[Example] Update README of transformers for multi-gpu support (dmlc#1435

)

* Update README.md

* Update README.md

examples/pytorch/transformer/README.md

@@ -1,7 +1,7 @@
 # Transformer in DGL
-In this example we implement the [Transformer](https://arxiv.org/pdf/1706.03762.pdf) and [Universal Transformer](https://arxiv.org/abs/1807.03819) with ACT in DGL.
+In this example we implement the [Transformer](https://arxiv.org/pdf/1706.03762.pdf) with ACT in DGL.
 
-The folder contains training module and inferencing module (beam decoder) for Transformer and training module for Universal Transformer
+The folder contains training module and inferencing module (beam decoder) for Transformer.
 
 ## Dependencies
 
@@ -18,6 +18,8 @@ The folder contains training module and inferencing module (beam decoder) for Tr
     python3 translation_train.py [--gpus id1,id2,...] [--N #layers] [--dataset DATASET] [--batch BATCHSIZE] [--universal]
     ```
 
+By specifying multiple gpu ids separated by comma, we will employ multi-gpu training with multiprocessing.
+
 - For evaluating BLEU score on test set(by enabling `--print` to see translated text):
 
     ```
@@ -28,19 +30,9 @@ Available datasets: `copy`, `sort`, `wmt14`, `multi30k`(default).
 
 ## Test Results
 
-### Transformer
-
 - Multi30k: we achieve BLEU score 35.41 with default setting on Multi30k dataset, without using pre-trained embeddings. (if we set the number of layers to 2, the BLEU score could reach 36.45).
 - WMT14: work in progress 
 
-### Universal Transformer
-
-- work in progress 
-
-## Notes
-
-- Currently we do not support Multi-GPU training(this will be fixed soon), you should only specify only one gpu\_id when running the training script.
-
 ## Reference
 
 - [The Annotated Transformer](http://nlp.seas.harvard.edu/2018/04/03/attention.html)