---
iso_doc_type: "Procedure"
iso_viewpoint: "ComponentView"
type: "module"
title: "regression_model_template::init_data — Synthetic Dataset Generator"
source_path: "src/regression_model_template/init_data.py"
description: "Generates reproducible synthetic regression datasets matching Pandera InputsSchema and TargetsSchema."
tags: ["iso42010", "okf", "component_view", "data"]
timestamp: "2026-10-01T17:50:00Z"
generated: "agent:wiki-iso-documentation-standard"
verified: "true"
last_verified_commit: "073b5fd"
---

# regression_model_template::init_data — Synthetic Dataset Generator

> **Source**: `src/regression_model_template/init_data.py` (Lines: L1-L66)  
> **Layer**: Utilities / Data Seeding  
> **Role**: Generates reproducible synthetic bike sharing demand regression datasets for development and testing.

---

## 1. Component Architecture & Signatures

```mermaid
classDiagram
    class InitData {
        <<script>>
        +generate_data(output_dir: str = "data") None
        +main() None
    }
```

### `generate_data(output_dir: str = "data") -> None`
- **Source Citation**: `src/regression_model_template/init_data.py:L10-L54`
- **Visibility**: Public (`+`)
- **Behavior**: Synthesizes 1,000 hourly bike-sharing records containing calendar, weather, temperature, humidity, and windspeed features, persisting outputs as `inputs.parquet` and `targets.parquet`.

---

> *Related: [Data Model](../../architecture/mission_data_model.md) · [Datasets](io/datasets.md)*