Skip to content

Custom KPI Definition

CustomKPIDefinition

Bases: KPIDefinition

Base class for custom KPI computation logic.

Supports two computation patterns
  1. Batch computation: Override compute_batch()
  2. Per-object computation: Override compute_for_object()
Choose the pattern that best fits your use case
  • Use batch for efficient vectorized operations across all objects
  • Use per_object for complex logic that varies significantly per object
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
class CustomKPIDefinition(KPIDefinition):
    """
    Base class for custom KPI computation logic.

    Supports two computation patterns:
        1. Batch computation: Override compute_batch()
        2. Per-object computation: Override compute_for_object()

    Choose the pattern that best fits your use case:
        - Use batch for efficient vectorized operations across all objects
        - Use per_object for complex logic that varies significantly per object
    """

    def __init__(
        self,
        flag: FlagTypeProtocol,
        model_flag: FlagTypeProtocol | None = None,
        objects: list[Hashable] | Literal['auto'] = 'auto',
        name_prefix: str = '',
        name_suffix: str = '',
        extra_attributes: dict = None,
        aggregation: Aggregation | None = None
    ):
        """
        Initialize custom KPI definition.

        Args:
            flag: Variable flag for the KPI
            model_flag: Optional model flag (auto-inferred if None)
            objects: List of objects or 'auto' to discover
            name_prefix: Prefix for KPI names
            name_suffix: Suffix for KPI names
            aggregation: Optional aggregation (for metadata only, not used in computation)
        """
        self.flag = flag
        self.model_flag = model_flag
        self.objects = objects
        self.name_prefix = name_prefix
        self.name_suffix = name_suffix
        self.extra_attributes = extra_attributes
        self.aggregation = aggregation

    def generate_kpis(self, dataset: Dataset) -> list[KPI]:
        """
        Generate KPIs using either per-object or batch computation.

        Args:
            dataset: Dataset to compute KPIs for

        Returns:
            List of computed KPI instances
        """
        model_flag = self.model_flag or dataset.flag_index.get_linked_model_flag(self.flag)

        if self.objects == 'auto':
            objects = dataset.fetch(self.flag).columns.tolist()
        else:
            objects = self.objects

        try:
            return self._generate_kpis_batch(dataset, model_flag, objects)
        except NotImplementedError:
            pass
        try:
            return self._generate_kpis_per_object(dataset, model_flag, objects)
        except NotImplementedError:
            raise NotImplementedError("Must override compute_for_object() or compute_batch().")

    def _generate_kpis_per_object(
        self,
        dataset: Dataset,
        model_flag: FlagTypeProtocol,
        objects: list[Hashable]
    ) -> list[KPI]:
        """
        Generate KPIs by calling compute_for_object() for each object.

        Args:
            dataset: Dataset to compute for
            model_flag: Model flag for objects
            objects: List of object names

        Returns:
            List of KPI instances
        """
        kpis = []
        for obj in objects:
            value = self.compute_for_object(dataset, obj)

            attributes = self._build_attributes(obj, dataset, model_flag)

            kpi = KPI(
                value=value,
                attributes=attributes,
                dataset=dataset
            )
            kpis.append(kpi)

        return kpis

    def _generate_kpis_batch(
        self,
        dataset: Dataset,
        model_flag: FlagTypeProtocol,
        objects: list[Hashable]
    ) -> list[KPI]:
        """
        Generate KPIs by calling compute_batch() once.

        Args:
            dataset: Dataset to compute for
            model_flag: Model flag for objects
            objects: List of object names

        Returns:
            List of KPI instances
        """
        # Compute all values at once
        values_dict = self.compute_batch(dataset, objects)

        kpis = []
        for obj, value in values_dict.items():
            attributes = self._build_attributes(obj, dataset, model_flag)

            kpi = KPI(
                value=value,
                attributes=attributes,
                dataset=dataset
            )
            kpis.append(kpi)

        return kpis

    def compute_for_object(self, dataset: Dataset, object_name: Hashable) -> Any:
        """
        Compute KPI value for a single object.

        Override this for per-object computation pattern.

        Args:
            dataset: Dataset to fetch data from
            object_name: Name of the object to compute for

        Returns:
            Computed KPI value

        Raises:
            NotImplementedError: If not overridden
        """
        raise NotImplementedError("Must override compute_for_object() or compute_batch()")

    def compute_batch(self, dataset: Dataset, objects: list[Hashable]) -> dict[Hashable, Any]:
        """
        Compute KPI values for all objects at once.

        Override this for batch computation pattern.

        Args:
            dataset: Dataset to fetch data from
            objects: List of object names to compute for

        Returns:
            Dict mapping object_name → value

        Raises:
            NotImplementedError: If not overridden
        """
        raise NotImplementedError("Must override compute_for_object() or compute_batch()")

    @abstractmethod
    def get_unit(self) -> Units.Unit:
        """
        Return the unit for this KPI type.

        Returns:
            Physical unit for the KPI values
        """
        pass

    def _build_attributes(
        self,
        object_name: Hashable,
        dataset: Dataset,
        model_flag: FlagTypeProtocol
    ) -> KPIAttributes:
        """
        Build KPIAttributes for a KPI instance.

        Args:
            object_name: Object identifier
            dataset: Source dataset
            model_flag: Model flag for the object

        Returns:
            KPIAttributes instance
        """

        return KPIAttributes(
            flag=self.flag,
            model_flag=model_flag,
            object_name=object_name,
            aggregation=self.aggregation,
            dataset_name=dataset.name,
            dataset_type=type(dataset),
            name_prefix=self.name_prefix,
            name_suffix=self.name_suffix,
            unit=self.get_unit(),
            dataset_attributes=dataset.attributes,
            extra_attributes=self.extra_attributes or dict()
        )

__init__ 

__init__(flag: FlagTypeProtocol, model_flag: FlagTypeProtocol | None = None, objects: list[Hashable] | Literal['auto'] = 'auto', name_prefix: str = '', name_suffix: str = '', extra_attributes: dict = None, aggregation: Aggregation | None = None)

Initialize custom KPI definition.

Parameters:

Name Type Description Default
flag FlagTypeProtocol

Variable flag for the KPI

 required
model_flag FlagTypeProtocol | None

Optional model flag (auto-inferred if None)

 None
objects list[Hashable] | Literal['auto']

List of objects or 'auto' to discover

 'auto'
name_prefix str

Prefix for KPI names

 ''
name_suffix str

Suffix for KPI names

 ''
aggregation Aggregation | None

Optional aggregation (for metadata only, not used in computation)

 None
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
def __init__(
    self,
    flag: FlagTypeProtocol,
    model_flag: FlagTypeProtocol | None = None,
    objects: list[Hashable] | Literal['auto'] = 'auto',
    name_prefix: str = '',
    name_suffix: str = '',
    extra_attributes: dict = None,
    aggregation: Aggregation | None = None
):
    """
    Initialize custom KPI definition.

    Args:
        flag: Variable flag for the KPI
        model_flag: Optional model flag (auto-inferred if None)
        objects: List of objects or 'auto' to discover
        name_prefix: Prefix for KPI names
        name_suffix: Suffix for KPI names
        aggregation: Optional aggregation (for metadata only, not used in computation)
    """
    self.flag = flag
    self.model_flag = model_flag
    self.objects = objects
    self.name_prefix = name_prefix
    self.name_suffix = name_suffix
    self.extra_attributes = extra_attributes
    self.aggregation = aggregation

generate_kpis 

generate_kpis(dataset: Dataset) -> list[KPI]

Generate KPIs using either per-object or batch computation.

Parameters:

Name Type Description Default
dataset Dataset

Dataset to compute KPIs for

 required

Returns:

Type Description
list[KPI]

List of computed KPI instances
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
def generate_kpis(self, dataset: Dataset) -> list[KPI]:
    """
    Generate KPIs using either per-object or batch computation.

    Args:
        dataset: Dataset to compute KPIs for

    Returns:
        List of computed KPI instances
    """
    model_flag = self.model_flag or dataset.flag_index.get_linked_model_flag(self.flag)

    if self.objects == 'auto':
        objects = dataset.fetch(self.flag).columns.tolist()
    else:
        objects = self.objects

    try:
        return self._generate_kpis_batch(dataset, model_flag, objects)
    except NotImplementedError:
        pass
    try:
        return self._generate_kpis_per_object(dataset, model_flag, objects)
    except NotImplementedError:
        raise NotImplementedError("Must override compute_for_object() or compute_batch().")

compute_for_object 

compute_for_object(dataset: Dataset, object_name: Hashable) -> Any

Compute KPI value for a single object.

Override this for per-object computation pattern.

Parameters:

Name Type Description Default
dataset Dataset

Dataset to fetch data from

 required
object_name Hashable

Name of the object to compute for

 required

Returns:

Type Description
Any

Computed KPI value

Raises:

Type Description
NotImplementedError

If not overridden
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
def compute_for_object(self, dataset: Dataset, object_name: Hashable) -> Any:
    """
    Compute KPI value for a single object.

    Override this for per-object computation pattern.

    Args:
        dataset: Dataset to fetch data from
        object_name: Name of the object to compute for

    Returns:
        Computed KPI value

    Raises:
        NotImplementedError: If not overridden
    """
    raise NotImplementedError("Must override compute_for_object() or compute_batch()")

compute_batch 

compute_batch(dataset: Dataset, objects: list[Hashable]) -> dict[Hashable, Any]

Compute KPI values for all objects at once.

Override this for batch computation pattern.

Parameters:

Name Type Description Default
dataset Dataset

Dataset to fetch data from

 required
objects list[Hashable]

List of object names to compute for

 required

Returns:

Type Description
dict[Hashable, Any]

Dict mapping object_name → value

Raises:

Type Description
NotImplementedError

If not overridden
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
def compute_batch(self, dataset: Dataset, objects: list[Hashable]) -> dict[Hashable, Any]:
    """
    Compute KPI values for all objects at once.

    Override this for batch computation pattern.

    Args:
        dataset: Dataset to fetch data from
        objects: List of object names to compute for

    Returns:
        Dict mapping object_name → value

    Raises:
        NotImplementedError: If not overridden
    """
    raise NotImplementedError("Must override compute_for_object() or compute_batch()")

get_unit abstractmethod 

get_unit() -> Unit

Return the unit for this KPI type.

Returns:

Type Description
Unit

Physical unit for the KPI values
Source code in submodules/mesqual/mesqual/kpis/definitions/custom.py 
187
188
189
190
191
192
193
194
195
@abstractmethod
def get_unit(self) -> Units.Unit:
    """
    Return the unit for this KPI type.

    Returns:
        Physical unit for the KPI values
    """
    pass