def test_pipeline_arguments(self):
with pytest.raises(ArgumentsError):
Pipeline()
with pytest.raises(ArgumentsError):
Pipeline(steps=[custom_read, times_two, squared, split])
with pytest.raises(TypeError):
Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[custom_read, "times_two", squared, split],
params=["hello"],
)
with pytest.raises(TypeError):
Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[custom_read, times_two, squared, split],
params=["hello"],
)
with pytest.raises(TypeError):
Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, squared, split],
params={"col_1": "A"},
custom_reader="custom_read",
)
def test_pipeline_with_custom_reader(self):
df = pd.DataFrame({"A": np.arange(1, 100), "B": np.arange(1, 100)})
_ = df.to_csv("./datasets/configs/dataset.csv", index=False)
params = {
"col_1": "A",
"col_2": "B",
"test_size": 0.2,
"df": "./datasets/configs/dataset.csv",
}
pipeline = Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, squared, split],
params=params,
custom_reader=custom_read,
)
pipeline.process()
assert (pipeline.params["df"].loc[69, "A"] ==
pipeline.params["df_copy"].loc[69, "A"] * 2)
assert (pipeline.params["df"].loc[42, "B"] ==
pipeline.params["df_copy"].loc[42, "B"]**2)
assert len(pipeline.params["X_train"]) == 80
def test_pipeline_with_default_reader(self):
df = pd.DataFrame({"A": np.arange(1, 100), "B": np.arange(1, 100)})
_ = df.to_csv("./datasets/configs/dataset.csv", index=False)
params = {
"col_1": "A",
"col_2": "B",
"test_size": 0.2,
}
pipeline = Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, squared, split],
params=params,
)
pipeline.process()
assert "df" in pipeline.params.keys()
assert "summary" in pipeline.params.keys()
assert "stats" in pipeline.params.keys()
def test_pipeline_arguments(error, df_path, steps, config_file, params,
custom_reader):
with pytest.raises(error):
Pipeline(
df_path=df_path,
steps=steps,
config_file=config_file,
params=params,
custom_reader=custom_reader,
)
def test_remove(self):
df = pd.DataFrame({"A": np.arange(1, 100), "B": np.arange(1, 100)})
_ = df.to_csv("./datasets/configs/dataset.csv", index=False)
params = {
"col_1": "A",
"col_2": "B",
"test_size": 0.2,
}
pipeline = Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, squared, split],
params=params,
)
pipeline.process()
assert len(pipeline.params["X_train"]) == 80
pipeline.remove("split")
pipeline.process()
assert pipeline.params["df"].shape[0] == df.shape[0]
def test_config():
df = pd.DataFrame({"A": np.arange(1, 100), "B": np.arange(1, 100)})
_ = df.to_csv("./datasets/configs/dataset.csv", index=False)
params = {
"df": "./datasets/configs/dataset.csv",
"col_1": "A",
"col_2": "B",
"test_size": 0.2,
}
config_path = "./datasets/configs/pipeline_config.json"
save_config(config_path, params)
pipeline = Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, squared, split],
config_file=config_path,
custom_reader=custom_read,
)
pipeline.process()
assert len(pipeline.params["X_train"]) == 80
pipeline.remove("split")
pipeline.process()
assert (
pipeline.params["df"].shape[0] == pipeline.params["df_copy"].shape[0])
def test_add(self):
df = pd.DataFrame({"A": np.arange(1, 100), "B": np.arange(1, 100)})
_ = df.to_csv("./datasets/configs/dataset.csv", index=False)
params = {
"col_1": "A",
"test_size": 0.2,
}
pipeline = Pipeline(
df_path="./datasets/configs/dataset.csv",
steps=[times_two, split],
params=params,
)
pipeline.process()
assert pipeline.params["df"].loc[42, "A"] == df.loc[42, "A"] * 2
pipeline.add(
squared,
{
"col_2": "A",
},
before="times_two",
)
pipeline.process()
num_0 = pipeline.params["df"].loc[42, "A"]
num_1 = df.loc[42, "A"]
assert num_0 == (num_1**2) * 2
pipeline.remove("squared")
pipeline.add(squared, {"col_2": "A"}, after="read_file")
pipeline.process()
num_0 = pipeline.params["df"].loc[42, "A"]
num_1 = df.loc[42, "A"]
assert num_0 == (num_1**2) * 2