Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix column_name and null generation #337

Merged
merged 4 commits into from
Sep 27, 2023
Merged

Fix column_name and null generation #337

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
e1dcd8f
fix for column name exclusion bug
ksneab7 Sep 27, 2023
c1bfa3d
pytest fixes
ksneab7 Sep 27, 2023
8d65506
Update tests/test_generators.py
ksneab7 Sep 27, 2023
a2ab09b
added null generation
ksneab7 Sep 27, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 14 additions & 0 deletions synthetic_data/distinct_generators/null_generator.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
"""Contains a random Null generator."""
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This function is used to generate fully null columns

import numpy as np


def null_generation(num_rows: int = 1) -> np.array:
"""
Randomly generates an array of integers between the given min and max values.

:param num_rows: the number of rows in np array generated
:type num_rows: int, optional

:return: np array of null values
"""
return np.array([None] * num_rows)
77 changes: 40 additions & 37 deletions synthetic_data/generators.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@
from synthetic_data.distinct_generators.datetime_generator import random_datetimes
from synthetic_data.distinct_generators.float_generator import random_floats
from synthetic_data.distinct_generators.int_generator import random_integers
from synthetic_data.distinct_generators.null_generator import null_generation
from synthetic_data.distinct_generators.text_generator import random_text
from synthetic_data.graph_synthetic_data import GraphDataGenerator
from synthetic_data.synthetic_data import make_data_from_report
Expand Down Expand Up @@ -42,6 +43,7 @@ def __init__(
"datetime": random_datetimes,
"string": random_text,
"text": random_text,
"null_generator": null_generation,
}

@classmethod
Expand Down Expand Up @@ -105,49 +107,46 @@ def _generate_uncorrelated_column_data(self, num_samples):
col_ = copy.deepcopy(col)

generator_name = col_.get("data_type", None)

if not generator_name:
logging.warning(
f"Generator of type {generator_name} is not implemented."
)
continue
column_header = col_.get("column_name", None)

col_["rng"] = self.rng
col_["num_rows"] = num_samples
if generator_name:
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is needed in place of the removed warning, as we account for a fully nulled column

if generator_name in ["string", "text"]:
if col_.get("categorical", False):
generator_name = "categorical"
total = 0
for count in col["statistics"]["categorical_count"].values():
total += count

if generator_name in ["string", "text"]:
if col_.get("categorical", False):
generator_name = "categorical"
total = 0
for count in col["statistics"]["categorical_count"].values():
total += count

probabilities = []
for count in col["statistics"]["categorical_count"].values():
probabilities.append(count / total)
probabilities = []
for count in col["statistics"]["categorical_count"].values():
probabilities.append(count / total)

col_["probabilities"] = probabilities
col_["categories"] = col_["statistics"].get("categories", None)
col_["probabilities"] = probabilities
col_["categories"] = col_["statistics"].get("categories", None)

col_["vocab"] = col_["statistics"].get("vocab", None)
col_["vocab"] = col_["statistics"].get("vocab", None)

col_["min"] = col_["statistics"].get("min", None)
col_["max"] = col_["statistics"].get("max", None)
col_["min"] = col_["statistics"].get("min", None)
col_["max"] = col_["statistics"].get("max", None)

# edge cases for extracting data from profiler report.
if generator_name == "datetime":
col_["format"] = col_["statistics"].get("format", None)
col_["min"] = pd.to_datetime(
col_["statistics"].get("min", None), format=col_["format"][0]
)
col_["max"] = pd.to_datetime(
col_["statistics"].get("max", None), format=col_["format"][0]
)
# edge cases for extracting data from profiler report.
if generator_name == "datetime":
col_["format"] = col_["statistics"].get("format", None)
col_["min"] = pd.to_datetime(
col_["statistics"].get("min", None), format=col_["format"][0]
)
col_["max"] = pd.to_datetime(
col_["statistics"].get("max", None), format=col_["format"][0]
)

if generator_name == "float":
col_["precision"] = int(
col_["statistics"].get("precision", None).get("max", None)
)
if generator_name == "float":
col_["precision"] = int(
col_["statistics"].get("precision", None).get("max", None)
)
elif not generator_name:
generator_name = "null_generator"

generator_func = self.gen_funcs.get(generator_name, None)
params_gen_funcs = inspect.signature(generator_func)
Expand All @@ -157,7 +156,9 @@ def _generate_uncorrelated_column_data(self, num_samples):
param_build[param[0]] = col_[param[0]]

generated_data = generator_func(**param_build)
if col_["order"] in sorting_types:
if (not generator_name == "null_generator") and col_[
"order"
] in sorting_types:
dataset.append(
self.get_ordered_column(
generated_data,
Expand All @@ -166,7 +167,9 @@ def _generate_uncorrelated_column_data(self, num_samples):
)
)
else:
if col_["order"] is not None:
if (not generator_name == "null_generator") and col_[
"order"
] is not None:
Comment on lines 169 to +172
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

just an else here (only way it gets is if generator_name not in the dictionary keys) and just logging.warning)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The warning above is not needed anymore just because of the null columns now being accounted for.

logging.warning(
f"""{generator_name} is passed with sorting type of {col_["order"]}.
Ascending and descending are the only supported options.
Expand All @@ -178,7 +181,7 @@ def _generate_uncorrelated_column_data(self, num_samples):
else:
dataset.append(generated_data)

column_names.append(generator_name)
column_names.append(column_header)

return self.convert_data_to_df(dataset, column_names=column_names)

Expand Down
84 changes: 57 additions & 27 deletions tests/test_generators.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,7 +74,25 @@ def test_synthesize_uncorrelated_output(self):
np.testing.assert_array_equal(
actual_synthetic_data.columns.values,
np.array(
["datetime", "categorical", "int", "string", "float"], dtype="object"
[
"datetime",
"host",
"src",
"proto",
"type",
"srcport",
"destport",
"srcip",
"locale",
"localeabbr",
"postalcode",
"latitude",
"longitude",
"owner",
"comment",
"int_col",
],
dtype="object",
),
)

Expand Down Expand Up @@ -301,13 +319,13 @@ def test_generate_uncorrelated_column_data(
else:
self.assertEqual(call_args_list[key], expected_calls[j][key])

@mock.patch("synthetic_data.generators.logging.warning")
@mock.patch("dataprofiler.profilers.StructuredProfiler.report")
def test_get_ordered_column_integration(self, mock_report, mock_warning):
def test_get_ordered_column_integration(self, mock_report):
mock_report.return_value = {
"data_stats": [
{
"data_type": "int",
"column_name": "test_column_1",
"order": "ascending",
"statistics": {
"min": 1.0,
Expand All @@ -316,6 +334,7 @@ def test_get_ordered_column_integration(self, mock_report, mock_warning):
},
{
"data_type": "string",
"column_name": "test_column_2",
"categorical": False,
"order": "ascending",
"statistics": {
Expand All @@ -326,6 +345,7 @@ def test_get_ordered_column_integration(self, mock_report, mock_warning):
},
{
"data_type": "string",
"column_name": "test_column_3",
"categorical": True,
"order": "ascending",
"statistics": {
Expand All @@ -342,11 +362,13 @@ def test_get_ordered_column_integration(self, mock_report, mock_warning):
},
{
"data_type": "float",
"column_name": "test_column_4",
"order": "ascending",
"statistics": {"min": 2.11234, "max": 8.0, "precision": {"max": 6}},
},
{
"data_type": "datetime",
"column_name": "test_column_5",
"order": "ascending",
"statistics": {
"format": ["%Y-%m-%d"],
Expand All @@ -355,6 +377,7 @@ def test_get_ordered_column_integration(self, mock_report, mock_warning):
},
},
{
"column_name": "test_column_6",
"data_type": None,
},
]
Expand All @@ -363,36 +386,43 @@ def test_get_ordered_column_integration(self, mock_report, mock_warning):
self.assertFalse(generator.is_correlated)

expected_array = [
[1, "arif", "blue", 2.246061, "2003-06-02"],
[1, "daips", "blue", 2.628393, "2003-10-08"],
[1, "dree", "orange", 2.642511, "2006-02-17"],
[1, "drqs", "orange", 2.807119, "2006-11-18"],
[1, "dwdaa", "orange", 3.009102, "2008-12-07"],
[2, "fswfe", "orange", 3.061853, "2009-12-03"],
[2, "fwqe", "orange", 3.677692, "2013-02-24"],
[2, "ipdpd", "orange", 3.887541, "2013-08-18"],
[3, "pdis", "red", 4.24257, "2014-02-19"],
[3, "peii", "red", 4.355663, "2014-04-29"],
[3, "pepie", "red", 4.739156, "2017-12-13"],
[3, "qrdq", "red", 4.831716, "2018-02-03"],
[3, "qrps", "yellow", 5.062321, "2019-05-13"],
[3, "rrqp", "yellow", 5.82323, "2020-01-09"],
[4, "sasr", "yellow", 6.212038, "2021-12-29"],
[4, "sspwe", "yellow", 6.231978, "2022-01-25"],
[4, "sssi", "yellow", 6.365346, "2023-03-20"],
[4, "wpfsi", "yellow", 7.461754, "2023-10-23"],
[4, "wqfed", "yellow", 7.775666, "2026-02-04"],
[4, "wsde", "yellow", 7.818521, "2027-06-13"],
[1, "arif", "blue", 2.246061, "2003-06-02", None],
[1, "daips", "blue", 2.628393, "2003-10-08", None],
[1, "dree", "orange", 2.642511, "2006-02-17", None],
[1, "drqs", "orange", 2.807119, "2006-11-18", None],
[1, "dwdaa", "orange", 3.009102, "2008-12-07", None],
[2, "fswfe", "orange", 3.061853, "2009-12-03", None],
[2, "fwqe", "orange", 3.677692, "2013-02-24", None],
[2, "ipdpd", "orange", 3.887541, "2013-08-18", None],
[3, "pdis", "red", 4.24257, "2014-02-19", None],
[3, "peii", "red", 4.355663, "2014-04-29", None],
[3, "pepie", "red", 4.739156, "2017-12-13", None],
[3, "qrdq", "red", 4.831716, "2018-02-03", None],
[3, "qrps", "yellow", 5.062321, "2019-05-13", None],
[3, "rrqp", "yellow", 5.82323, "2020-01-09", None],
[4, "sasr", "yellow", 6.212038, "2021-12-29", None],
[4, "sspwe", "yellow", 6.231978, "2022-01-25", None],
[4, "sssi", "yellow", 6.365346, "2023-03-20", None],
[4, "wpfsi", "yellow", 7.461754, "2023-10-23", None],
[4, "wqfed", "yellow", 7.775666, "2026-02-04", None],
[4, "wsde", "yellow", 7.818521, "2027-06-13", None],
]
expected_column_names = [
"test_column_1",
"test_column_2",
"test_column_3",
"test_column_4",
"test_column_5",
"test_column_6",
]
categories = ["int", "string", "categorical", "float", "datetime"]

expected_data = [dict(zip(categories, item)) for item in expected_array]
expected_data = [
dict(zip(expected_column_names, item)) for item in expected_array
]
expected_df = pd.DataFrame(expected_data)

actual_df = generator.synthesize(20)

self.assertEqual(mock_warning.call_count, 1)
mock_warning.assert_called_with(f"Generator of type None is not implemented.")
pd.testing.assert_frame_equal(expected_df, actual_df)

@mock.patch("dataprofiler.profilers.StructuredProfiler.report")
Expand Down