Restores Benchmarks

.github/workflows/periodic_benchmarks.yaml

@@ -10,9 +10,9 @@
 # - Publish website
 name: Benchmarks
 on:
-  # Everyday at 12 pm UTC
+  # Every Monday and Thursday at 12 pm UTC
   schedule:
-    - cron: "0 12 * * *"
+    - cron: "0 12 * * 1,4"
   # Make it possible to trigger the
   # workflow manually
   workflow_dispatch:

.github/workflows/scheduled_tests.yaml

@@ -6,10 +6,9 @@ on:
     branches:
       - main
 
-  # runs every day at 09:00 and 15:00 UTC
+  # runs every day at 09:00 UTC
   schedule:
     - cron: '0 9 * * *'
-    - cron: '0 15 * * *'
 
 # Check noxfile.py for associated environment variables
 env:

.pre-commit-config.yaml

@@ -4,7 +4,7 @@ ci:
 
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: "v0.4.5"
+    rev: "v0.4.6"
     hooks:
       - id: ruff
         args: [--fix, --show-fixes]

CHANGELOG.md

@@ -20,6 +20,7 @@
 
 ## Bug Fixes
 
+- [#337](https://github.com/pybop-team/PyBOP/issues/337) - Restores benchmarks, relaxes CI schedule for benchmarks and scheduled tests.
 - [#231](https://github.com/pybop-team/PyBOP/issues/231) - Allows passing of keyword arguments to PyBaMM models and disables build on initialisation.
 - [#321](https://github.com/pybop-team/PyBOP/pull/321) - Improves `integration/test_spm_parameterisation.py` stability, adds flakly pytest plugin, and `test_thevenin_parameterisation.py` integration test.
 - [#330](https://github.com/pybop-team/PyBOP/issues/330) - Fixes implementation of default plotting options.

benchmarks/benchmark_parameterisation.py

@@ -82,12 +82,26 @@ def setup(self, model, parameter_set, optimiser):
         # Create cost function
         cost = pybop.SumSquaredError(problem=problem)
 
-        # Create optimization instance
-        self.optim = pybop.Optimisation(cost, optimiser=optimiser)
+        # Create optimization instance and set options for consistent benchmarking
         if optimiser in [pybop.GradientDescent]:
-            self.optim.optimiser.set_learning_rate(
-                0.008
-            )  # Compromise between stability & performance
+            self.optim = pybop.Optimisation(
+                cost,
+                optimiser=optimiser,
+                max_iterations=250,
+                max_unchanged_iterations=25,
+                threshold=1e-5,
+                min_iterations=2,
+                learning_rate=0.008,  # Compromise between stability & performance
+            )
+        else:
+            self.optim = pybop.Optimisation(
+                cost,
+                optimiser=optimiser,
+                max_iterations=250,
+                max_unchanged_iterations=25,
+                threshold=1e-5,
+                min_iterations=2,
+            )
 
     def time_parameterisation(self, model, parameter_set, optimiser):
         """
@@ -99,10 +113,6 @@ def time_parameterisation(self, model, parameter_set, optimiser):
             parameter_set (str): The name of the parameter set being used (unused).
             optimiser (pybop.Optimiser): The optimizer class being used (unused).
         """
-        # Set optimizer options for consistent benchmarking
-        self.optim.set_max_unchanged_iterations(iterations=25, threshold=1e-5)
-        self.optim.set_max_iterations(250)
-        self.optim.set_min_iterations(2)
         self.optim.run()
 
     def time_optimiser_ask(self, model, parameter_set, optimiser):
@@ -115,4 +125,4 @@ def time_optimiser_ask(self, model, parameter_set, optimiser):
             optimiser (pybop.Optimiser): The optimizer class being used.
         """
         if optimiser not in [pybop.SciPyMinimize, pybop.SciPyDifferentialEvolution]:
-            self.optim.optimiser.ask()
+            self.optim.pints_optimiser.ask()

benchmarks/benchmark_track_parameterisation.py

@@ -82,12 +82,26 @@ def setup(self, model, parameter_set, optimiser):
         # Create cost function
         cost = pybop.SumSquaredError(problem=problem)
 
-        # Create optimization instance
-        self.optim = pybop.Optimisation(cost, optimiser=optimiser)
+        # Create optimization instance and set options for consistent benchmarking
         if optimiser in [pybop.GradientDescent]:
-            self.optim.optimiser.set_learning_rate(
-                0.008
-            )  # Compromise between stability & performance
+            self.optim = pybop.Optimisation(
+                cost,
+                optimiser=optimiser,
+                max_iterations=250,
+                max_unchanged_iterations=25,
+                threshold=1e-5,
+                min_iterations=2,
+                learning_rate=0.008,  # Compromise between stability & performance
+            )
+        else:
+            self.optim = pybop.Optimisation(
+                cost,
+                optimiser=optimiser,
+                max_iterations=250,
+                max_unchanged_iterations=25,
+                threshold=1e-5,
+                min_iterations=2,
+            )
 
         # Track output results
         self.x = self.results_tracking(model, parameter_set, optimiser)
@@ -110,10 +124,5 @@ def results_tracking(self, model, parameter_set, optimiser):
             parameter_set (str): The name of the parameter set being used (unused).
             optimiser (pybop.Optimiser): The optimizer class being used (unused).
         """
-
-        # Set optimizer options for consistent benchmarking
-        self.optim.set_max_unchanged_iterations(iterations=25, threshold=1e-5)
-        self.optim.set_max_iterations(250)
-        self.optim.set_min_iterations(2)
         x, _ = self.optim.run()
         return x

pybop/optimisers/base_optimiser.py

@@ -90,9 +90,13 @@ def __init__(
         self.set_base_options()
         self._set_up_optimiser()
 
-        # Throw an error if any options remain
+        # Throw an warning if any options remain
         if self.unset_options:
-            raise ValueError(f"Unrecognised keyword arguments: {self.unset_options}")
+            warnings.warn(
+                f"Unrecognised keyword arguments: {self.unset_options} will not be used.",
+                UserWarning,
+                stacklevel=2,
+            )
 
     def set_base_options(self):
         """

tests/integration/test_spm_parameterisations.py

@@ -231,14 +231,14 @@ def test_model_misparameterisation(self, parameters, model, init_soc):
         optimiser = pybop.CMAES
 
         # Build the optimisation problem
-        parameterisation = optimiser(cost=cost)
+        optim = optimiser(cost=cost)
+        initial_cost = optim.cost(cost.x0)
 
         # Run the optimisation problem
-        x, final_cost = parameterisation.run()
+        x, final_cost = optim.run()
 
         # Assertion for final_cost
-        with np.testing.assert_raises(AssertionError):
-            np.testing.assert_allclose(final_cost, 0, atol=1e-2)
+        assert initial_cost > final_cost
 
         # Assertion for x
         with np.testing.assert_raises(AssertionError):

tests/unit/test_optimisation.py

@@ -1,3 +1,5 @@
+import warnings
+
 import numpy as np
 import pytest
 
@@ -155,11 +157,12 @@ def test_optimiser_kwargs(self, cost, optimiser):
                 threshold=1e-2,
                 max_evaluations=20,
             )
-            with pytest.raises(
-                ValueError,
-                match="Unrecognised keyword arguments",
+            with pytest.warns(
+                UserWarning,
+                match="Unrecognised keyword arguments: {'unrecognised': 10} will not be used.",
             ):
-                optim = optimiser(cost=cost, tol=1e-3)
+                warnings.simplefilter("always")
+                optim = optimiser(cost=cost, unrecognised=10)
         else:
             # Check bounds in list format and update tol
             bounds = [