Home
Sake is a modular Python machine-learning framework for budget execution, statistical analysis, visualization, and model benchmarking.
🧠Purpose¶
This documentation site explains how to install, operate, extend, and validate Sake.
🧱 Core Capabilities¶
| Capability | Description |
|---|---|
| Data loading | Load CSV, Excel, Account Balances, and Pandas DataFrame inputs. |
| Descriptive statistics | Summarize budget execution data using traditional statistical measures. |
| Inferential statistics | Evaluate correlations, group differences, and statistical significance. |
| Feature engineering | Prepare machine-learning features through encoding, coercion, and dimensionality reduction. |
| Classification | Train and evaluate supervised classification models. |
| Regression | Train and evaluate supervised regression models. |
| Visualization | Review diagnostics, residuals, confusion matrices, ROC curves, and feature importance. |
🚀 Quick Start¶
Run these commands from the project root.
python -m venv .venv
.venv\Scripts\Activate.ps1
pip install -r requirements.txt
streamlit run app.py
🔗 Documentation Sections¶
Use the navigation tabs to review architecture, data sources, model evaluation, user workflows, API reference, and development guidance.