1b5a498c2c0283a27428861255743ba7023ca0a3,fairseq/models/fconv.py,,fconv_wmt_en_de,#Any#,490

Before Change

@register_model_architecture("fconv", "fconv_wmt_en_de")
def fconv_wmt_en_de(args):
    base_architecture(args)
    convs = "[(512, 3)] * 9"       // first 9 layers have 512 units
    convs += " + [(1024, 3)] * 4"  // next 4 layers have 1024 units
    convs += " + [(2048, 1)] * 2"  // final 2 layers use 1x1 convolutions
    args.encoder_embed_dim = 768

After Change

    convs += " + [(1024, 3)] * 4"  // next 4 layers have 1024 units
    convs += " + [(2048, 1)] * 2"  // final 2 layers use 1x1 convolutions

    args.encoder_embed_dim = getattr(args, "encoder_embed_dim", 768)
    args.encoder_layers = getattr(args, "encoder_layers", convs)
    args.decoder_embed_dim = getattr(args, "decoder_embed_dim", 768)
    args.decoder_layers = getattr(args, "decoder_layers", convs)
    args.decoder_out_embed_dim = getattr(args, "decoder_out_embed_dim", 512)
    base_architecture(args)


@register_model_architecture("fconv", "fconv_wmt_en_fr")
def fconv_wmt_en_fr(args):

Instances