# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import inspect
import cuml
import pytest
import numpydoc.docscrape
from cuml.test.utils import (get_classes_from_package,
small_classification_dataset)
all_base_children = get_classes_from_package(cuml, import_sub_packages=True)
def test_base_class_usage():
# Ensure base class returns the 3 main properties needed by all classes
base = cuml.Base()
base.handle.sync()
base_params = base.get_param_names()
assert "handle" in base_params
assert "verbose" in base_params
assert "output_type" in base_params
del base
umap_model = {"UMAP": cuml.UMAP}
rf_module = ClassEnumerator(module=cuml.ensemble)
rf_models = rf_module.get_models()
k_neighbors_config = ClassEnumerator(
module=cuml.neighbors, exclude_classes=[cuml.neighbors.NearestNeighbors])
k_neighbors_models = k_neighbors_config.get_models()
unfit_pickle_xfail = ['ARIMA', 'KalmanFilter', 'ForestInference']
unfit_clone_xfail = [
'ARIMA', 'ExponentialSmoothing', 'KalmanFilter', 'MBSGDClassifier',
'MBSGDRegressor'
]
all_models = get_classes_from_package(cuml)
all_models.update({
**regression_models,
**solver_models,
**cluster_models,
**decomposition_models,
**decomposition_models_xfail,
**neighbor_models,
**dbscan_model,
**umap_model,
**rf_models,
**k_neighbors_models,
'ARIMA':
lambda: ARIMA((1, 1, 1), np.array([-217.72, -206.77]), [
np.array([0.03]), np.array([-0.03])
], [np.array([-0.99]), np.array([-0.99])],
'ARIMA', 'AutoARIMA', 'KalmanFilter', 'BaseRandomForestModel',
'ForestInference', 'MulticlassClassifier', 'OneVsOneClassifier',
'OneVsRestClassifier'
]
unfit_clone_xfail = [
'AutoARIMA',
"ARIMA",
"BaseRandomForestModel",
"GaussianRandomProjection",
'MulticlassClassifier',
'OneVsOneClassifier',
'OneVsRestClassifier',
"SparseRandomProjection",
]
all_models = get_classes_from_package(cuml, import_sub_packages=True)
all_models.update({
**regression_models,
**solver_models,
**cluster_models,
**decomposition_models,
**decomposition_models_xfail,
**neighbor_models,
**dbscan_model,
**agglomerative_model,
**umap_model,
**rf_models,
**k_neighbors_models,
'ARIMA':
lambda: ARIMA(np.random.normal(0.0, 1.0, (10, ))),
'ExponentialSmoothing':