issues
search
lonePatient
/
BERT-NER-Pytorch
Chinese NER(Named Entity Recognition) using BERT(Softmax, CRF, Span)
MIT License
2.1k
stars
427
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
ValueError: This scheme is not supported: {<Prefix.I: 1>, <Prefix.B: 4>, <Prefix.S: 16>, <Prefix.O: 2>}
#107
Lowryyyy
opened
1 month ago
0
实体嵌套是只能解同类型的实体嵌套问题吗？
#106
yuye2133
opened
7 months ago
0
为啥我下载的代码里面models里面没有transformers
#105
SNW12
opened
7 months ago
1
LOC Entity结果全是0是怎么回事，其他的是正常的？
#104
1693146833
opened
8 months ago
0
关于跑出来loss为nan问题
#103
xiaoyingwang055
opened
9 months ago
1
一个关于参数的问题
#102
pppeninsula
opened
11 months ago
5
疑惑
#101
SongBaoru
opened
11 months ago
2
如果我需要添加或修改标签的话，需要在哪些地方进行操作？
#100
Violettttee
opened
11 months ago
3
请问如何用训练好的模型来做文本的预测？
#99
Violettttee
opened
11 months ago
0
终端输入运行指令后提示[Errno 22] Invalid argument:…………，似乎是日志处file地址的问题，请问如何解决？
#98
Violettttee
closed
11 months ago
1
在预测的时候遇到ValueError: expected sequence of length 517 at dim 1 (got 512)怎么办，下面是详细报错内容
#97
huangshuai-ustc
opened
1 year ago
0
自定义数据集训练的话， 是不是只需要把数据集做成cluener的样子，放在dataset/cluener文件就可以了，需不需要为自己的数据集重写processor？
#96
LXXiaogege
opened
1 year ago
4
Cner数据集
#95
LinXiaojie1998
opened
1 year ago
0
目前最好的效果是基于 globalpointer 的做法吗
#94
652994331
opened
1 year ago
1
Problem：num_samples should be a positive integer value, but got num_samples=0
#93
git-saner
opened
1 year ago
2
softmax和crf性能差距很大，参数一致，请教如何看原因调效果？
#92
northfun
opened
1 year ago
0
英文分词
#91
jenfung
opened
2 years ago
5
bert_extract_item(start_logits, end_logits) 看抽取方式是不支持嵌套实体的？
#90
DSXiangLi
opened
2 years ago
1
problem
#89
langzigaoi
opened
2 years ago
0
请问required parameters的list是什么呢？
#88
xiaohou1112
opened
2 years ago
1
Fix do_lower_case
#87
entropy2333
closed
1 year ago
0
关于tokenizer.tokenize的疑问
#86
lsx0930
opened
2 years ago
1
label标签疑问
#85
SCU-JJkinging
opened
2 years ago
1
Some weights of the model checkpoint at/ BERT-NER-Pytorch-master/prev_trained_model/bert-base-chinese were not used when initializing BertCrfForNe
#84
zacharykzhao
opened
2 years ago
1
tokenizer.tokenize()问题
#83
472027909
closed
2 years ago
1
BERT-SPAN模型中SPAN输出的标签1具有严重的稀疏性
#82
milktang
closed
2 years ago
0
为什么数据经过DataLoader后变短了
#81
wulalahalala
opened
2 years ago
1
callback目录
#80
liuweishuo
opened
2 years ago
0
prev_trained_model 预训练模型下载
#79
nanggn
opened
2 years ago
1
为什么所有指标为0
#78
LastMonody
opened
2 years ago
4
我想问下您写的这个模型发的论文是那一篇我想看看 引用 谢谢了
#77
Kissingbymodi
opened
2 years ago
0
cner测试集与验证集中CONT类型的precision与recall均为0
#76
yangjianxin1
closed
2 years ago
1
您好，我想知道如何利用训练结果去预测新数据
#75
haogegenizhenbang
opened
2 years ago
9
新手求教！是否可以bert+crf+focalloss呢？
#74
deidei266
opened
2 years ago
0
在train过程中注释掉了evaluate 会影响模型的结果
#73
nicholasbao
opened
2 years ago
1
作NER任务求分类损失的时候为啥【clc】标签忽略了，而【seq】没有忽略？
#72
cqray1990
opened
2 years ago
0
这里的spanbert 为啥预训练还用bert的？这个不是原版的spanbert吧？
#71
cqray1990
opened
2 years ago
1
cuda10.2 pytorch 1.8 can run this code
#70
cqray1990
opened
2 years ago
0
关于crf的问题
#69
zhijiezhong
opened
2 years ago
0
总觉 在bert span中start位置提供的信息结合倒end的方式有问题。
#68
orangetwo
opened
2 years ago
1
softmax
#67
GekiKaraa
closed
2 years ago
0
span是指什么？
#66
milktang
opened
2 years ago
0
代码似乎存在一些bug，span标签会错位
#65
xujunrt
opened
2 years ago
1
How to enforce constraint on CRF transition matrix?
#64
lkqnaruto
opened
2 years ago
0
求大佬指点，如何设置命令行参数？为什么每次运行都报错？
#63
wones
closed
2 years ago
0
CRF.forward问题
#62
zhoujiang2013
closed
2 years ago
0
the following arguments are required: --task_name, --data_dir, --model_type, --model_name_or_path, --output_dir
#61
shidonghao-aki
opened
3 years ago
2
可以弄个运行实例吗？
#60
nlpCSir
opened
3 years ago
1
BertSoftmaxForNer do not use Softmax ？
#59
lj976264709
closed
3 years ago
0
准确率为0
#58
FangWu2
opened
3 years ago
3
Next