MCPcopy Create free account
hub / github.com/HA-Transformer/MAT / __init__

Method __init__

fairseq/models/fconv_self_att.py:276–379  ·  view source on GitHub ↗
(
        self, dictionary, embed_dim=512, out_embed_dim=256, max_positions=1024,
        convolutions=((512, 3),) * 8, attention=True, dropout=0.1,
        selfattention=False, attention_nheads=1, selfattention_nheads=1,
        project_input=False, gated_attention=False, downsample=False,
        pretrained=False, trained_decoder=None,
    )

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls 9

LayerNormFunction · 0.90
SelfAttentionClass · 0.85
padMethod · 0.80
EmbeddingFunction · 0.70
PositionalEmbeddingFunction · 0.70
LinearFunction · 0.70
LinearizedConv1dFunction · 0.70
__init__Method · 0.45

Tested by

no test coverage detected