hummingbird
/
fastNLP
Not watched
2
0
Fork 0
Code
Releases 13 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
949 Commits
3 Branches
25 MB
0 Download
 
 
 
 
 
Tree: d1f531c049
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'd1f531c049'
${ noResults }
fastNLP/reproduction
xuyige d1f531c049 update matching dataloader in reproduction/matching 5 years ago
..
Biaffine_parser 1. 删除Trainer中对train_data必须为DataSet的assert 5 years ago
CNN-sentence_classification * fixing unit tests 6 years ago
Char-aware_NLM remove unused codes; add more tests 6 years ago
Chinese_word_segmentation Dev0.4.0 (#149) 5 years ago
HAN-document_classification optimize CWS example 6 years ago
LSTM+self_attention_sentiment_analysis Dev0.4.0 (#149) 5 years ago
POS_tagging 1. 删除Trainer中对train_data必须为DataSet的assert 5 years ago
Star_transformer 1. 删除Trainer中对train_data必须为DataSet的assert 5 years ago
matching update matching dataloader in reproduction/matching 5 years ago
seqence_labelling 1. 在vocabulary的from_dataset中增加no_create_entry_dataset选项，用于传递dev和test 5 years ago
text_classification add TC/MTL16Loader 5 years ago
README.md update reproduction/README.md 5 years ago
__init__.py 1. 将pad的功能从FieldArray中剥离出来，使用Padder完成各种padding操作。 5 years ago
utils.py 1. 适配将Batch修改为pytorch的DataLoader的修改 5 years ago

README.md

模型复现

这里复现了在fastNLP中实现的模型，旨在达到与论文中相符的性能。

复现的模型有:

任务复现

Text Classification (文本分类)

  • still in progress

Matching (自然语言推理/句子匹配)

  • still in progress

Sequence Labeling (序列标注)

  • still in progress

Coreference resolution (指代消解)

  • still in progress

Summarization (摘要)

  • still in progress

...

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

一款轻量级的自然语言处理（NLP）工具包，目标是减少用户项目中的工程型代码，例如数据处理循环、训练循环、多卡运行等

Python Jupyter Notebook Text CSV Markdown