Metadata-Version: 2.4
Name: tosa-ml
Version: 0.1.0
Summary: High-level machine learning preprocessing toolkit
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: pandas>=2.0
Requires-Dist: numpy>=1.24
Requires-Dist: scikit-learn>=1.3
Dynamic: license-file

# tosa-ml

High-level machine learning preprocessing toolkit.

## Install (local)

```bash
python -m venv .venv
# Windows: .venv\Scripts\activate
# Linux/Mac: source .venv/bin/activate
pip install pandas numpy scikit-learn pytest
pip install -e .
```

## Auto mode (1 line)

```python
import pandas as pd
from tosa_ml import preprocess

df = pd.read_csv("data.csv")
X_train, X_test, y_train, y_test = preprocess(df, target="price")
```

## Controlled mode

```python
X_train, X_test, y_train, y_test = preprocess(
    df,
    target="price",
    missing="auto",
    encoding="onehot",
    scaling="minmax",
    outliers="iqr",
    test_size=0.2,
    random_state=42,
    duplicate_removal=True,
    outlier_threshold=1.5,
)
```

## Run tests

```bash
pytest
```

## Status

MVP. Known TODO: fit/transform separation (leakage-free), persistence (joblib),
feature selection integration, datetime/text/high-cardinality columns.
