novatium 0.1.0

Novatium — Cosmic-Grade Machine Learning with NovaRegressor and NovaClassifier

Novatium

Cosmic-Grade Machine Learning — supercharged Base+Delta models for regression and classification.

Install (from TestPyPI first)

pip install -i https://test.pypi.org/simple/ novatium

(Once released on PyPI: pip install novatium)

Quickstart

NovaRegressor

from novatium import NovaRegressor
from sklearn.datasets import load_diabetes
from sklearn.linear_model import LinearRegression
from sklearn.ensemble import RandomForestRegressor

X, y = load_diabetes(return_X_y=True)
model = NovaRegressor(base_model=LinearRegression(), delta_model=RandomForestRegressor(n_estimators=200, random_state=42))
model.fit(X, y)
pred = model.predict(X[:5])
print(pred)

NovaClassifier

from novatium import NovaClassifier
from sklearn.datasets import load_breast_cancer
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import GradientBoostingRegressor

X, y = load_breast_cancer(return_X_y=True)
clf = NovaClassifier(base_model=LogisticRegression(max_iter=1000), delta_model=GradientBoostingRegressor(random_state=42))
clf.fit(X, y)
proba = clf.predict_proba(X[:5])
print(proba)

Why Novatium?

• Drop-in Base+Delta pattern to uplift strong baselines.
• Sklearn-compatible: works with pipelines, grid search, etc.
• Domain-agnostic: tabular now; CV/NLP-ready APIs.

License

Apache-2.0 — see LICENSE.