24912462866

Committed 24 Apr 2026 09:21PM UTC coverage: 76.362% (-3.1%) from 79.464%

Build # 24912462866

Build Type

push

github

Committed by

web-flow

Commit Message

Merge pull request #73 from rafaelpadilla/eduardo/refactor_data_operations

Refactor of data operations, trainers and models.

Coverage Stats

244 of 339 branches covered (71.98%)

Branch coverage included in aggregate %.

1317 of 1706 new or added lines in 50 files covered. (77.2%)

28 existing lines in 5 files now uncovered.

2124 of 2762 relevant lines covered (76.9%)

0.77 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

92.31

/toolkit/ThreeWToolkit/preprocessing/fill_labels.py

from typing import Literal
from pydantic import Field, field_validator, PrivateAttr, ValidationInfo

from ..core.base_dataset import BaseDataset
from ..core.base_preprocessing import BasePreprocessing, BasePreprocessingConfig
from ..core.dataset_outputs import DatasetOutputs


class FillLabelsConfig(BasePreprocessingConfig):
    """Configuration for FillLabels preprocessing step."""

    fill_method: Literal["nearest", "ffill", "bfill", "constant"] = Field(
        default="nearest", description="Method to fill missing values in labels."
    )
    fill_value: int | None = Field(
        default=None, description="Value to use when fill_method is 'constant'."
    )
    _target: type = PrivateAttr(default_factory=lambda: FillLabels)

    @field_validator("fill_value")
    @classmethod
    def validate_fill_value(cls, value: int | None, info: ValidationInfo) -> int | None:
        if info.data.get("fill_method") == "constant" and value is None:
            raise ValueError(
                "fill_value must be provided when fill_method is 'constant'."
            )
        return value


class FillLabels(BasePreprocessing):
    """
    Preprocessing step to fill missing values in label Series using bfill then ffill.
    Applies to each event's label data.
    """

    def __init__(self, config: FillLabelsConfig):
        """Initialize FillLabels preprocessing step with the given configuration.
        Args:
            config: Configuration for FillLabels preprocessing step.
        """
        self.config: FillLabelsConfig = config

    def fit(self, data: BaseDataset) -> None:
        """No fitting needed for this preprocessing step.

        Args:
            data: Dataset (not used).
        """
        pass

    def transform(self, data: DatasetOutputs) -> DatasetOutputs:
        """Fill missing label values using the configured strategy.

        Args:
            data: Dataset outputs with labels to fill.

        Returns:
            Dataset outputs with filled labels.
        """

        if data.label is None:  # Noop if there are no labels to fill
            return data

        if self.config.fill_method == "constant":
            label = data.label.fillna(self.config.fill_value)
        elif self.config.fill_method == "ffill":
            label = (
                data.label.ffill().bfill()
            )  # Fill missing values using forward fill then backward fill
        elif self.config.fill_method == "bfill":
            label = (
                data.label.bfill().ffill()
            )  # Fill missing values using backward fill then forward fill
        elif self.config.fill_method == "nearest":
            # Fill missing values using nearest interpolation then bfill and ffill to handle any remaining NaNs
            label = data.label.interpolate(method="nearest").bfill().ffill()
        else:
            raise ValueError(f"Unsupported fill method: {self.config.fill_method}")

        return DatasetOutputs(signal=data.signal, label=label, metadata=data.metadata)

1	from typing import Literal	1✔
2	from pydantic import Field, field_validator, PrivateAttr, ValidationInfo	1✔
3
4	from ..core.base_dataset import BaseDataset	1✔
5	from ..core.base_preprocessing import BasePreprocessing, BasePreprocessingConfig	1✔
6	from ..core.dataset_outputs import DatasetOutputs	1✔
7
8
9	class FillLabelsConfig(BasePreprocessingConfig):	1✔
10	"""Configuration for FillLabels preprocessing step."""
11
12	fill_method: Literal["nearest", "ffill", "bfill", "constant"] = Field(	1✔
13	default="nearest", description="Method to fill missing values in labels."
14	)
15	fill_value: int \| None = Field(	1✔
16	default=None, description="Value to use when fill_method is 'constant'."
17	)
18	_target: type = PrivateAttr(default_factory=lambda: FillLabels)	1✔
19
20	@field_validator("fill_value")	1✔
21	@classmethod	1✔
22	def validate_fill_value(cls, value: int \| None, info: ValidationInfo) -> int \| None:	1✔
23	if info.data.get("fill_method") == "constant" and value is None:	1✔
NEW 24	raise ValueError(	×
25	"fill_value must be provided when fill_method is 'constant'."
26	)
27	return value	1✔
28
29
30	class FillLabels(BasePreprocessing):	1✔
31	"""
32	Preprocessing step to fill missing values in label Series using bfill then ffill.
33	Applies to each event's label data.
34	"""
35
36	def __init__(self, config: FillLabelsConfig):	1✔
37	"""Initialize FillLabels preprocessing step with the given configuration.
38	Args:
39	config: Configuration for FillLabels preprocessing step.
40	"""
41	self.config: FillLabelsConfig = config	1✔
42
43	def fit(self, data: BaseDataset) -> None:	1✔
44	"""No fitting needed for this preprocessing step.
45
46	Args:
47	data: Dataset (not used).
48	"""
49	pass	1✔
50
51	def transform(self, data: DatasetOutputs) -> DatasetOutputs:	1✔
52	"""Fill missing label values using the configured strategy.
53
54	Args:
55	data: Dataset outputs with labels to fill.
56
57	Returns:
58	Dataset outputs with filled labels.
59	"""
60
61	if data.label is None: # Noop if there are no labels to fill	1✔
NEW 62	return data	×
63
64	if self.config.fill_method == "constant":	1✔
65	label = data.label.fillna(self.config.fill_value)	1✔
66	elif self.config.fill_method == "ffill":	1✔
67	label = (	1✔
68	data.label.ffill().bfill()
69	) # Fill missing values using forward fill then backward fill
70	elif self.config.fill_method == "bfill":	1✔
71	label = (	1✔
72	data.label.bfill().ffill()
73	) # Fill missing values using backward fill then forward fill
74	elif self.config.fill_method == "nearest":	1✔
75	# Fill missing values using nearest interpolation then bfill and ffill to handle any remaining NaNs
76	label = data.label.interpolate(method="nearest").bfill().ffill()	1✔
77	else:
NEW 78	raise ValueError(f"Unsupported fill method: {self.config.fill_method}")	×
79
80	return DatasetOutputs(signal=data.signal, label=label, metadata=data.metadata)	1✔

rafaelpadilla / 3W / 24912462866

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous