Code Monkey home page Code Monkey logo

lit-ie's Issues

UIE的效果一般

uie的loss极小但是表现一般,是UIE的设置不对吗还是性能就这样?全部使用默认参数和提供的example data:
image

跑提供的实体识别中的crf的例子,p,r,f1都是0

你好,我使用这个项目中提供的crf.sh跑实体识别,跑了大概8个epoch,p,r,f1都是0,使用的数据也是项目中提供的,换成span方式是正常的。另外,选择softmax或者cascade_crf方式时是不是只需要修改脚本中的TASK_NAME为对应的名称就行了,我修改之后报以下错误:
Traceback (most recent call last):
File "main.py", line 33, in
main()
File "main.py", line 27, in main
model.finetune(data_args)
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/models/base.py", line 71, in finetune
self.engine = self.create_engine()
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/models/ner.py", line 14, in create_engine
return NerEngine(
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/engines/ner.py", line 24, in init
super().init(model_type, task_model_name, model_config_kwargs=model_config_kwargs, **kwargs)
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/engines/base.py", line 80, in init
self.initialize_model(self.pretrained_model_name_or_path, config, model)
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/engines/base.py", line 118, in initialize_model
model = self.get_auto_model(self.model_type, self.task_model_name)
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/engines/ner.py", line 32, in get_auto_model
return AutoNerTaskModel.create(
File "/home/nlp_service/anaconda3/envs/uie/lib/python3.8/site-packages/litie/nn/ner/init.py", line 34, in create
return cls.registry[class_key](model_type, **kwargs)
TypeError: get_auto_cascade_crf_ner_model() got an unexpected keyword argument 'base_model'

请教下efficient global pointer在超长矩阵下的效率问题

苏神,请教下,我现在需要处理的序列是4096长度的,在bert上接入了efficient global pointer结构,26分类,结果非常慢,debug看了下有两个问题:

  1. 内存大小从20G->75G, 设置如下:
self.global_pointer = EfficientGlobalPointer(
    768,
    self.num_labels,
    64,
    use_rope=True
)
  1. 因为内部代码结构的问题,实现的是稠密loss

请问如果换做sparse矩阵加速会很多吗?在超长序列下EGP是否有效率问题?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.