# hugbert

**Repository Path**: shenhuaze/hugbert

## Basic Information

- **Project Name**: hugbert
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 1
- **Forks**: 0
- **Created**: 2021-01-21
- **Last Updated**: 2021-10-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# hugbert
My misc testing/tutorial/helper scripts for [Huggingface transformers](https://github.com/huggingface/transformers). Let's hug BERT together. 

## Articles

* [【HugBert01】Huggingface Transformers，一个顶级自然语言处理框架 - 知乎](https://zhuanlan.zhihu.com/p/141527015)
* [【HugBert02】热身运动：安装及向量编码 - 知乎](https://zhuanlan.zhihu.com/p/143161582)
* [【HugBert03】最后一英里：下游NLP任务的微调 - 知乎](https://zhuanlan.zhihu.com/p/149779660)
* [【HugBert04】GLUE：BERT类模型的通用评估基准 - 知乎](https://zhuanlan.zhihu.com/p/151818251)
* [【HugBert05】照猫画虎：理解from_pretrained，攒个模型下载器 - 知乎](https://zhuanlan.zhihu.com/p/158974273)
* [【HugBert06】避虚击实：中文命名实体识别任务微调 - 知乎](https://zhuanlan.zhihu.com/p/164966421)
* [【HugBert07】提纲挈领：中文文本摘要任务微调 - 知乎](https://zhuanlan.zhihu.com/p/168719940)

持续更新中。

## Scripts

* [hf_model_downloader.py](https://github.com/hspuppy/hugbert/blob/master/hf_model_downloader.py): given a model short name, list and/or download model files.
* add ner_dataset folder, for some public Chinese NER data