Add option to disable positional embeddings in TransformerModel (#421)

Summary:
Add argument `--no-token-positional-embeddings` to TransformerModel (currently only available in TransformerLanguageModel) to disable positional embeddings.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/421

Differential Revision: D13548450

Pulled By: myleott

fbshipit-source-id: b352c702ed1609e3b84d9a8404941d3274a7f883

fairseq/models/transformer.py

@@ -89,6 +89,8 @@ class TransformerModel(FairseqModel):
         parser.add_argument('--share-all-embeddings', action='store_true',
                             help='share encoder, decoder and output embeddings'
                                  ' (requires shared dictionary and embed dim)')
+        parser.add_argument('--no-token-positional-embeddings', default=False, action='store_true',
+                            help='if set, disables positional embeddings (outside self attention)')
         parser.add_argument('--adaptive-softmax-cutoff', metavar='EXPR',
                             help='comma separated list of adaptive softmax cutoff points. '
                                  'Must be used with adaptive_loss criterion'),