Forces test (ORNL#283)

* force tests, which required model arg, and some typo fixing in Lennard Jones

* Add PNAPlus since it uses positions as well

* formatting

examples/LennardJones/LJ.json

@@ -25,6 +25,8 @@
       "int_emb_size": 32,
       "out_emb_size": 16,
       "basis_emb_size": 8,
+      "num_gaussians": 10,
+      "num_filters": 8,
       "num_before_skip": 1,
       "num_after_skip": 1,
       "envelope_exponent": 5,
@@ -55,6 +57,7 @@
     "Training": {
       "num_epoch": 15,
       "batch_size": 64,
+      "perc_train": 0.7,
       "patience": 20,
       "early_stopping": true,
       "Optimizer": {

examples/LennardJones/LJ_data.py

@@ -103,9 +103,9 @@ def __init__(self, dirpath, config, dist=False, sampling=None):
 
         for file in rx:
             filepath = os.path.join(dirpath, file)
-            self.dataset.append(self.transform_inumpyut_to_data_object_base(filepath))
+            self.dataset.append(self.transform_input_to_data_object_base(filepath))
 
-    def transform_inumpyut_to_data_object_base(self, filepath):
+    def transform_input_to_data_object_base(self, filepath):
 
         # Using readline()
         file = open(filepath, "r")

examples/LennardJones/LennardJones.py

@@ -61,6 +61,7 @@
         help="preprocess only (no training)",
     )
     parser.add_argument("--inputfile", help="input file", type=str, default="LJ.json")
+    parser.add_argument("--model_type", help="model type", type=str, default=None)
     parser.add_argument("--mae", action="store_true", help="do mae calculation")
     parser.add_argument("--ddstore", action="store_true", help="ddstore dataset")
     parser.add_argument("--ddstore_width", type=int, help="ddstore width", default=None)
@@ -98,6 +99,11 @@
     # Configurable run choices (JSON file that accompanies this example script).
     with open(input_filename, "r") as f:
         config = json.load(f)
+    config["NeuralNetwork"]["Architecture"]["model_type"] = (
+        args.model_type
+        if args.model_type
+        else config["NeuralNetwork"]["Architecture"]["model_type"]
+    )
     verbosity = config["Verbosity"]["level"]
     config["NeuralNetwork"]["Variables_of_interest"][
         "graph_feature_names"
@@ -159,7 +165,7 @@
         ## This is a local split
         trainset, valset, testset = split_dataset(
             dataset=total,
-            perc_train=0.9,
+            perc_train=config["NeuralNetwork"]["Training"]["perc_train"],
             stratify_splitting=False,
         )
         print("Local splitting: ", len(total), len(trainset), len(valset), len(testset))

tests/test_forces_equivariant.py

@@ -0,0 +1,27 @@
+##############################################################################
+# Copyright (c) 2024, Oak Ridge National Laboratory                          #
+# All rights reserved.                                                       #
+#                                                                            #
+# This file is part of HydraGNN and is distributed under a BSD 3-clause      #
+# license. For the licensing terms see the LICENSE file in the top-level     #
+# directory.                                                                 #
+#                                                                            #
+# SPDX-License-Identifier: BSD-3-Clause                                      #
+##############################################################################
+
+import os
+import pytest
+
+import subprocess
+
+
+@pytest.mark.parametrize("example", ["LennardJones"])
+@pytest.mark.parametrize("model_type", ["SchNet", "EGNN", "DimeNet", "PNAPlus"])
+@pytest.mark.mpi_skip()
+def pytest_examples(example, model_type):
+    path = os.path.join(os.path.dirname(__file__), "..", "examples", example)
+    file_path = os.path.join(path, example + ".py")  # Assuming different model scripts
+    return_code = subprocess.call(["python", file_path, "--model_type", model_type])
+
+    # Check the file ran without error.
+    assert return_code == 0