Add code for "Pay Less Attention with Lightweight and Dynamic Convolutions" (#473)

Summary:
Changelog:
- `e330f56`: Add code for the "Pay Less Attention with Lightweight and Dynamic Convolutions" paper
- `5e3b98c`: Add scripts for computing tokenized BLEU with compound splitting and sacrebleu
- update READMEs
- misc fixes
Pull Request resolved: https://github.com/pytorch/fairseq/pull/473

Differential Revision: D13819717

Pulled By: myleott

fbshipit-source-id: f2dc12ea89a436b950cafec3593ed1b04af808e9

train.py

@@ -328,6 +328,9 @@ def save_checkpoint(args, trainer, epoch_itr, val_loss):
 def load_checkpoint(args, trainer, epoch_itr):
     """Load a checkpoint and replay dataloader to match."""
     os.makedirs(args.save_dir, exist_ok=True)
+    if os.path.isabs(args.restore_file):
+        checkpoint_path = args.restore_file
+    else:
         checkpoint_path = os.path.join(args.save_dir, args.restore_file)
     if os.path.isfile(checkpoint_path):
         extra_state = trainer.load_checkpoint(checkpoint_path, args.reset_optimizer, args.reset_lr_scheduler,
@@ -344,6 +347,8 @@ def load_checkpoint(args, trainer, epoch_itr):
             if 'best' in extra_state:
                 save_checkpoint.best = extra_state['best']
         return True
+    else:
+        print('| no existing checkpoint found {}'.format(checkpoint_path))
     return False