def regress_predict(exp, field, model): pred = model.predict(exp.data) df = pd.DataFrame({ 'Y_PRED': pred, 'Y_TRUE': exp.sample_metadata[field].values, 'SAMPLE': exp.sample_metadata[field].index.values, 'CV': 0 }) plot_scatter(df, cv=False) return df
def test_plot_scatter(self): res = pd.read_table(join(self.test_data_dir, 'diabetes_pred.txt'), index_col=0) title = 'foo' ax = plot_scatter(res, title=title) self.assertEqual(title, ax.get_title()) cor = 'r=-0.62 p-value=0.078' self.assertEqual(cor, ax.texts[0].get_text()) dots = [] for collection in ax.collections: dots.append(collection.get_offsets()) assert_array_equal(np.concatenate(dots, axis=0), res[['Y_TRUE', 'Y_PRED']].values)