File size: 2,100 Bytes
319aff6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
from transformers import PretrainedConfig
class DebertaV2Config(PretrainedConfig):
def __init__(
self,
vocab_size=128100,
hidden_size=1536,
sep_token_id=2,
mask_token_id=128000,
num_hidden_layers=24,
num_attention_heads=24,
intermediate_size=6144,
hidden_act="gelu",
hidden_dropout_prob=0.1,
attention_probs_dropout_prob=0.1,
max_position_embeddings=512,
type_vocab_size=0,
initializer_range=0.02,
layer_norm_eps=1e-7,
relative_attention=False,
max_relative_positions=-1,
pad_token_id=0,
position_biased_input=True,
pos_att_type=None,
pooler_dropout=0,
pooler_hidden_act="gelu",
**kwargs,
):
super().__init__(**kwargs)
self.hidden_size = hidden_size
self.mask_token_id = mask_token_id
self.sep_token_id = sep_token_id
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.intermediate_size = intermediate_size
self.hidden_act = hidden_act
self.hidden_dropout_prob = hidden_dropout_prob
self.attention_probs_dropout_prob = attention_probs_dropout_prob
self.max_position_embeddings = max_position_embeddings
self.type_vocab_size = type_vocab_size
self.initializer_range = initializer_range
self.relative_attention = relative_attention
self.max_relative_positions = max_relative_positions
self.pad_token_id = pad_token_id
self.position_biased_input = position_biased_input
# Backwards compatibility
if isinstance(pos_att_type, str):
pos_att_type = [x.strip() for x in pos_att_type.lower().split("|")]
self.pos_att_type = pos_att_type
self.vocab_size = vocab_size
self.layer_norm_eps = layer_norm_eps
self.pooler_hidden_size = kwargs.get("pooler_hidden_size", hidden_size)
self.pooler_dropout = pooler_dropout
self.pooler_hidden_act = pooler_hidden_act
|