def test_X_None():
X = None
y = None
from tests.mock_model import MockModel
model = MockModel()
models = [model]
with pytest.raises(TypeError):
eval_models(X, y, models)
def test_y_None():
path = os.path.join(THIS_DIR, os.pardir, 'tests/dummy.txt')
df = read_data(path)
X = df[['value']]
y = None
from tests.mock_model import MockModel
model = MockModel()
models = [model]
with pytest.raises(TypeError):
eval_models(X, y, models)
def test_y_None():
path = os.path.join(THIS_DIR, os.pardir, 'tests/data/dummy.txt')
df = read_data(path)
X = df[['value']]
y = None
model = MockModel()
models = [model]
with pytest.raises(TypeError):
eval_models(X, y, models, label_col_name='is_anomaly')
def test_X_None():
X = None
y = None
model = MockModel()
models = [model]
try:
res = eval_models(X, y, models, label_col_name='is_anomaly')
except TypeError:
assert True
return
assert False
def test_eval_models_all_false():
path = os.path.join(THIS_DIR, os.pardir, 'tests/dummy.txt')
df = read_data(path)
df['is_anomaly'] = 0
from tests.mock_model import MockModel
model = MockModel(prediction=0)
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X, y, models)
assert math.isnan(res['MockModel(prediction=0)']['f1'])
assert math.isnan(res['MockModel(prediction=0)']['precision'])
assert math.isnan(res['MockModel(prediction=0)']['recall'])
def test_n_splits_big():
path = os.path.join(THIS_DIR, os.pardir, 'tests/dummy2.txt')
df = read_data(path)
from tests.mock_model import MockModel
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X, y, models, n_splits=40000, verbose=True)
assert res['MockModel(prediction=1)']['f1'] == 1.0
assert res['MockModel(prediction=1)']['precision'] == 1.0
assert res['MockModel(prediction=1)']['recall'] == 1.0
def test_eval_models_all_false():
path = os.path.join(THIS_DIR, os.pardir, 'tests/data/dummy.txt')
df = read_data(path)
df['is_anomaly'] = 0
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X, y, models, label_col_name='is_anomaly')
assert math.isnan(res['MockModel']['f1'])
assert res['MockModel']['precision'] == 0
assert math.isnan(res['MockModel']['recall'])
def test_eval_models_half_false():
path = os.path.join(THIS_DIR, os.pardir, 'tests/dummy2.txt')
df = read_data(path)
df['is_anomaly'] = 0
df.iloc[-1]['is_anomaly'] = 1
df.iloc[-2]['is_anomaly'] = 1
from tests.mock_model import MockModel
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X, y, models)
assert res['MockModel(prediction=1)']['precision'] == 0.5
assert res['MockModel(prediction=1)']['recall'] == 1.0
def test_eval_models_all_true():
path = os.path.join(THIS_DIR, os.pardir, 'tests/dummy2.txt')
df = read_data(path)
from tests.mock_model import MockModel
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X, y, models)
print(res)
assert res['MockModel(prediction=1)']['f1'] == 1.0
assert res['MockModel(prediction=1)']['precision'] == 1.0
assert res['MockModel(prediction=1)']['recall'] == 1.0
def test_n_splits_big():
path = os.path.join(THIS_DIR, os.pardir, 'tests/data/dummy2.txt')
df = read_data(path)
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models_CV(X,
y,
models,
n_splits=40000,
verbose=True,
label_col_name='is_anomaly')
assert res['MockModel']['f1'] == 1.0
assert res['MockModel']['precision'] == 1.0
assert res['MockModel']['recall'] == 1.0
def test_articles_not_modified_by_predictor(self):
"""
Test if articles fields 'themes' and 'verified_themes' are not modified
by the predictor.
:return:
"""
tokenizer_init_article = Article(
id="0",
title="",
summary="theme1 theme2 theme3",
themes=["theme1", "theme2", "theme3"],
verified_themes=["theme1", "theme2", "theme3"],
predicted_themes=[])
articleOne = Article(id="1",
title="",
summary="theme1 theme2",
themes=["one", "two"],
verified_themes=["one", "two", "three"],
predicted_themes=["three"])
article_tokenizer = ArticleTextTokenizer(
Articles([tokenizer_init_article]), 3)
theme_tokenizer = ArticleThemeTokenizer(
Articles([tokenizer_init_article]))
predictor = ArticlePredictor(
classifier_model=MockModel.get_model(),
supported_themes=["theme1", "theme2", "theme3"],
preprocessor=MockPreprocessor(),
article_tokenizer=article_tokenizer,
theme_tokenizer=theme_tokenizer)
prediction = predictor.predict_preprocessed(
Articles(article=articleOne))
article_with_predictions = prediction.get_articles_with_predictions(
)[0]
self.assertEqual(["one", "two"], article_with_predictions.themes)
self.assertEqual(["one", "two", "three"],
article_with_predictions.verified_themes)
self.assertEqual(["theme1", "theme2"],
article_with_predictions.predicted_themes)
def test_eval_models_half_false():
path = os.path.join(THIS_DIR, os.pardir, 'tests/data/dummy2.txt')
df = read_data(path)
df['is_anomaly'] = 0
df.iloc[-1]['is_anomaly'] = 1
df.iloc[-2]['is_anomaly'] = 1
model = MockModel()
models = [model]
X = df[['value']]
y = df[['is_anomaly']]
res = eval_models(X,
y,
models,
label_col_name='is_anomaly',
window_size_for_metrics=0)
assert res['MockModel']['precision'] == 0.5
assert res['MockModel']['recall'] == 1.0
def build_response_data(req):
"""build_response_data
:param req: request dict
"""
model = MockModel(req=req)
predictions = req.get("test_predictions", [])
sample_predictions = req.get("test_predictions", [])
rounded = req.get("test_predictions", [])
accuracy = req.get("test_accuracy", {"accuracy": 52.5})
error = req.get("test_error", None)
image_file = req.get("image_file", None)
history = req.get("history", None)
histories = req.get("histories", None)
indexes = req.get("test_indexes", None)
scores = req.get("test_scores", None)
cm = req.get("test_cm", None)
predicts_merged = req.get("test_predicts_merged", False)
merge_df = req.get("test_merge_df", None)
data = {
"predictions": predictions,
"rounded_predictions": rounded,
"sample_predictions": sample_predictions,
"acc": accuracy,
"scores": scores,
"history": history,
"histories": histories,
"image_file": image_file,
"model": model,
"indexes": indexes,
"confusion_matrix": cm,
"are_predicts_merged": predicts_merged,
"merge_df": merge_df,
"err": error
}
return data
