microsoft / list Goto Github PK

View Code? Open in Web Editor NEW

29.0 6.0 7.0 33.53 MB

Lite Self-Training

License: MIT License

Shell 2.69% Python 97.31%

nlp liteweight machine-learning-algorithms few-shot-learning nlu pytorch deep-learning prompt-based

list's People

Contributors

Stargazers

Watchers

Forkers

pqmsoft1 martin6336 liangbin123 sdc777 jianzhu test-mass-forker-org-1 ycahn82

list's Issues

What is the LICENSE on this code ?

Hi, could you please include a LICENSE file with this code ?

This is great work and I really enjoyed going through your paper. Are there next steps that you'll are working on ?

This repo is missing important files

There are important files that Microsoft projects should all have that are not present in this repository. A pull request has been opened to add the missing file(s). When the pr is merged this issue will be closed automatically.

Microsoft teams can learn more about this effort and share feedback within the open source guidance available internally.

Merge this pull request

请教代码中问题

LiST/src/trainer.py

Lines 1335 to 1341 in 4d461bc

    
           if un_inputs is not None and (not finetune): 
        
               try: 
        
                   tr_loss += self.training_step(self.model,  un_inputs=un_inputs) 
        
               except: 
        
                   print("One error here") 
        
                   self.clear_memory() 
        
               tr_loss += self.training_step(self.model, inputs=inputs)

尊敬的作者，您好，请问上面这段代码中，在warmup student 的同时，为什么也要同时在有标签的数据上train student，这样混合的训练，与论文中的算法流程好像有别？请问是我的理解错了，还是确实代码有问题？如果代码有问题，请问如何修改？谢谢

microsoft / list Goto Github PK

list's People

Contributors

Stargazers

Watchers

Forkers

list's Issues

What is the LICENSE on this code ?

This repo is missing important files

请教代码中问题

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	if un_inputs is not None and (not finetune):
	try:
	tr_loss += self.training_step(self.model, un_inputs=un_inputs)
	except:
	print("One error here")
	self.clear_memory()
	tr_loss += self.training_step(self.model, inputs=inputs)