fix #808

2019-08-20 11:29:26 +02:00
parent 901dde0e45
commit 53c8f700f4
7 changed files with 20 additions and 1 deletions
--- a/pytorch_transformers/modeling_bert.py
+++ b/pytorch_transformers/modeling_bert.py
@@ -599,7 +599,10 @@ BERT_INPUTS_DOCSTRING = r"""
                ``tokens:         [CLS] the dog is hairy . [SEP]``
                
                ``token_type_ids:   0   0   0   0  0     0   0``
-    
+
+            Bert is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
            Indices can be obtained using :class:`pytorch_transformers.BertTokenizer`.
            See :func:`pytorch_transformers.PreTrainedTokenizer.encode` and
            :func:`pytorch_transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.