def test_second_anomaly_prediction_endpoint_all_columns(
second_base_route,
sensors_str,
influxdb,
gordo_ml_server_client,
sensors,
resp_format,
):
data_to_post = {
"X": np.random.random(size=(10, len(sensors_str))).tolist(),
"y": np.random.random(size=(10, len(sensors_str))).tolist(),
}
endpoint = (
f"{second_base_route}/anomaly/prediction?all_columns=yes&format={resp_format}"
)
resp = gordo_ml_server_client.post(endpoint, json=data_to_post)
assert resp.status_code == 200
if resp_format in (None, "json"):
assert "data" in resp.json
data = server_utils.dataframe_from_dict(resp.json["data"])
else:
data = server_utils.dataframe_from_parquet_bytes(resp.data)
assert "smooth-tag-anomaly-scaled" in data
assert "smooth-tag-anomaly-unscaled" in data
assert "smooth-total-anomaly-scaled" in data
assert "smooth-total-anomaly-unscaled" in data
def test_ml_server_dataframe_to_dict_and_back(sensors_str, use_test_project_tags):
"""
Tests the flow of the server creating a dataframe from the model's data, putting into
a dict of string to df. lists of values, and the client being able to reconstruct it back
to the original dataframe (less the second level names)
"""
# Run test with test project tag names
if use_test_project_tags:
tags = sensors_str
# Run project with random names
else:
tags = [string.ascii_uppercase[i] for i in range(len(sensors_str))]
# Some synthetic data
original_input = np.random.random((10, len(tags)))
model_output = np.random.random((10, len(tags)))
# Convert this data into a dataframe with multi index columns
df = model_utils.make_base_dataframe(tags, original_input, model_output)
# Server then converts this into a dict which maps top level names to lists
serialized = server_utils.dataframe_to_dict(df)
# Client reproduces this dataframe
df_clone = server_utils.dataframe_from_dict(serialized)
# each subset of column under the top level names should be equal
top_lvl_names = df.columns.get_level_values(0)
for top_lvl_name in filter(lambda n: n not in ("start", "end"), top_lvl_names):
assert np.allclose(df[top_lvl_name].values, df_clone[top_lvl_name].values)
def test_dataframe_to_from_dict(expect_multi_lvl: bool, data: dict):
"""
Creating dataframes from various raw data structures should have determined behavior
such as not creating MultiIndex columns with a dict of simple key to array mappings.
"""
df = server_utils.dataframe_from_dict(data)
if expect_multi_lvl:
assert isinstance(df.columns, pd.MultiIndex)
else:
assert not isinstance(df.columns, pd.MultiIndex)
def test_anomaly_prediction_endpoint(
base_route,
sensors_str,
influxdb,
gordo_ml_server_client,
data_size,
sensors,
resp_format,
):
"""
Anomaly GET and POST responses are the same
"""
data_to_post = {
"X": np.random.random(size=(data_size, len(sensors_str))).tolist(),
"y": np.random.random(size=(data_size, len(sensors_str))).tolist(),
}
endpoint = f"{base_route}/anomaly/prediction"
if resp_format is not None:
endpoint += f"?format={resp_format}"
resp = gordo_ml_server_client.post(endpoint, json=data_to_post)
# From here, the response should be (pretty much) the same format from GET or POST
assert resp.status_code == 200
if resp_format in (None, "json"):
assert "data" in resp.json
data = server_utils.dataframe_from_dict(resp.json["data"])
else:
data = server_utils.dataframe_from_parquet_bytes(resp.data)
# Only different between POST and GET is POST will return None for
# start and end dates, because the server can't know what those are
assert "start" in data
assert "end" in data
if data_to_post is not None:
assert np.all(data["start"].isna())
assert np.all(data["end"].isna())
else:
assert not np.any(data["start"].isna())
assert not np.any(data["end"].isna())
assert all(key in data for key in (
"total-anomaly-scaled",
"total-anomaly-unscaled",
"tag-anomaly-scaled",
"tag-anomaly-unscaled",
"model-input",
"model-output",
))
def test_dataframe_from_to_dict(df):
"""
Test (de)serializations back and forth between dataframe -> dict -> dataframe
"""
index_was_datetimes: bool = isinstance(df.index, pd.DatetimeIndex)
cloned = server_utils.dataframe_from_dict(
server_utils.dataframe_to_dict(df))
if index_was_datetimes:
# Ensure the function hasn't mutated the index.
assert isinstance(df.index, pd.DatetimeIndex)
assert np.allclose(df.values, cloned.values)
assert df.columns.tolist() == cloned.columns.tolist()
assert df.index.tolist() == cloned.index.tolist()
def test_prediction_endpoint_post_ok(
base_route,
sensors,
sensors_str,
gordo_ml_server_client,
data_size,
to_dict_arg,
resp_format,
send_as_parquet,
):
"""
Test the expected successful data posts, by sending a variety of valid
JSON formats of a dataframe, as well as parquet serializations.
"""
data_to_post = np.random.random(size=(data_size, len(sensors))).tolist()
if to_dict_arg is not None:
df = pd.DataFrame(data_to_post, columns=sensors_str)
data_to_post = df.to_dict(to_dict_arg)
endpoint = f"{base_route}/prediction"
if resp_format is not None:
endpoint += f"?format={resp_format}"
if send_as_parquet:
X = pd.DataFrame.from_dict(data_to_post)
kwargs = dict(data={
"X": (io.BytesIO(server_utils.dataframe_into_parquet_bytes(X)),
"X")
})
else:
kwargs = dict(json={"X": data_to_post})
resp = gordo_ml_server_client.post(endpoint, **kwargs)
assert resp.status_code == 200
if resp_format in (None, "json"):
data = server_utils.dataframe_from_dict(resp.json["data"])
else:
data = server_utils.dataframe_from_parquet_bytes(resp.data)
# Expected column names
assert all(key in data for key in ("model-output", "model-input"))
def dataframe_from_response(
response: typing.Union[dict, bytes]) -> pd.DataFrame:
"""
The response from the server, parsed as either JSON / dict or raw bytes,
of which would be expected to be loadable from :func:`server.utils.dataframe_from_parquet_bytes`
Parameters
----------
response: Union[dict, bytes]
The parsed response from the ML server.
Returns
-------
pandas.DataFrame
"""
if isinstance(response, dict):
predictions = server_utils.dataframe_from_dict(response["data"])
else:
predictions = server_utils.dataframe_from_parquet_bytes(response)
return predictions
def test_dataframe_from_dict_ordering(index):
"""
We expect that from_dict should order based on the index, and will parse the index
either as datetime or integers and sort in ascending order from there.
"""
df = pd.DataFrame(np.random.random((10, 5)))
df.index = index
original = df.copy()
# What we want
if isinstance(original.index[0], str):
# Parse as datetime or integers if index is string
try:
original.index = original.index.map(dateutil.parser.isoparse)
except ValueError:
original.index = original.index.map(int)
original.sort_index(inplace=True)
# What we get
df_out = server_utils.dataframe_from_dict(
server_utils.dataframe_to_dict(df))
assert np.alltrue(df_out.index == original.index)
assert np.alltrue(df_out.values == original.values)
