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

feat: add data quality for regression #78

Merged
merged 3 commits into from
Jul 4, 2024
Merged

feat: add data quality for regression #78

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fa75f79
feat: add data quality for regression
dtria91 Jul 2, 2024
c9c252a
fix: run ruff fix
dtria91 Jul 2, 2024
2f4f3c5
fix: edit (sdk) current regression test
dtria91 Jul 3, 2024
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
25 changes: 24 additions & 1 deletion api/app/models/metrics/data_quality_dto.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,20 @@ class NumericalFeatureMetrics(FeatureMetrics):
)


class NumericalTargetMetrics(FeatureMetrics):
type: str = 'numerical'
mean: float
std: float
min: float
max: float
median_metrics: MedianMetrics
histogram: Histogram

model_config = ConfigDict(
populate_by_name=True, alias_generator=to_camel, protected_namespaces=()
)


class CategoryFrequency(BaseModel):
name: str
count: int
Expand Down Expand Up @@ -116,7 +130,16 @@ class ClassificationDataQuality(BaseModel):


class RegressionDataQuality(BaseModel):
pass
n_observations: int
target_metrics: NumericalTargetMetrics
feature_metrics: List[NumericalFeatureMetrics]

model_config = ConfigDict(
arbitrary_types_allowed=True,
populate_by_name=True,
alias_generator=to_camel,
protected_namespaces=(),
)


class DataQualityDTO(BaseModel):
Expand Down
56 changes: 53 additions & 3 deletions api/tests/commons/db_mock.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -265,7 +265,7 @@ def get_sample_current_dataset(
},
}

data_quality_dict = {
classification_data_quality_dict = {
'nObservations': 200,
'classMetrics': [
{'name': 'classA', 'count': 100, 'percentage': 50.0},
Expand Down Expand Up @@ -312,6 +312,56 @@ def get_sample_current_dataset(
],
}

regression_data_quality_dict = {
'nObservations': 200,
'targetMetrics': {
'max': 3410.0,
'min': 2.0,
'std': 686.62,
'mean': 848.12,
'type': 'numerical',
'histogram': {
'buckets': [2.0, 342.8, 683.6, 1024.4],
'reference_values': [204, 144, 165, 89],
},
'feature_name': 'ground_truth',
'missing_value': {'count': 0, 'percentage': 0.0},
'median_metrics': {'median': 713.0, 'perc_25': 315.0, 'perc_75': 1097.0},
},
'featureMetrics': [
{
'max': 731.0,
'min': 1.0,
'std': 211.16,
'mean': 366.0,
'type': 'numerical',
'histogram': {
'buckets': [1.0, 74.0, 147.0, 220.0],
'reference_values': [73, 73, 73, 73],
},
'feature_name': 'instant',
'missing_value': {'count': 0, 'percentage': 0.0},
'median_metrics': {'median': 366.0, 'perc_25': 183.5, 'perc_75': 548.5},
'class_median_metrics': [],
},
{
'max': 4.0,
'min': 1.0,
'std': 1.12,
'mean': 2.49,
'type': 'numerical',
'histogram': {
'buckets': [1.0, 1.3, 1.6, 1.9],
'reference_values': [181, 0, 0, 184],
},
'feature_name': 'season',
'missing_value': {'count': 0, 'percentage': 0.0},
'median_metrics': {'median': 3.0, 'perc_25': 2.0, 'perc_75': 3.0},
'class_median_metrics': [],
},
],
}

drift_dict = {
'featureMetrics': [
{
Expand All @@ -333,7 +383,7 @@ def get_sample_current_dataset(
def get_sample_reference_metrics(
reference_uuid: uuid.UUID = REFERENCE_UUID,
model_quality: Dict = binary_model_quality_dict,
data_quality: Dict = data_quality_dict,
data_quality: Dict = classification_data_quality_dict,
statistics: Dict = statistics_dict,
) -> ReferenceDatasetMetrics:
return ReferenceDatasetMetrics(
Expand All @@ -347,7 +397,7 @@ def get_sample_reference_metrics(
def get_sample_current_metrics(
current_uuid: uuid.UUID = CURRENT_UUID,
model_quality: Dict = binary_current_model_quality_dict,
data_quality: Dict = data_quality_dict,
data_quality: Dict = classification_data_quality_dict,
statistics: Dict = statistics_dict,
drift: Dict = drift_dict,
) -> CurrentDatasetMetrics:
Expand Down
58 changes: 58 additions & 0 deletions api/tests/services/metrics_service_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -257,6 +257,36 @@ def test_get_empty_reference_data_quality_by_model_by_uuid(self):
data_quality_data=None,
)

def test_get_reference_regression_data_quality_by_model_by_uuid(self):
status = JobStatus.SUCCEEDED
reference_dataset = db_mock.get_sample_reference_dataset(status=status.value)
reference_metrics = db_mock.get_sample_reference_metrics(
data_quality=db_mock.regression_data_quality_dict
)
model = db_mock.get_sample_model(model_type=ModelType.REGRESSION)
self.model_service.get_model_by_uuid = MagicMock(return_value=model)
self.reference_dataset_dao.get_reference_dataset_by_model_uuid = MagicMock(
return_value=reference_dataset
)
self.reference_metrics_dao.get_reference_metrics_by_model_uuid = MagicMock(
return_value=reference_metrics
)
res = self.metrics_service.get_reference_data_quality_by_model_by_uuid(
model_uuid
)
self.reference_dataset_dao.get_reference_dataset_by_model_uuid.assert_called_once_with(
model_uuid
)
self.reference_metrics_dao.get_reference_metrics_by_model_uuid.assert_called_once_with(
model_uuid
)

assert res == DataQualityDTO.from_dict(
model_type=model.model_type,
job_status=reference_dataset.status,
data_quality_data=reference_metrics.data_quality,
)

def test_get_current_statistics_by_model_by_uuid(self):
status = JobStatus.SUCCEEDED
current_dataset = db_mock.get_sample_current_dataset(status=status.value)
Expand Down Expand Up @@ -462,6 +492,34 @@ def test_get_empty_current_data_quality_by_model_by_uuid(self):
data_quality_data=None,
)

def test_get_current_regression_data_quality_by_model_by_uuid(self):
status = JobStatus.SUCCEEDED
current_dataset = db_mock.get_sample_current_dataset(status=status.value)
current_metrics = db_mock.get_sample_current_metrics(data_quality=db_mock.regression_data_quality_dict)
model = db_mock.get_sample_model(model_type=ModelType.REGRESSION)
self.model_service.get_model_by_uuid = MagicMock(return_value=model)
self.current_dataset_dao.get_current_dataset_by_model_uuid = MagicMock(
return_value=current_dataset
)
self.current_metrics_dao.get_current_metrics_by_model_uuid = MagicMock(
return_value=current_metrics
)
res = self.metrics_service.get_current_data_quality_by_model_by_uuid(
model_uuid, current_dataset.uuid
)
self.current_dataset_dao.get_current_dataset_by_model_uuid.assert_called_once_with(
model_uuid, current_dataset.uuid
)
self.current_metrics_dao.get_current_metrics_by_model_uuid.assert_called_once_with(
model_uuid, current_dataset.uuid
)

assert res == DataQualityDTO.from_dict(
model_type=model.model_type,
job_status=current_dataset.status,
data_quality_data=current_metrics.data_quality,
)

def test_get_current_binary_class_model_quality_by_model_by_uuid(self):
status = JobStatus.SUCCEEDED
current_dataset = db_mock.get_sample_current_dataset(status=status.value)
Expand Down
8 changes: 6 additions & 2 deletions api/tests/validation/model_type_validator_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,9 @@ def test_prediction_for_binary():
def test_prediction_for_multiclass():
"""Tests that for ModelType.MULTI_CLASS: prediction must be a number or string."""
with pytest.raises(ValidationError) as excinfo:
model_data = get_model_sample_wrong(['outputs.prediction'], ModelType.MULTI_CLASS)
model_data = get_model_sample_wrong(
['outputs.prediction'], ModelType.MULTI_CLASS
)
ModelIn.model_validate(ModelIn(**model_data))
assert 'prediction must be a number or string for a ModelType.MULTI_CLASS' in str(
excinfo.value
Expand All @@ -62,7 +64,9 @@ def test_prediction_for_multiclass():
def test_prediction_for_regression():
"""Tests that for ModelType.REGRESSION: prediction must be a number."""
with pytest.raises(ValidationError) as excinfo:
model_data = get_model_sample_wrong(['outputs.prediction'], ModelType.REGRESSION)
model_data = get_model_sample_wrong(
['outputs.prediction'], ModelType.REGRESSION
)
ModelIn.model_validate(ModelIn(**model_data))
assert 'prediction must be a number for a ModelType.REGRESSION' in str(
excinfo.value
Expand Down
22 changes: 21 additions & 1 deletion sdk/radicalbit_platform_sdk/models/dataset_data_quality.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,6 +64,18 @@ class NumericalFeatureMetrics(FeatureMetrics):
model_config = ConfigDict(populate_by_name=True, alias_generator=to_camel)


class NumericalTargetMetrics(FeatureMetrics):
type: str = 'numerical'
mean: float
std: float
min: float
max: float
median_metrics: MedianMetrics
histogram: Histogram

model_config = ConfigDict(populate_by_name=True, alias_generator=to_camel)


class CategoryFrequency(BaseModel):
name: str
count: int
Expand Down Expand Up @@ -97,4 +109,12 @@ class ClassificationDataQuality(DataQuality):


class RegressionDataQuality(DataQuality):
pass
n_observations: int
target_metrics: NumericalTargetMetrics
feature_metrics: List[NumericalFeatureMetrics]

model_config = ConfigDict(
arbitrary_types_allowed=True,
populate_by_name=True,
alias_generator=to_camel,
)
59 changes: 57 additions & 2 deletions sdk/tests/apis/model_current_dataset_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -461,14 +461,69 @@ def test_regression_data_quality_ok(self):
body="""{
"datetime": "something_not_used",
"jobStatus": "SUCCEEDED",
"dataQuality": {}
"dataQuality": {
"n_observations":731,
"target_metrics": {
"max":3410.0,
"min":2.0,
"std":686.62,
"mean":848.17,
"type":"numerical",
"histogram":{
"buckets":[2.0, 342.8, 683.6, 1024.4],
"reference_values":[204, 144, 165, 89],
"current_values":[123, 231, 122, 89]
},
"feature_name":"ground_truth",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":713.0, "perc_25":315.0, "perc_75":1097.0}
},
"featureMetrics": [
{
"max":731.0,
"min":1.0,
"std":211.16,
"mean":366.0,
"type":"numerical",
"histogram":{
"buckets":[1.0, 74.0, 147.0, 220.0],
"reference_values":[73, 73, 73, 73],
"current_values":[73, 73, 73, 73]
},
"feature_name":"instant",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":366.0, "perc_25":183.5, "perc_75":548.5},
"class_median_metrics":[]
},
{
"max":4.0,
"min":1.0,
"std":1.11,
"mean":2.49,
"type":"numerical",
"histogram":{
"buckets":[1.0, 1.3, 1.6, 1.9],
"reference_values":[181, 0, 0, 184],
"current_values":[123, 0, 0, 212]
},
"feature_name":"season",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":3.0, "perc_25":2.0, "perc_75":3.0},
"class_median_metrics":[]
}
]
}
}""",
)

metrics = model_current_dataset.data_quality()

assert isinstance(metrics, RegressionDataQuality)
# TODO: add asserts to properties
assert metrics.n_observations == 731
assert metrics.target_metrics.feature_name == 'ground_truth'
assert metrics.target_metrics.median_metrics.median == 713.0
assert metrics.feature_metrics[0].max == 731.0
assert len(metrics.feature_metrics) == 2
assert model_current_dataset.status() == JobStatus.SUCCEEDED

@responses.activate
Expand Down
62 changes: 57 additions & 5 deletions sdk/tests/apis/model_reference_dataset_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -639,16 +639,68 @@ def test_regression_data_quality_ok(self):
url=f'{base_url}/api/models/{str(model_id)}/reference/data-quality',
status=200,
body="""{
"datetime": "something_not_used",
"jobStatus": "SUCCEEDED",
"dataQuality": {}
}""",
"datetime": "something_not_used",
"jobStatus": "SUCCEEDED",
"dataQuality": {
"n_observations":731,
"target_metrics": {
"max":3410.0,
"min":2.0,
"std":686.62,
"mean":848.17,
"type":"numerical",
"histogram":{
"buckets":[2.0, 342.8, 683.6, 1024.4],
"reference_values":[204, 144, 165, 89]
},
"feature_name":"ground_truth",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":713.0, "perc_25":315.0, "perc_75":1097.0}
},
"featureMetrics": [
{
"max":731.0,
"min":1.0,
"std":211.16,
"mean":366.0,
"type":"numerical",
"histogram":{
"buckets":[1.0, 74.0, 147.0, 220.0],
"reference_values":[73, 73, 73, 73]
},
"feature_name":"instant",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":366.0, "perc_25":183.5, "perc_75":548.5},
"class_median_metrics":[]
},
{
"max":4.0,
"min":1.0,
"std":1.11,
"mean":2.49,
"type":"numerical",
"histogram":{
"buckets":[1.0, 1.3, 1.6, 1.9],
"reference_values":[181, 0, 0, 184]
},
"feature_name":"season",
"missing_value":{"count":0, "percentage":0.0},
"median_metrics":{"median":3.0, "perc_25":2.0, "perc_75":3.0},
"class_median_metrics":[]
}
]
}
}""",
)

metrics = model_reference_dataset.data_quality()

assert isinstance(metrics, RegressionDataQuality)
# TODO: add asserts to properties
assert metrics.n_observations == 731
assert metrics.target_metrics.feature_name == 'ground_truth'
assert metrics.target_metrics.median_metrics.median == 713.0
assert metrics.feature_metrics[0].max == 731.0
assert len(metrics.feature_metrics) == 2
assert model_reference_dataset.status() == JobStatus.SUCCEEDED

@responses.activate
Expand Down