These operations should be delayed to avoid larger than memory issues

1b84b956 · Tiago de Freitas Pereira · ffa826bb · 1b84b956
Commit 1b84b956 authored Jul 06, 2021 by Tiago de Freitas Pereira
--- a/bob/bio/base/score/load.py
+++ b/bob/bio/base/score/load.py
@@ -165,7 +165,7 @@ def get_split_dataframe(filename):
    genuines = df[df.probe_subject_id == df.bio_ref_subject_id]
    impostors = df[df.probe_subject_id != df.bio_ref_subject_id]

-    return impostors.compute(), genuines.compute()
+    return impostors, genuines


 def split_csv_writer(filename):