ML Workflows

DeTime provides a workflow layer around decomposition for machine-learning pipelines. It wraps decomposition methods with a shared configuration contract, result object, and CLI/Python surface.

Where it fits in ML pipelines

Common machine-learning-facing uses include:

  • denoising a series before feature engineering or model fitting,
  • separating trend and seasonal structure before downstream regression or classification,
  • generating components that can be summarized into tabular features,
  • inspecting shared structure across channels before multivariate modeling.

The package contribution is that these steps use one configuration contract, one result object, and one CLI/Python surface rather than a mix of notebooks, method-specific wrappers, and one-off scripts.

Raw series sensor, experimental, or panel data
DeTime trend / season / residual / components
Features component summaries, residual diagnostics, channel structure
Model regression, classification, clustering, or inspection

Small scikit-learn-facing example

import numpy as np
from sklearn.linear_model import LinearRegression

from detime import DecompositionConfig, decompose

t = np.arange(120, dtype=float)
series = 0.02 * t + np.sin(2.0 * np.pi * t / 12.0)

result = decompose(
    series,
    DecompositionConfig(
        method="SSA",
        params={"window": 24, "rank": 6, "primary_period": 12},
    ),
)

X = np.column_stack([result.trend, result.season, result.residual])
y = series

model = LinearRegression().fit(X, y)
print(model.score(X, y))

This example is intentionally small. The point is not that DeTime replaces scikit-learn, but that decomposition outputs can feed a downstream estimator through a stable package-level workflow.

Project and review notes

Release checks, coverage boundaries, and reviewer-facing evidence live in Reproducibility. This page stays focused on where the package fits in downstream ML work.