def test_hyper_param_populate(driver):
driver.get('/')
with create_test_project() as p, create_test_featureset(
p) as fs, create_test_model(fs) as m:
driver.refresh()
proj_select = Select(
driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-6').click()
driver.find_element_by_partial_link_text('Create New Model').click()
driver.find_element_by_partial_link_text(
'Choose Model Parameters').click()
model_select = Select(
driver.find_element_by_css_selector('[name=modelType]'))
model_select.select_by_visible_text("RandomForestClassifier")
driver.implicitly_wait(0.1)
driver.find_element_by_xpath(
"//label[contains(text(),'n_estimators')]")
driver.find_element_by_xpath(
"//label[contains(text(),'max_features')]")
model_select.select_by_visible_text("RidgeClassifierCV")
driver.implicitly_wait(0.1)
driver.find_element_by_xpath("//label[contains(text(),'alphas')]")
driver.find_element_by_xpath("//label[contains(text(),'scoring')]")
model_select.select_by_visible_text("BayesianRidgeRegressor")
driver.implicitly_wait(0.1)
driver.find_element_by_xpath("//label[contains(text(),'n_iter')]")
driver.find_element_by_xpath("//label[contains(text(),'alpha_1')]")
def test_delete_model(driver):
driver.get('/')
with create_test_project() as p, create_test_featureset(p) as fs,\
create_test_model(fs) as m:
driver.refresh()
proj_select = Select(driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-6').click()
driver.find_element_by_partial_link_text('Delete').click()
driver.implicitly_wait(1)
status_td = driver.find_element_by_xpath(
"//div[contains(text(),'Model deleted')]")
def test_predict_raw_data():
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs, create_test_model(fs) as m:
ts_data = [[1, 2, 3, 4], [32.2, 53.3, 32.3, 32.52], [0.2, 0.3, 0.6, 0.3]]
impute_kwargs = {'strategy': 'constant', 'value': None}
query_string = ('{}/predict_raw_data?ts_data={}'
'&modelID={}&impute_kwargs={}').format(
cfg['server']['url'], json.dumps(ts_data),
json.dumps(m.id), json.dumps(impute_kwargs))
response = requests.post(query_string).json()
assert response['status'] == 'success'
assert response['data']['0']['features']['total_time'] == 3.0
assert 'Mira' in response['data']['0']['prediction']
def test_prediction_to_csv_class():
"""Test util.prediction_to_csv"""
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs,\
create_test_model(fs, model_type='LinearSGDClassifier') as m,\
create_test_prediction(ds, m) as pred:
pred = featureset.from_netcdf(pred.file.uri)
assert util.prediction_to_csv(pred) ==\
[['ts_name', 'true_target', 'prediction'],
['0', 'Mira', 'Mira'],
['1', 'Classical_Cepheid', 'Classical_Cepheid'],
['2', 'Mira', 'Mira'],
['3', 'Classical_Cepheid', 'Classical_Cepheid'],
['4', 'Mira', 'Mira']]
def test_model_info_display(driver):
with create_test_project() as p, create_test_featureset(p) as fs,\
create_test_model(fs) as m:
driver.refresh()
proj_select = Select(driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-6').click()
driver.find_element_by_xpath("//td[contains(text(),'{}')]".format(m.name)).click()
assert driver.find_element_by_xpath("//th[contains(text(),'Model Type')]")\
.is_displayed()
assert driver.find_element_by_xpath("//th[contains(text(),'Hyper "
"Parameters')]").is_displayed()
assert driver.find_element_by_xpath("//th[contains(text(),'Training "
"Data Score')]").is_displayed()
def test_prediction_to_csv_regr():
"""Test util.prediction_to_csv"""
with create_test_project() as p, create_test_dataset(p, label_type='regr') as ds,\
create_test_featureset(p, label_type='regr') as fs,\
create_test_model(fs, model_type='LinearRegressor') as m,\
create_test_prediction(ds, m) as pred:
pred = xr.open_dataset(pred.file.uri)
results = util.prediction_to_csv(pred)
assert results[0] == ['ts_name', 'true_target', 'prediction']
npt.assert_array_almost_equal(
[[float(e) for e in row] for row in results[1:]],
[[0, 2.2, 2.2], [1, 3.4, 3.4], [2, 4.4, 4.4], [3, 2.2, 2.2],
[4, 3.1, 3.1]])
def test_plot_features(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs:
driver.refresh()
proj_select = Select(driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-4').click()
driver.implicitly_wait(1)
driver.find_element_by_xpath("//td[contains(text(),'{}')]".format(fs.name)).click()
driver.implicitly_wait(1)
driver.find_element_by_xpath("//b[contains(text(),'Please wait while we load your plotting data...')]")
driver.implicitly_wait(3)
driver.find_element_by_css_selector("[class=svg-container]")
def test_pred_results_table_lsgdc(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs,\
create_test_model(fs, model_type='LinearSGDClassifier') as m,\
create_test_prediction(ds, m):
_click_prediction_row(p.id, driver)
try:
rows = _grab_pred_results_table_rows(driver, 'Mira')
rows = [row.text for row in rows]
npt.assert_array_equal([
'0 Mira Mira', '1 Classical_Cepheid Classical_Cepheid',
'2 Mira Mira', '3 Classical_Cepheid Classical_Cepheid',
'4 Mira Mira'
], rows)
except:
driver.save_screenshot("/tmp/pred_click_tr_fail.png")
raise
def test_delete_prediction(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs, create_test_model(fs) as m,\
create_test_prediction(ds, m):
driver.refresh()
proj_select = Select(
driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-8').click()
driver.implicitly_wait(1)
driver.find_element_by_xpath(
"//td[contains(text(),'Completed')]").click()
time.sleep(0.2)
driver.find_element_by_partial_link_text('Delete').click()
driver.implicitly_wait(1)
status_td = driver.find_element_by_xpath(
"//div[contains(text(),'Prediction deleted')]")
def test_pred_results_table_regr(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p, label_type='regr') as fs,\
create_test_model(fs, model_type='LinearRegressor') as m,\
create_test_prediction(ds, m):
_click_prediction_row(p.id, driver)
try:
rows = _grab_pred_results_table_rows(driver, '3.4')
rows = [[float(el) for el in row.text.split(' ')] for row in rows]
npt.assert_array_almost_equal(
[[0.0, 2.2, 2.2000000000000006
], [1.0, 3.4, 3.4000000000000004],
[2.0, 4.4, 4.400000000000001], [3.0, 2.2, 2.1999999999999993],
[4.0, 3.1, 3.099999999999999]], rows)
except:
driver.save_screenshot("/tmp/pred_click_tr_fail.png")
raise
def test_download_prediction_csv_class(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs,\
create_test_model(fs, model_type='LinearSGDClassifier') as m,\
create_test_prediction(ds, m):
_click_download(p.id, driver)
assert os.path.exists('/tmp/cesium_prediction_results.csv')
try:
npt.assert_equal(
np.genfromtxt('/tmp/cesium_prediction_results.csv',
dtype='str'),
[
'ts_name,true_target,prediction', '0,Mira,Mira',
'1,Classical_Cepheid,Classical_Cepheid', '2,Mira,Mira',
'3,Classical_Cepheid,Classical_Cepheid', '4,Mira,Mira'
])
finally:
os.remove('/tmp/cesium_prediction_results.csv')
def test_cannot_build_model_unlabeled_data(driver):
driver.get('/')
with create_test_project() as p,\
create_test_featureset(p, label_type='none') as fs:
driver.refresh()
proj_select = Select(driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-6').click()
driver.find_element_by_partial_link_text('Create New Model').click()
model_name = driver.find_element_by_css_selector('[name=modelName]')
model_name.send_keys(str(uuid.uuid4()))
driver.find_element_by_class_name('btn-primary').click()
driver.implicitly_wait(2)
driver.find_element_by_xpath(
"//div[contains(.,'Cannot build model for unlabeled feature set.')]")
def test_pred_results_table_rfc(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs, create_test_model(fs) as m,\
create_test_prediction(ds, m):
_click_prediction_row(p.id, driver)
try:
rows = _grab_pred_results_table_rows(driver, 'Mira')
for row in rows:
probs = [
float(v.text)
for v in row.find_elements_by_tag_name('td')[3::2]
]
assert sorted(probs, reverse=True) == probs
driver.find_element_by_xpath(
"//th[contains(text(),'Time Series')]")
except:
driver.save_screenshot("/tmp/pred_click_tr_fail.png")
raise
def test_prediction_to_csv_regr():
"""Test util.prediction_to_csv"""
with create_test_project() as p, create_test_dataset(p, label_type='regr') as ds,\
create_test_featureset(p, label_type='regr') as fs,\
create_test_model(fs, model_type='LinearRegressor') as m,\
create_test_prediction(ds, m) as pred:
pred = featureset.from_netcdf(pred.file.uri)
results = util.prediction_to_csv(pred)
assert results[0] == ['ts_name', 'true_target', 'prediction']
npt.assert_array_almost_equal(
[[float(e) for e in row] for row in results[1:]],
[[0, 2.2, 2.2],
[1, 3.4, 3.4],
[2, 4.4, 4.4],
[3, 2.2, 2.2],
[4, 3.1, 3.1]])
def test_delete_featureset(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs:
driver.refresh()
proj_select = Select(driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-4').click()
driver.find_element_by_partial_link_text('Delete').click()
driver.implicitly_wait(1)
status_td = driver.find_element_by_xpath(
"//div[contains(text(),'Feature set deleted')]")
try:
el = driver.find_element_by_xpath(
"//td[contains(text(),'{}')]".format(test_featureset_name))
except NoSuchElementException:
pass
else:
raise Exception("Featureset still present in table after delete.")
def test_download_prediction_csv_regr(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p, label_type='regr') as ds,\
create_test_featureset(p, label_type='regr') as fs,\
create_test_model(fs, model_type='LinearRegressor') as m,\
create_test_prediction(ds, m):
_click_download(p.id, driver)
assert os.path.exists('/tmp/cesium_prediction_results.csv')
try:
results = np.genfromtxt('/tmp/cesium_prediction_results.csv',
dtype='str',
delimiter=',')
npt.assert_equal(results[0],
['ts_name', 'true_target', 'prediction'])
npt.assert_array_almost_equal(
[[float(e) for e in row] for row in results[1:]],
[[0, 2.2, 2.2], [1, 3.4, 3.4], [2, 4.4, 4.4], [3, 2.2, 2.2],
[4, 3.1, 3.1]])
finally:
os.remove('/tmp/cesium_prediction_results.csv')
def test_delete_featureset(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs:
driver.refresh()
proj_select = Select(
driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-4').click()
driver.find_element_by_partial_link_text('Delete').click()
driver.implicitly_wait(1)
status_td = driver.find_element_by_xpath(
"//div[contains(text(),'Feature set deleted')]")
try:
el = driver.find_element_by_xpath(
"//td[contains(text(),'{}')]".format(test_featureset_name))
except NoSuchElementException:
pass
else:
raise Exception("Featureset still present in table after delete.")
def test_plot_features(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs:
driver.refresh()
proj_select = Select(
driver.find_element_by_css_selector('[name=project]'))
proj_select.select_by_value(str(p.id))
driver.find_element_by_id('react-tabs-4').click()
driver.implicitly_wait(1)
driver.find_element_by_xpath("//td[contains(text(),'{}')]".format(
fs.name)).click()
driver.implicitly_wait(1)
driver.find_element_by_xpath(
"//b[contains(text(),'Please wait while we load your plotting data...')]"
)
driver.implicitly_wait(3)
driver.find_element_by_css_selector("[class=svg-container]")
def test_add_prediction_rfr(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p, label_type='regr') as ds,\
create_test_featureset(p, label_type='regr') as fs,\
create_test_model(fs, model_type='RandomForestRegressor') as m:
_add_prediction(p.id, driver)
def test_build_model_lr(driver):
driver.get('/')
with create_test_project() as p,\
create_test_featureset(p, label_type='regr') as fs:
_build_model(p.id, 'LinearRegressor', driver)
def test_build_model_lsgdc(driver):
driver.get('/')
with create_test_project() as p, create_test_featureset(p) as fs:
_build_model(p.id, 'LinearSGDClassifier', driver)
def test_build_model_rfc(driver):
driver.get('/')
with create_test_project() as p, create_test_featureset(p) as fs:
_build_model(p.id, 'RandomForestClassifier', driver)
def test_add_prediction_rfc(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs, create_test_model(fs) as m:
_add_prediction(p.id, driver)
def test_add_prediction_lsgdc(driver):
driver.get('/')
with create_test_project() as p, create_test_dataset(p) as ds,\
create_test_featureset(p) as fs,\
create_test_model(fs, model_type='LinearSGDClassifier') as m:
_add_prediction(p.id, driver)