Save grid_search performance metrics

cognoma · Sep 21, 2016 · eb1b670 · eb1b670
1 parent a050db0
commit eb1b670
Show file tree

Hide file tree

Showing 2 changed files with 1,269 additions and 1 deletion.
diff --git a/cognoml/analysis.py b/cognoml/analysis.py
@@ -54,7 +54,6 @@ def classify(sample_id, mutation_status, **kwargs):
     obs_df['testing'] = obs_df.sample_id.isin(X_test.index).astype(int)
 
     pipeline.fit(X=X_train, y=y_train)
-    #cv_score_df = grid_scores_to_df(clf_grid.grid_scores_)
 
     predict_df = pd.DataFrame.from_items([
         ('sample_id', X_whole.index),
@@ -95,6 +94,12 @@ def classify(sample_id, mutation_status, **kwargs):
     performance['cv'] = {'auroc': round(clf_grid.best_score_, 5)}
     results['performance'] = performance
 
+    gs = collections.OrderedDict()
+    gs['mean_scores'] = utils.mean_grid_scores_to_df(clf_grid.grid_scores_)
+    gs['fold_scores'] = utils.grid_scores_to_df(clf_grid.grid_scores_)
+    gs = utils.value_map(gs, utils.df_to_datatables)
+    results['grid_search'] = gs
+
     results['model'] = utils.model_info(clf_grid.best_estimator_)
 
     feature_df = utils.get_feature_df(clf_grid.best_estimator_, X.columns)