neptune-sklearn 0.9.5

Neptune.ai scikit-learn integration library

Neptune: sklearn integration

See the official docs.

Minimal example:

from sklearn.datasets import load_boston
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
import neptune.new as neptune
import neptune.new.integrations.sklearn as npt_utils

run = neptune.init(project='common/sklearn-integration',
                   api_token='ANONYMOUS',
                   name='regression-example',
                   tags=['RandomForestRegressor', 'regression'])

parameters = {'n_estimators': 70,
              'max_depth': 7,
              'min_samples_split': 3}


rfr = RandomForestRegressor(**parameters)

X, y = load_boston(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20, random_state=28743)

rfr.fit(X_train, y_train)

run['rfr_summary'] = npt_utils.create_regressor_summary(rfr, X_train, X_test, y_train, y_test)