modelzoo.transformers.pytorch.bert.bert_pretrain_models.BertMLMHead

class modelzoo.transformers.pytorch.bert.bert_pretrain_models.BertMLMHead

Bases: torch.nn.Module

Methods

forward
get_output_embeddings
reset_parameters

__call__(*args: Any, **kwargs: Any) → Any

Call self as a function.

__init__(hidden_size=768, vocab_size=30522, embedding_size=None, use_ffn_bias_in_mlm=True, use_output_bias_in_mlm=True, activation='gelu', layer_norm_epsilon=1e-05, dropout=None, kernel_initializer='xavier_uniform')

static __new__(cls, *args: Any, **kwargs: Any) → Any

previous

modelzoo.transformers.pytorch.bert.bert_pretrain_models.BertClassifierHead

next

modelzoo.transformers.pytorch.bert.bert_pretrain_models.BertMLMHeadTransform