|
from transformers import PretrainedConfig |
|
|
|
class RankingPrompterConfig(PretrainedConfig): |
|
model_type = "umt5" |
|
|
|
def __init__( |
|
self, |
|
vocab_size=250112, |
|
d_model=512, |
|
d_kv=64, |
|
d_ff=1024, |
|
num_layers=8, |
|
num_decoder_layers=None, |
|
num_heads=6, |
|
relative_attention_num_buckets=32, |
|
relative_attention_max_distance=128, |
|
num_answer_query=128, |
|
dropout_rate=0.1, |
|
layer_norm_epsilon=1e-6, |
|
initializer_factor=1.0, |
|
feed_forward_proj="gated-gelu", |
|
is_encoder_decoder=True, |
|
use_cache=True, |
|
tokenizer_class="T5Tokenizer", |
|
tie_word_embeddings=True, |
|
pad_token_id=0, |
|
eos_token_id=1, |
|
decoder_start_token_id=0, |
|
classifier_dropout=0.0, |
|
**kwargs, |
|
): |
|
super().__init__( |
|
is_encoder_decoder=is_encoder_decoder, |
|
tokenizer_class=tokenizer_class, |
|
tie_word_embeddings=tie_word_embeddings, |
|
pad_token_id=pad_token_id, |
|
eos_token_id=eos_token_id, |
|
decoder_start_token_id=decoder_start_token_id, |
|
**kwargs, |
|
) |
|
self.vocab_size = vocab_size |
|
self.d_model = d_model |
|
self.d_kv = d_kv |
|
self.d_ff = d_ff |
|
self.num_layers = num_layers |
|
self.num_decoder_layers = ( |
|
num_decoder_layers if num_decoder_layers is not None else self.num_layers |
|
) |
|
self.num_heads = num_heads |
|
self.relative_attention_num_buckets = relative_attention_num_buckets |
|
self.relative_attention_max_distance = relative_attention_max_distance |
|
self.num_answer_query = num_answer_query |
|
self.dropout_rate = dropout_rate |
|
self.classifier_dropout = classifier_dropout |
|
self.layer_norm_epsilon = layer_norm_epsilon |
|
self.initializer_factor = initializer_factor |
|
self.feed_forward_proj = feed_forward_proj |
|
self.use_cache = use_cache |
|
|
|
act_info = self.feed_forward_proj.split("-") |
|
self.dense_act_fn = act_info[-1] |
|
self.is_gated_act = act_info[0] == "gated" |
|
|
|
if len(act_info) > 1 and act_info[0] != "gated" or len(act_info) > 2: |
|
raise ValueError( |
|
f"`feed_forward_proj`: {feed_forward_proj} is not a valid activation function of the dense layer." |
|
"Please make sure `feed_forward_proj` is of the format `gated-{ACT_FN}` or `{ACT_FN}`, e.g. " |
|
"'gated-gelu' or 'relu'" |
|
) |
|
|
|
if feed_forward_proj == "gated-gelu": |
|
self.dense_act_fn = "gelu_new" |
|
|
|
@property |
|
def hidden_size(self): |
|
return self.d_model |
|
|
|
@property |
|
def num_attention_heads(self): |
|
return self.num_heads |
|
|
|
@property |
|
def num_hidden_layers(self): |
|
return self.num_layers |
|
|
