def test_experiment_loadsavemodel(self):
(train, label) = get_X_y(train_file, label_column, sep=',')
(test, label1) = get_X_y(test_file, label_column, sep=',')
cat = OneHotVectorizer() << categorical_columns
ftree = FastTreesBinaryClassifier()
pipeline = Pipeline([cat, ftree])
pipeline.fit(train, label)
metrics1, scores1 = pipeline.test(test,
label1,
'binary',
output_scores=True)
sum1 = metrics1.sum().sum()
(fd, modelfilename) = tempfile.mkstemp(suffix='.model.bin')
fl = os.fdopen(fd, 'w')
fl.close()
pipeline.save_model(modelfilename)
pipeline2 = Pipeline()
pipeline2.load_model(modelfilename)
metrics2, scores2 = pipeline2.test(test,
label1,
'binary',
output_scores=True)
sum2 = metrics2.sum().sum()
assert_equal(sum1, sum2,
"model metrics don't match after loading model")
def test_pipeline_loaded_from_zip_has_feature_contributions(self):
features = ['age', 'education-num', 'hours-per-week']
model_nimbusml = Pipeline(
steps=[FastLinearBinaryClassifier(feature=features)])
model_nimbusml.fit(train, label)
fc = model_nimbusml.get_feature_contributions(test)
# Save the model to zip
model_filename = get_temp_file(suffix='.zip')
model_nimbusml.save_model(model_filename)
# Load the model from zip
model_nimbusml_zip = Pipeline()
model_nimbusml_zip.load_model(model_filename)
fc_zip = model_nimbusml_zip.get_feature_contributions(test)
assert ['FeatureContributions.' + feature in fc_zip.columns
for feature in features]
assert [fc['FeatureContributions.' + feature].equals(
fc_zip['FeatureContributions.' + feature])
for feature in features]
os.remove(model_filename)
def test_model_datastream(self):
model_nimbusml = Pipeline(
steps=[
('cat',
OneHotVectorizer() << categorical_columns),
('linear',
FastLinearBinaryClassifier(
shuffle=False,
number_of_threads=1))])
model_nimbusml.fit(train, label)
# Save with pickle
pickle_filename = get_temp_file(suffix='.p')
with open(pickle_filename, 'wb') as f:
pickle.dump(model_nimbusml, f)
with open(pickle_filename, "rb") as f:
model_nimbusml_pickle = pickle.load(f)
os.remove(pickle_filename)
score1 = model_nimbusml.predict(test).head(5)
score2 = model_nimbusml_pickle.predict(test).head(5)
metrics, score = model_nimbusml.test(test, test_label, output_scores=True)
metrics_pickle, score_pickle = model_nimbusml_pickle.test(
test, test_label, output_scores=True)
assert_almost_equal(score1.sum().sum(), score2.sum().sum(), decimal=2)
assert_almost_equal(
metrics.sum().sum(),
metrics_pickle.sum().sum(),
decimal=2)
# Save load with pipeline methods
model_filename = get_temp_file(suffix='.m')
model_nimbusml.save_model(model_filename)
model_nimbusml_load = Pipeline()
model_nimbusml_load.load_model(model_filename)
score1 = model_nimbusml.predict(test).head(5)
score2 = model_nimbusml_load.predict(test).head(5)
metrics, score = model_nimbusml.test(test, test_label, output_scores=True)
model_nimbusml_load, score_load = model_nimbusml_load.test(
test, test_label, evaltype='binary', output_scores=True)
assert_almost_equal(score1.sum().sum(), score2.sum().sum(), decimal=2)
assert_almost_equal(
metrics.sum().sum(),
model_nimbusml_load.sum().sum(),
decimal=2)
os.remove(model_filename)
def test_pass_predict_proba_from_load_model(selfs):
pipeline = Pipeline([LogisticRegressionBinaryClassifier()])
pipeline.fit(X_train, y_train)
probs1 = pipeline.predict_proba(X_test)
sum1 = probs1.sum().sum()
(fd, modelfilename) = tempfile.mkstemp(suffix='.model.bin')
fl = os.fdopen(fd, 'w')
fl.close()
pipeline.save_model(modelfilename)
pipeline2 = Pipeline()
pipeline2.load_model(modelfilename)
probs2 = pipeline2.predict_proba(X_test)
sum2 = probs2.sum().sum()
assert_equal(sum1, sum2,
"model probabilities don't match after loading model")
def test_model_datastream(self):
model_nimbusml = Pipeline(
steps=[('cat', OneHotVectorizer() << categorical_columns),
('linear',
FastLinearBinaryClassifier(shuffle=False, train_threads=1)
)])
model_nimbusml.fit(train, label)
# Save with pickle
pickle.dump(model_nimbusml, open('nimbusml_model.p', 'wb'))
model_nimbusml_pickle = pickle.load(open("nimbusml_model.p", "rb"))
score1 = model_nimbusml.predict(test).head(5)
score2 = model_nimbusml_pickle.predict(test).head(5)
metrics, score = model_nimbusml.test(test,
test_label,
output_scores=True)
metrics_pickle, score_pickle = model_nimbusml_pickle.test(
test, test_label, output_scores=True)
assert_almost_equal(score1.sum().sum(), score2.sum().sum(), decimal=2)
assert_almost_equal(metrics.sum().sum(),
metrics_pickle.sum().sum(),
decimal=2)
# Save load with pipeline methods
model_nimbusml.save_model('model.nimbusml.m')
model_nimbusml_load = Pipeline()
model_nimbusml_load.load_model('model.nimbusml.m')
score1 = model_nimbusml.predict(test).head(5)
score2 = model_nimbusml_load.predict(test).head(5)
metrics, score = model_nimbusml.test(test,
test_label,
output_scores=True)
model_nimbusml_load, score_load = model_nimbusml_load.test(
test, test_label, evaltype='binary', output_scores=True)
assert_almost_equal(score1.sum().sum(), score2.sum().sum(), decimal=2)
assert_almost_equal(metrics.sum().sum(),
model_nimbusml_load.sum().sum(),
decimal=2)
