Print name when sampling

glee2429 · Nov 10, 2020 · 0776650 · 0776650
1 parent 07edce4
commit 0776650
Showing 1 changed file with 1 addition and 0 deletions.
diff --git a/the_pile/pile.py b/the_pile/pile.py
@@ -315,6 +315,7 @@ def lang_stats(pile):
 def sample_from_sets(datasets, n_docs):
  random.seed(42)
  for dset, _ in datasets:
+ print(dset.name())
  fname = 'dataset_samples/{}.json'.format(dset.name().replace(' ', '_'))
  if os.path.exists(fname): continue