edits.

modelforge/dataset/dataset.py

@@ -146,11 +146,15 @@ def _from_hdf5(self) -> None:
                     for mol in tqdm.tqdm(list(hf.keys())):
                         n_configs = hf[mol]["n_configs"][()]
                         for i in range(n_configs):
+                            temp_data = {}
+                            contains_nan = False
                             for value in self.properties_of_interest:
                                 # if we have a series, we will index into it
                                 if hf[mol][value].attrs["series"]:
-                                    data[value].append(hf[mol][value][i])
+                                    temp_data[value] = hf[mol][value][i]
+                                    if
                                 else:  # if we do not have a series, just append the value
+                                    temp_data[value] = hf[mol][value][()]
                                     data[value].append(hf[mol][value][()])
 
         self.hdf5data = data

modelforge/tests/test_curation.py

@@ -451,6 +451,19 @@ def test_an1_process_download_short(prep_temp_dir):
     ani1_data._process_downloaded(str(local_data_path), hdf5_file)
 
     # ani1_n5.hdf5 datafile includes entries [843, 861, 872, 930, 932] from the full datafile
+    # Example code snippet used to generate this file
+    #
+    # input_file_name = "ani1x-release.h5"
+    #
+    # with h5py.File(input_file_name, "r") as hf_in:
+    #     with h5py.File('ani1_n5.hdf5', "w") as hf_out:
+    #         test_names = list(hf_in.keys())
+    #         test_names2 = []
+    #         for i in [843, 861, 872, 930, 932]:
+    #             test_names2.append(test_names[i])
+    #         for test_name in test_names2:
+    #             hf_in.copy(hf_in[test_name], hf_out)
+
     assert len(ani1_data.data) == 5
 
     assert ani1_data.data[0]["name"] == "C1H4N4O4"