entity_extractor_by_ner icon indicating copy to clipboard operation
entity_extractor_by_ner copied to clipboard

基于Tensorflow2.3开发的NER模型,都是CRF范式,包含Bilstm(IDCNN)-CRF、Bert-Bilstm(IDCNN)-CRF、Bert-CRF,可微调预训练模型,可对抗学习,用于命名实体识别,配置后可直接运行。

Results 12 entity_extractor_by_ner issues
Sort by recently updated
recently updated
newest added

您好 再更换自己数据集从头训练·时,固定在2%的位置跳这个错 但是不是很理解原因 劳请提点一下 2%|▏ | 20/1022 [00:18

您好!我改成mode=train训练模式之后,在训练过程中就报错 ![image](https://user-images.githubusercontent.com/63388342/190111857-c9c247d3-018d-4955-98ea-9f92a8b150b0.png)

你好,请问训练自己的数据集时,checkpoints_dir和checkpoints_name要怎么改呢

Bumps [transformers](https://github.com/huggingface/transformers) from 4.6.1 to 4.36.0. Release notes Sourced from transformers's releases. v4.36: Mixtral, Llava/BakLlava, SeamlessM4T v2, AMD ROCm, F.sdpa wide-spread support New model additions Mixtral Mixtral is the new...

dependencies

我下载了英文的预训练bert模型,并且把我自己的英文数据集处理了格式之后,放上去训练结果全是0,是不支持英文吗 ![image](https://github.com/stanleylsx/entity_extractor_by_ner/assets/102421327/7e1ab77f-754a-4a3d-a0d9-422094f81d81)

The problem is in the file engines/data.py line 108. old: ``` tokens = [tokens for token in tokens if token if token not in [' ', '']] ``` modified: ```...

![Snipaste_2024-05-10_21-19-22](https://github.com/stanleylsx/entity_extractor_by_ner/assets/115866701/494c574b-0548-43ed-bdea-6dcd034883b9)

这个papers的文件夹里面的PDF文件是用来干什么的呢

from tensorflow_addons.text.crf import crf_decode 这个导入的时候报错,是因为什么呢,而且文件夹里也没有这个文件