def test_OneHotEncoder(self): df = get_baseball_df() p = Pipeline([('e', OneHotEncoder(columns=['League']))]) df = p.transform(df) info(df)
def test_ScopedTransformer(self): df = get_baseball_df() p = Pipeline([('e', ScopedTransformer(transformer=MinMaxScaler(), columns=['RS']))]) df = p.fit_transform(df, []) info(df)
def test_scatter_matrix(self): df = get_baseball_df() scatter_matrix(df)
import pandas as pd from litkit.dash.viewer import view from litkit.data import get_baseball_df from litkit.inspect import info p = Pipeline([('estimator', Lasso())]) pg = [{'estimator': [Lasso(), LinearRegression(), ElasticNet(), Ridge()]}] gs = GridSearchCV(estimator=p, param_grid=pg, n_jobs=1, return_train_score=True) df = get_baseball_df() y = df['RS'] X = df[['RA', 'W', 'OBP', 'SLG']] gs.fit(X, y) r_df = pd.DataFrame(gs.cv_results_) del r_df['params'] r_df['param_estimator'] = r_df['param_estimator'].apply( lambda x: x.__class__.__name__) view(r_df) pp(p.named_steps["estimator"])
def test_info(self): df = get_baseball_df() info(df) info(df.values)