Example #1
0
def base_architecture(args):
    args.compressed = getattr(args, "compressed", 4)
    args.shared_kv_compressed = getattr(args, "shared_kv_compressed", 0)
    args.shared_layer_kv_compressed = getattr(args,
                                              "shared_layer_kv_compressed", 0)
    args.freeze_compress = getattr(args, "freeze_compress", 0)
    roberta_base_architecture(args)
Example #2
0
def xlm_align_base(args):
  roberta_base_architecture(args)
Example #3
0
def reload_roberta_base(args):
    roberta_base_architecture(args)
Example #4
0
def mpnet_base_architecture(args):
    roberta_base_architecture(args)
Example #5
0
def infoxlm_base(args):
  roberta_base_architecture(args)
Example #6
0
def model_parallel_roberta_postnorm_architecture(args):
    # the original BERT/RoBERTa uses the "Post-LN" formulation
    roberta_base_architecture(args)