def test_print_available_datasets(self):
# Go over all parameter combinations,
# for each check that returned dataset is correct
for sort_method in ['alphabetical', 'num_entries']:
datasets = get_available_datasets(sort_method=sort_method)
if sort_method == 'alphabetical':
self.assertEqual(datasets, sorted(self.dataset_names))
else:
self.assertEqual(
datasets,
sorted(self.dataset_names,
key=lambda x: self.dataset_dict[x]['num_entries'],
reverse=True))
def test_get_available_datasets(self):
# Go over all parameter combinations,
# for each check that returned dataset is correct
for parameter_combo in product([True, False], [True, False],
['alphabetical', 'num_entries']):
datasets = get_available_datasets(*parameter_combo)
if parameter_combo[2] == 'alphabetical':
self.assertEqual(datasets, sorted(self.dataset_names))
else:
self.assertEqual(
datasets,
sorted(self.dataset_names,
key=lambda x: self.dataset_dict[x]['num_entries'],
reverse=True))
def test_get_available_datasets(self):
# Go over all parameter combinations,
# for each check that returned dataset is correct
for parameter_combo in product([True, False], [True, False],
['alphabetical', 'num_entries']):
datasets = get_available_datasets(*parameter_combo)
if parameter_combo[2] == 'alphabetical':
self.assertEqual(datasets, sorted(self.dataset_names))
else:
self.assertEqual(
datasets,
sorted(self.dataset_names,
key=lambda x: self.dataset_dict[x]['num_entries'],
reverse=True)
)
def get_available_datasets(self):
datasets = get_available_datasets()
return
from matminer.datasets.dataset_retrieval import load_dataset, get_available_datasets, get_all_dataset_info
datasets = get_available_datasets(print_format=None)
for dataset in datasets:
if "matbench_" in dataset:
df = load_dataset(dataset)
target_col = [col for col in df.columns if col not in ["structure", "composition"]][0]
print(f" * - :code:`{dataset}`\n - :code:`{target_col}`\n - {df.shape[0]}")
# print(get_all_dataset_info("matbench_steels"))
2. composition features are actually desired. (deduced from whether
composition featurizers are present in self.featurizers).
Args:
df (pandas.DataFrame): May or may not contain composition column.
Returns:
df (pandas.DataFrame): Contains composition column if desired
"""
if "structure" in df.columns and "composition" not in df.columns:
if self.auto_featurizer or (set(_composition_aliases)
& set(self.featurizers.keys())):
df = self._tidy_column(df, "structure")
struct2comp = StructureToComposition(
target_col_id="composition", overwrite_data=False)
df = struct2comp.featurize_dataframe(df, "structure")
self.logger.debug("Adding compositions from structures.")
return df
if __name__ == "__main__":
from matminer.datasets.dataset_retrieval import load_dataset, get_available_datasets
print(get_available_datasets())
# df = load_dataset("steel_strength").rename(columns={"formula": "composition"})[["yield strength", "composition"]]
# af = AutoFeaturizer()
# print(df)
# df = af.fit_transform(df, "yield strength")
from pymatgen import Structure
# s = Structure()
# s.