create separate threadpool config for GHA runner

.github/workflows/run_pytest.yaml

@@ -39,5 +39,5 @@ jobs:
           pip install git+https://github.com/mir-group/allegro --no-deps
           pip install git+https://github.com/svandenhaute/openmm-ml.git@triclinic
           pip install 'psiflow[test] @ git+https://github.com/molmod/psiflow.git'
-          pytest --skip-gpu
+          pytest --skip-gpu --psiflow-config=.github/workflows/threadpool.py
         shell: micromamba-shell {0}

.github/workflows/threadpool.py

@@ -0,0 +1,67 @@
+from parsl.config import Config
+from parsl.executors import ThreadPoolExecutor
+from parsl.providers import LocalProvider
+
+from psiflow.execution import (
+    Default,
+    ModelEvaluation,
+    ModelTraining,
+    ReferenceEvaluation,
+)
+
+default = Default(
+    parsl_provider=LocalProvider(),  # unused
+)
+model_evaluation = ModelEvaluation(
+    parsl_provider=LocalProvider(),
+    cores_per_worker=1,
+    max_walltime=1,  # timeout after 10 seconds
+    simulation_engine="openmm",
+    gpu=False,
+)
+model_training = ModelTraining(
+    parsl_provider=LocalProvider(),
+    gpu=False,
+    max_walltime=1,
+)
+reference_evaluation = ReferenceEvaluation(
+    parsl_provider=LocalProvider(),
+    cores_per_worker=1,
+    max_walltime=1,
+    mpi_command=lambda x: f"mpirun -np {x}",
+)
+definitions = [
+    default,
+    model_evaluation,
+    model_training,
+    reference_evaluation,
+]
+
+
+def get_config(path_internal):
+    executors = [
+        ThreadPoolExecutor(
+            label="Default",
+            max_threads=default.cores_per_worker,
+            working_dir=str(path_internal),
+        ),
+        ThreadPoolExecutor(
+            label="ModelTraining",
+            max_threads=model_training.cores_per_worker,
+            working_dir=str(path_internal),
+        ),
+        ThreadPoolExecutor(
+            label="ModelEvaluation",
+            max_threads=model_evaluation.cores_per_worker,
+            working_dir=str(path_internal),
+        ),
+        ThreadPoolExecutor(
+            label="ReferenceEvaluation",
+            max_threads=reference_evaluation.cores_per_worker,
+            working_dir=str(path_internal),
+        ),
+    ]
+    config = Config(
+        executors, run_dir=str(path_internal), usage_tracking=False, app_cache=False
+    )
+    return config, definitions