested.
     startpoint_latest_mle:
-        If `True`, one of the startpoints in the multistart optimization
+        If ``True``, one of the startpoints in the multistart optimization
         will be the MLE of the latest model.
     """
 
     def __init__(
         self,
         petab_select_problem: petab_select.Problem,
-        calibrated_models: Dict[str, Model],
+        calibrated_models: dict[str, Model],
         # Arguments/attributes that can simply take the default value here.
         criterion_threshold: float = 0.0,
         limit: int = np.inf,
-        minimize_options: Dict = None,
+        minimize_options: dict = None,
         model_postprocessor: TYPE_POSTPROCESSOR = None,
         objective_customizer: Callable = None,
         select_first_improvement: bool = False,
@@ -318,15 +317,15 @@ def __init__(
     def __call__(
         self,
         predecessor_model: Optional[Union[Model, None]] = None,
-        newly_calibrated_models: Optional[Dict[str, Model]] = None,
-    ) -> Tuple[List[Model], Dict[str, Model]]:
+        newly_calibrated_models: Optional[dict[str, Model]] = None,
+    ) -> tuple[list[Model], dict[str, Model]]:
         """Run a single iteration of the model selection method.
 
         A single iteration here refers to calibration of all candidate models.
         For example, given a predecessor model with 3 estimated parameters,
         with the forward method, a single iteration would involve calibration
         of all models that have both: the same 3 estimated parameters; and 1
-        additional estimated paramenter.
+        additional estimated parameter.
 
         The input `newly_calibrated_models` is from the previous iteration. The
         output `newly_calibrated_models` is from the current iteration.
@@ -342,12 +341,11 @@ def __call__(
 
         Returns
         -------
-        tuple
-            A 2-tuple, with the following values:
+        A 2-tuple, with the following values:
 
-               1. the predecessor model for the newly calibrated models; and
-               2. the newly calibrated models, as a `dict` where keys are model
-                  hashes and values are models.
+           1. the predecessor model for the newly calibrated models; and
+           2. the newly calibrated models, as a `dict` where keys are model
+              hashes and values are models.
         """
         # All calibrated models in this iteration (see second return value).
         self.logger.new_selection()
@@ -406,8 +404,7 @@ def handle_calibrated_model(
 
         Returns
         -------
-        MethodSignal
-            A `MethodSignal` that describes the result.
+        A :class:`MethodSignal` that describes the result.
         """
         # Use the predecessor model from `__init__` if an iteration-specific
         # predecessor model was not supplied to `__call__`.
@@ -465,9 +462,8 @@ def model1_gt_model0(
 
         Returns
         -------
-        bool
-            `True`, if `model1` is superior to `model0` by the criterion,
-            else `False`.
+        ``True``, if `model1` is superior to `model0` by the criterion,
+        else ``False``.
         """
         if self.criterion in [
             Criterion.AIC,
@@ -509,8 +505,7 @@ def new_model_problem(
 
         Returns
         -------
-        ModelProblem
-            The model selection problem.
+        The model selection problem.
         """
         x_guess = None
         if (

pypesto/select/misc.py

@@ -1,6 +1,6 @@
 """Miscellaneous methods."""
 import logging
-from typing import Dict, Iterable
+from typing import Iterable
 
 import pandas as pd
 import petab
@@ -19,7 +19,7 @@
 def model_to_pypesto_problem(
     model: Model,
     objective: Objective = None,
-    x_guesses: Iterable[Dict[str, float]] = None,
+    x_guesses: Iterable[dict[str, float]] = None,
     hierarchical: bool = False,
 ) -> Problem:
     """Create a pyPESTO problem from a PEtab Select model.
@@ -43,8 +43,7 @@ def model_to_pypesto_problem(
 
     Returns
     -------
-    Problem
-        The pyPESTO select problem.
+    The pyPESTO select problem.
     """
     petab_problem = petab_select.ui.model_to_petab(model=model)[PETAB_PROBLEM]
 
@@ -78,7 +77,7 @@ def model_to_pypesto_problem(
 def model_to_hierarchical_pypesto_problem(*args, **kwargs) -> Problem:
     """Create a hierarchical pyPESTO problem from a PEtab Select model.
 
-    See `model_to_pypesto_problem`.
+    See :func:`model_to_pypesto_problem`.
     """
     pypesto_problem = model_to_pypesto_problem(
         *args,
@@ -89,7 +88,7 @@ def model_to_hierarchical_pypesto_problem(*args, **kwargs) -> Problem:
 
 
 def correct_x_guesses(
-    x_guesses: Iterable[Dict[str, float]],
+    x_guesses: Iterable[dict[str, float]],
     model: Model,
     petab_problem: petab.Problem = None,
     hierarchical: bool = False,

pypesto/select/model_problem.py

@@ -1,6 +1,6 @@
 """Calibrate a PEtab Select model with pyPESTO."""
 import time
-from typing import Any, Callable, Dict, List, Optional
+from typing import Any, Callable, Optional
 
 from petab_select import Criterion, Model
 
@@ -33,7 +33,7 @@ class ModelProblem:
         Keyword argument options that will be passed on to
         :func:`pypesto.optimize.minimize`.
     minimize_result:
-        A pyPESTO result with an optimize result.
+        A pyPESTO result with an `optimize` result.
     model:
         A PEtab Select model.
     model_id:
@@ -61,8 +61,8 @@ def __init__(
         criterion: Criterion,
         valid: bool = True,
         autorun: bool = True,
-        x_guess: List[float] = None,
-        minimize_options: Dict = None,
+        x_guess: list[float] = None,
+        minimize_options: dict = None,
         objective_customizer: Optional[OBJECTIVE_CUSTOMIZER_TYPE] = None,
         postprocessor: Optional["TYPE_POSTPROCESSOR"] = None,
         model_to_pypesto_problem_method: Callable[[Any], Problem] = None,
@@ -141,7 +141,7 @@ def set_result(self, result: Result):
         Parameters
         ----------
         result:
-            A pyPESTO result with an optimize result.
+            A pyPESTO result with an `optimize` result.
         """
         self.minimize_result = result
         # TODO extract best parameter estimates, to use as start point for

pypesto/select/postprocessors.py

@@ -1,6 +1,5 @@
 """Process a model selection :class:`ModelProblem` after calibration."""
 from pathlib import Path
-from typing import List
 
 import matplotlib.pyplot as plt
 import numpy as np
@@ -20,7 +19,7 @@
 
 def multi_postprocessor(
     problem: ModelProblem,
-    postprocessors: List[TYPE_POSTPROCESSOR] = None,
+    postprocessors: list[TYPE_POSTPROCESSOR] = None,
 ):
     """Combine multiple postprocessors into a single postprocessor.
 
@@ -118,7 +117,7 @@ def model_id_binary_postprocessor(problem: ModelProblem):
 def report_postprocessor(
     problem: ModelProblem,
     output_filepath: TYPE_PATH,
-    criteria: List[Criterion] = None,
+    criteria: list[Criterion] = None,
 ):
     """Create a TSV table of model selection results.
 

pypesto/select/problem.py

@@ -1,5 +1,5 @@
 """Manage all components of a pyPESTO model selection problem."""
-from typing import Any, Dict, Iterable, List, Optional, Tuple
+from typing import Any, Iterable, Optional
 
 import petab_select
 from petab_select import Model
@@ -57,8 +57,7 @@ def create_method_caller(self, **kwargs) -> MethodCaller:
 
         Returns
         -------
-        MethodCaller
-            A `MethodCaller` instance.
+        A :class:`MethodCaller` instance.
         """
         return MethodCaller(
             petab_select_problem=self.petab_select_problem,
@@ -69,8 +68,8 @@ def create_method_caller(self, **kwargs) -> MethodCaller:
 
     def set_state(
         self,
-        calibrated_models: Dict[str, Model],
-        newly_calibrated_models: Dict[str, Model],
+        calibrated_models: dict[str, Model],
+        newly_calibrated_models: dict[str, Model],
     ) -> None:
         """Set the state of the problem.
 
@@ -81,7 +80,7 @@ def set_state(
 
     def update_with_newly_calibrated_models(
         self,
-        newly_calibrated_models: Optional[Dict[str, Model]] = None,
+        newly_calibrated_models: Optional[dict[str, Model]] = None,
     ) -> None:
         """Update the state of the problem with newly calibrated models.
 
@@ -94,7 +93,7 @@ def update_with_newly_calibrated_models(
 
     def handle_select_kwargs(
         self,
-        kwargs: Dict[str, Any],
+        kwargs: dict[str, Any],
     ):
         """Check keyword arguments to select calls."""
         if "newly_calibrated_models" in kwargs:
@@ -111,7 +110,7 @@ def handle_select_kwargs(
     def select(
         self,
         **kwargs,
-    ) -> Tuple[Model, Dict[str, Model], Dict[str, Model]]:
+    ) -> tuple[Model, dict[str, Model], dict[str, Model]]:
         """Run a single iteration of a model selection algorithm.
 
         The result is the selected model for the current run, independent of
@@ -121,14 +120,13 @@ def select(
 
         Returns
         -------
-        tuple
-            A 3-tuple, with the following values:
-
-               1. the best model;
-               2. all candidate models in this iteration, as a `dict` with
-                  model hashes as keys and models as values; and
-               3. all candidate models from all iterations, as a `dict` with
-                  model hashes as keys and models as values.
+        A 3-tuple, with the following values:
+
+           1. the best model;
+           2. all candidate models in this iteration, as a `dict` with
+              model hashes as keys and models as values; and
+           3. all candidate models from all iterations, as a `dict` with
+              model hashes as keys and models as values.
         """
         # TODO move some options to PEtab Select? e.g.:
         # - startpoint_latest_mle
@@ -159,7 +157,7 @@ def select(
     def select_to_completion(
         self,
         **kwargs,
-    ) -> List[Model]:
+    ) -> list[Model]:
         """Run an algorithm until an exception `StopIteration` is raised.
 
         ``kwargs`` are passed to the :class:`MethodCaller` constructor.
@@ -170,8 +168,7 @@ def select_to_completion(
 
         Returns
         -------
-        list
-            The best models (the best model at each iteration).
+        The best models (the best model at each iteration).
         """
         best_models = []
         self.handle_select_kwargs(kwargs)
@@ -208,7 +205,7 @@ def multistart_select(
         self,
         predecessor_models: Iterable[Model] = None,
         **kwargs,
-    ) -> Tuple[Model, List[Model]]:
+    ) -> tuple[Model, list[Model]]:
         """Run an algorithm multiple times, with different predecessor models.
 
         Note that the same method caller is currently shared between all calls.
@@ -228,11 +225,10 @@ def multistart_select(
 
         Returns
         -------
-        tuple
-            A 2-tuple, with the following values:
+        A 2-tuple, with the following values:
 
-               1. the best model; and
-               2. the best models (the best model at each iteration).
+           1. the best model; and
+           2. the best models (the best model at each iteration).
         """
         self.handle_select_kwargs(kwargs)
         model_lists = []