def evaluate(data_conf, model_conf, **kwargs):
"""Python evaluate method called by AOA framework
Parameters:
data_conf (dict): The dataset metadata
model_conf (dict): The model configuration to use
Returns:
None:No return
"""
predict_df = pd.read_csv(data_conf['location'])
_, test = train_test_split(predict_df, test_size=0.5, random_state=42)
X_predict = test.drop("species", 1)
y_test = test['species']
jnius_configure_classpath()
backend = PyJNIusBackend()
evaluator = make_evaluator(backend, "models/model.pmml") \
.verify()
y_predict = evaluator.evaluateAll(X_predict)
scores = {}
scores['accuracy'] = metrics.accuracy_score(y_test, y_predict['y'])
print("model accuracy is ", scores['accuracy'])
# dump results as json file evaluation.json to models/ folder
with open("models/evaluation.json", "w+") as f:
json.dump(scores, f)
print("Evaluation complete...")
# 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 os
import unittest
import numpy as np
import pandas as pd
import xgboost
from jpmml_evaluator import make_evaluator
from jpmml_evaluator.pyjnius import PyJNIusBackend, jnius_configure_classpath
from sqlflow_submitter.xgboost.train import save_model_to_local_file
# Configure JVM
jnius_configure_classpath()
# Construct a PyJNIus backend
PMML_BACKEND = PyJNIusBackend()
# TODO(sneaxiy): Add XGBRanker unittest. XGBRanker requires group information
# but we do not support group now.
class TestXGBoostModelSavingBase(unittest.TestCase):
def tearDown(self):
filename = self.filename()
if filename is not None and os.path.exists(filename):
os.remove(filename)
os.remove(self.pmml_filename())
def filename(self):
