!5888 add bert hub config
Merge pull request !5888 from yoonlee666/master
This commit is contained in:
commit
947247bbc5
|
@ -0,0 +1,77 @@
|
|||
# Copyright 2020 Huawei Technologies Co., Ltd
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
# ============================================================================
|
||||
'''
|
||||
Bert hub interface for bert base and bert nezha
|
||||
'''
|
||||
from src.bert_model import BertModel
|
||||
from src.bert_model import BertConfig
|
||||
import mindspore.common.dtype as mstype
|
||||
bert_net_cfg_base = BertConfig(
|
||||
batch_size=32,
|
||||
seq_length=128,
|
||||
vocab_size=21128,
|
||||
hidden_size=768,
|
||||
num_hidden_layers=12,
|
||||
num_attention_heads=12,
|
||||
intermediate_size=3072,
|
||||
hidden_act="gelu",
|
||||
hidden_dropout_prob=0.1,
|
||||
attention_probs_dropout_prob=0.1,
|
||||
max_position_embeddings=512,
|
||||
type_vocab_size=2,
|
||||
initializer_range=0.02,
|
||||
use_relative_positions=False,
|
||||
input_mask_from_dataset=True,
|
||||
token_type_ids_from_dataset=True,
|
||||
dtype=mstype.float32,
|
||||
compute_type=mstype.float16
|
||||
)
|
||||
bert_net_cfg_nezha = BertConfig(
|
||||
batch_size=32,
|
||||
seq_length=128,
|
||||
vocab_size=21128,
|
||||
hidden_size=1024,
|
||||
num_hidden_layers=24,
|
||||
num_attention_heads=16,
|
||||
intermediate_size=4096,
|
||||
hidden_act="gelu",
|
||||
hidden_dropout_prob=0.1,
|
||||
attention_probs_dropout_prob=0.1,
|
||||
max_position_embeddings=512,
|
||||
type_vocab_size=2,
|
||||
initializer_range=0.02,
|
||||
use_relative_positions=True,
|
||||
input_mask_from_dataset=True,
|
||||
token_type_ids_from_dataset=True,
|
||||
dtype=mstype.float32,
|
||||
compute_type=mstype.float16
|
||||
)
|
||||
def create_network(name, *args, **kwargs):
|
||||
'''
|
||||
Create bert network for base and nezha.
|
||||
'''
|
||||
if name == 'bert_base':
|
||||
if "batch_size" in kwargs:
|
||||
bert_net_cfg_base.batch_size = kwargs["batch_size"]
|
||||
if "seq_length" in kwargs:
|
||||
bert_net_cfg_base.seq_length = kwargs["seq_length"]
|
||||
return BertModel(bert_net_cfg_base, *args)
|
||||
if name == 'bert_nezha':
|
||||
if "batch_size" in kwargs:
|
||||
bert_net_cfg_nezha.batch_size = kwargs["batch_size"]
|
||||
if "seq_length" in kwargs:
|
||||
bert_net_cfg_nezha.seq_length = kwargs["seq_length"]
|
||||
return BertModel(bert_net_cfg_nezha, *args)
|
||||
raise NotImplementedError(f"{name} is not implemented in the repo")
|
Loading…
Reference in New Issue