Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/autoliuweijie/K-BERT
/ functions
Functions
198 in github.com/autoliuweijie/K-BERT
⨍
Functions
198
◇
Types & classes
52
Method
__init__
(self, args, vocab, tokenizer)
uer/utils/data.py:809
Method
__init__
(self, args, dataset_path, batch_size, proc_id, proc_num, shuffle=False)
uer/utils/data.py:883
Method
__init__
(self, args, vocab, tokenizer)
uer/utils/data.py:947
Method
__init__
(self, args, dataset_path, batch_size, proc_id, proc_num, shuffle=False)
uer/utils/data.py:1119
Method
__init__
(self, args, vocab, tokenizer)
uer/utils/data.py:1185
Method
__init__
(self, args, dataset_path, batch_size, proc_id, proc_num, shuffle=False)
uer/utils/data.py:1264
Method
__init__
(self, params, lr, warmup=-1, t_total=-1, schedule='warmup_linear', b1=0.9, b2=0.999, e=1e-6,
uer/utils/optimizers.py:50
Method
__init__
(self, args)
uer/encoders/rnn_encoder.py:7
Method
__init__
(self, args)
uer/encoders/mixed_encoder.py:54
Method
__init__
(self, args)
uer/encoders/bert_encoder.py:13
Method
__init__
(self, args)
uer/encoders/birnn_encoder.py:7
Method
__init__
(self, args)
uer/encoders/gpt_encoder.py:14
Method
__init__
(self, args)
uer/encoders/attn_encoder.py:11
Method
__init__
(self, args)
uer/encoders/cnn_encoder.py:41
Method
__init__
(self, hidden_size, heads_num, dropout)
uer/layers/multi_headed_attn.py:11
Method
__init__
(self, args)
uer/layers/transformer.py:13
Method
__init__
(self, hidden_size, eps=1e-6)
uer/layers/layer_norm.py:7
Method
__init__
(self, args, vocab_size)
uer/layers/embeddings.py:12
Method
__init__
(self, hidden_size, feedforward_size)
uer/layers/position_ffn.py:8
Method
__init__
(self, args, vocab_size)
uer/subencoders/rnn_subencoder.py:6
Method
__init__
(self, args, vocab_size)
uer/subencoders/avg_subencoder.py:7
Method
__init__
(self, args, vocab_size)
uer/subencoders/cnn_subencoder.py:7
Method
__init__
(self, args, vocab_size)
uer/targets/bilm_target.py:13
Method
__init__
(self, args, vocab_size)
uer/targets/bert_target.py:14
Method
__init__
(self, args, vocab_size)
uer/targets/s2s_target.py:12
Method
__init__
(self, args, vocab_size)
uer/targets/mlm_target.py:14
Method
__init__
(self, args, vocab_size)
uer/targets/nsp_target.py:12
Method
__init__
(self, args, vocab_size)
uer/targets/lm_target.py:12
Method
__init__
(self, args, vocab_size)
uer/targets/cls_target.py:12
Method
__init__
(self, args, embedding, encoder, target)
uer/models/bert_model.py:12
Method
__init__
(self, args, embedding, encoder, target, subencoder = None)
uer/models/model.py:15
Method
__iter__
(self)
uer/utils/data.py:307
Method
__iter__
(self)
uer/utils/data.py:456
Method
__iter__
(self)
uer/utils/data.py:605
Method
__iter__
(self)
uer/utils/data.py:777
Method
__iter__
(self)
uer/utils/data.py:918
Method
__iter__
(self)
uer/utils/data.py:1154
Method
__iter__
(self)
uer/utils/data.py:1299
Method
__len__
(self)
uer/utils/vocab.py:52
Method
build
Build vocabulary from the given corpus.
uer/utils/vocab.py:101
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:88
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:350
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:497
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:649
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:819
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:956
Method
build_and_save
Build dataset from the given corpus. Start workers_num processes and each process deals with a part of data.
uer/utils/data.py:1194
Method
convert_ids_to_tokens
Converts a sequence of ids in wordpiece tokens using the vocab.
uer/utils/tokenizer.py:100
Method
convert_tokens_to_ids
Converts a sequence of tokens into ids using the vocab.
uer/utils/tokenizer.py:87
Method
forward
Args: src: [batch_size x seq_length] label: [batch_size x seq_length] mask: [batch_size x seq_length]
run_kbert_ner.py:32
Method
forward
Args: src: [batch_size x seq_length] label: [batch_size] mask: [batch_size x seq_length]
run_kbert_cls.py:39
Method
forward
(self, emb, seg)
uer/encoders/rnn_encoder.py:28
Method
forward
(self, emb, seg)
uer/encoders/rnn_encoder.py:65
Method
forward
(self, emb, seg)
uer/encoders/mixed_encoder.py:27
Method
forward
(self, emb, seg)
uer/encoders/mixed_encoder.py:74
Method
forward
Args: emb: [batch_size x seq_length x emb_size] seg: [batch_size x seq_length] vm: [batch_size x seq_leng
uer/encoders/bert_encoder.py:20
Method
forward
(self, emb, seg)
uer/encoders/birnn_encoder.py:29
Method
forward
Args: emb: [batch_size x seq_length x emb_size] seg: [batch_size x seq_length] Returns: hidden:
uer/encoders/gpt_encoder.py:21
Method
forward
Args: emb: [batch_size x seq_length x emb_size] mask: [batch_size x 1 x seq_length x seq_length] Returns:
uer/encoders/attn_encoder.py:24
Method
forward
(self, emb, seg)
uer/encoders/cnn_encoder.py:20
Method
forward
(self, emb, seg)
uer/encoders/cnn_encoder.py:57
Method
forward
Args: key: [batch_size x seq_length x hidden_size] value: [batch_size x seq_length x hidden_size] query:
uer/layers/multi_headed_attn.py:23
Method
forward
Args: hidden: [batch_size x seq_length x emb_size] mask: [batch_size x 1 x seq_length x seq_length] Returns:
uer/layers/transformer.py:29
Method
forward
(self, x)
uer/layers/layer_norm.py:13
Method
forward
(self, src, seg, pos=None)
uer/layers/embeddings.py:21
Method
forward
(self, x)
uer/layers/position_ffn.py:13
Method
forward
(self, ids)
uer/subencoders/rnn_subencoder.py:18
Method
forward
(self, ids)
uer/subencoders/avg_subencoder.py:11
Method
forward
(self, ids)
uer/subencoders/cnn_subencoder.py:14
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: [batch_size x seq_length] Returns:
uer/targets/bilm_target.py:21
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: tuple with tgt_mlm [batch_size x seq_length] and tgt_
uer/targets/bert_target.py:55
Method
forward
(self, memory_bank, tgt)
uer/targets/s2s_target.py:23
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: [batch_size x seq_length] Returns:
uer/targets/mlm_target.py:51
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: [batch_size] Returns: loss: Next
uer/targets/nsp_target.py:22
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: [batch_size x seq_length] Returns:
uer/targets/lm_target.py:20
Method
forward
Args: memory_bank: [batch_size x seq_length x hidden_size] tgt: [batch_size] Returns: loss: Clas
uer/targets/cls_target.py:23
Method
forward
(self, src, tgt_mlm, tgt_nsp, seg)
uer/models/bert_model.py:18
Method
forward
(self, src, tgt, seg, pos=None, vm=None)
uer/models/model.py:29
Method
get_lr
(self)
uer/utils/optimizers.py:70
Method
shape
(x)
uer/layers/multi_headed_attn.py:38
Method
tokenize
(self, text)
uer/utils/tokenizer.py:23
Method
tokenize
Splitting the sentence into words according to space.
uer/utils/tokenizer.py:32
Method
tokenize
(self, text)
uer/utils/tokenizer.py:77
Method
tokenize
Tokenizes a piece of text.
uer/utils/tokenizer.py:121
Method
tokenize
Tokenizes a piece of text into its word pieces. This uses a greedy longest-match-first algorithm to perform tokenization using the giv
uer/utils/tokenizer.py:232
Function
train_and_validate
(args)
uer/trainer.py:21
Function
train_bert
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:96
Function
train_bilm
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:254
Function
train_cls
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:335
Function
train_lm
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:183
Function
train_mlm
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:405
Function
train_nsp
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:477
Function
train_s2s
(args, gpu_id, rank, loader, model, optimizer)
uer/trainer.py:547
Method
unshape
(x)
uer/layers/multi_headed_attn.py:44
Function
warmup_constant
(x, warmup=0.002)
uer/utils/optimizers.py:18
Function
warmup_cosine
(x, warmup=0.002)
uer/utils/optimizers.py:13
Function
warmup_linear
(x, warmup=0.002)
uer/utils/optimizers.py:23
Method
worker
Worker that creates vocabulary from corpus[start:end].
uer/utils/vocab.py:55
← previous
101–198 of 198, ranked by callers