whylogs.core.metrics.condition_count_metric#

Module Contents#

Classes#

Condition

Condition to be evaluated by the ConditionCountMetric.

ConditionCountConfig

ConditionCountMetric

A whylogs metric that counts how many column entries satisfy a condition.

Functions#

relation(→ Callable[[Any], bool])

and_relations(→ Callable[[Any], bool])

or_relations(→ Callable[[Any], bool])

not_relation(→ Callable[[Any], bool])

Attributes#

logger

Relation

whylogs.core.metrics.condition_count_metric.logger#
whylogs.core.metrics.condition_count_metric.Relation#
whylogs.core.metrics.condition_count_metric.relation(op: Relation, value: Union[str, int, float]) → Callable[[Any], bool]#
Parameters
Return type

Callable[[Any], bool]

whylogs.core.metrics.condition_count_metric.and_relations(left: Callable[[Any], bool], right: Callable[[Any], bool]) → Callable[[Any], bool]#
Parameters

  • left (Callable[[Any], bool]) –

  • right (Callable[[Any], bool]) –

Return type

Callable[[Any], bool]

whylogs.core.metrics.condition_count_metric.or_relations(left: Callable[[Any], bool], right: Callable[[Any], bool]) → Callable[[Any], bool]#
Parameters

  • left (Callable[[Any], bool]) –

  • right (Callable[[Any], bool]) –

Return type

Callable[[Any], bool]

whylogs.core.metrics.condition_count_metric.not_relation(relation: Callable[[Any], bool]) → Callable[[Any], bool]#
Parameters

relation (Callable[[Any], bool]) –

Return type

Callable[[Any], bool]

class whylogs.core.metrics.condition_count_metric.Condition#

Condition to be evaluated by the ConditionCountMetric.

Parameters

  • relation (Callable[[Any], bool]) – The predicate to evaluate. The callable is passed a value from the column the ConditionCountMetric is attached to, and returns True if the value satisfies the condition.

  • throw_on_failure (bool) – If throw_on_failure is true, whylogs will immediately raise a ValueError if data that does not satisfy the condition is logged.

  • log_on_failure (bool) – If log_on_failure is true, whylogs will log a warning message if data that does not satisfy the conditon is logged.

  • actions (List[Callable[[str, str, Any], None]]) – A list of callables that will be invoked if data that does not satisfy the conditon is logged. The arguments passed to the callable are the metric’s name (“condition_count”), the name of the failed condition, and the value that caused the failure.

relation: Callable[[Any], bool]#
throw_on_failure: bool = False#
log_on_failure: bool = False#
actions: List[Callable[[str, str, Any], None]]#
class whylogs.core.metrics.condition_count_metric.ConditionCountConfig#

Bases: whylogs.core.metrics.metrics.MetricConfig

conditions: Dict[str, Condition]#
exclude_from_serialization: bool = False#
class whylogs.core.metrics.condition_count_metric.ConditionCountMetric#

Bases: whylogs.core.metrics.metrics.Metric

A whylogs metric that counts how many column entries satisfy a condition.

Parameters

conditions (Dict[str, Condition]) – The conditions evaluated by the metric. The key is the condition name, and the Condition value specifies the Callable condition predicate to evaluate & count.

Examples

This example counts the occurrances of email addresses in the some_text column and credit card numbers in the more_text column.

``` import pandas as pd import whylogs as why from whylogs.core.resolvers import STANDARD_RESOLVER from whylogs.core.specialized_resolvers import ConditionCountMetricSpec from whylogs.core.metrics.condition_count_metric import Condition from whylogs.core.relations import Predicate from whylogs.core.schema import DeclarativeSchema

email_condition = {“contiansEmail”: Condition(Predicate().fullmatch(“[w.]+[._]?[a-z0-9]+[@]w+[.]w{2,3}”))} cc_condition = {“containsCreditCard”: Condition(Predicate().matches(“.*4[0-9]{12}(?:[0-9]{3})?”))}

schema = DeclarativeSchema(STANDARD_RESOLVER) schema.add_resolver_spec(column_name=”some_text”, metrics=[ConditionCountMetricSpec(email_condition)]) schema.add_resolver_spec(column_name=”more_text”, metrics=[ConditionCountMetricspec(cc_condition)])

df = pd.DataFrame({“some_text”: [“not an email”, “bob@spam.com”], “more_text”: [“frogs”, “4000000000000”]}) view = why.log(df).view() view.to_pandas()[[‘condition_count/containsEmail’, ‘condition_count/containsCreditCard’, ‘condition_count/total’]]

# results in

condition_count/containsEmail condition_count/containsCreditCard condition_count/total

column some_text 1.0 NaN 2 more_text NaN 1.0 2 ```

property exclude_from_serialization: bool#
Return type

bool

property namespace: str#
Return type

str

conditions: Dict[str, Condition]#
total: whylogs.core.metrics.metric_components.IntegralComponent#
matches: Dict[str, whylogs.core.metrics.metric_components.IntegralComponent]#
hide_from_serialization: bool = False#
merge(other: ConditionCountMetric) → ConditionCountMetric#
Parameters

other (ConditionCountMetric) –

Return type

ConditionCountMetric

add_conditions(conditions: Dict[str, Condition]) → None#
Parameters

conditions (Dict[str, Condition]) –

Return type

None

get_component_paths() → List[str]#
Return type

List[str]

columnar_update(data: whylogs.core.preprocessing.PreprocessedColumn) → whylogs.core.metrics.metrics.OperationResult#
Parameters

data (whylogs.core.preprocessing.PreprocessedColumn) –

Return type

whylogs.core.metrics.metrics.OperationResult

classmethod zero(config: Optional[whylogs.core.metrics.metrics.MetricConfig] = None) → ConditionCountMetric#
Parameters

config (Optional[whylogs.core.metrics.metrics.MetricConfig]) –

Return type

ConditionCountMetric

to_protobuf() → whylogs.core.proto.MetricMessage#
Return type

whylogs.core.proto.MetricMessage

to_summary_dict(cfg: Optional[whylogs.core.configs.SummaryConfig] = None) → Dict[str, Any]#
Parameters

cfg (Optional[whylogs.core.configs.SummaryConfig]) –

Return type

Dict[str, Any]

classmethod from_protobuf(msg: whylogs.core.proto.MetricMessage) → ConditionCountMetric#
Parameters

msg (whylogs.core.proto.MetricMessage) –

Return type

ConditionCountMetric