THUDM P-tuning issues - Githubissues

THUDM / P-tuning

A novel method to tune language models. Codes and datasets for paper ``GPT understands, too''.

MIT License

924 stars 111 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

which paper is the first one to propose soft prompt

#51 lihuiliullh opened 9 months ago
1
LAMA link is bad

#50 TuuSiwei closed 2 days ago
1
python cli.py AttributeError: 'Namespace' object has no attribute '_get_node_flag'

#49 cqray1990 opened 1 year ago
0
LAMA数据集链接已经失效

#48 huzuoliang closed 1 year ago
1
Why discreteness of word embedding leads to the optimizer easily fall into local minima?

#47 xsc1234 opened 1 year ago
1
MASK and x_h reversed on bert

#46 wuflyh closed 1 year ago
0
The link to LAMA dataset is broken

#45 wuflyh opened 1 year ago
1
AttributeError: 'Namespace' object has no attribute '_get_node_flag'

#44 YCC-324 closed 1 year ago
1
why does LSTM can be discarded during inference?

#43 MXuer opened 1 year ago
1
P-tuning中模型参数冷冻的问题

#42 YCC-324 opened 1 year ago
2
fully-supervised learning那部分的实验的具体细节。

#41 DonceDace opened 1 year ago
0
Doesn't anyone see a problem with the code about the prompt construction for BERT-style transformer???

#40 lovekittynine opened 2 years ago
2
questions about discreteness in optimization

#39 skygl closed 2 years ago
2
Tuning joe

#38 Joeplus1234 closed 2 years ago
0
lstm的enbedding_dim与预训练模型的embedding_dim不一致的报错？

#37 SCU-JJkinging closed 2 years ago
7
loss, logits = output.loss, output.logits AttributeError: 'tuple' object has no attribute 'loss'

#36 Claymore715 closed 2 years ago
1
关于p-tuning中训练loss变为nan

#35 effort-yq closed 2 years ago
1
P-tuning如何用于Question answering

#34 yuhangjiang22 closed 2 years ago
1
Inconsistent SuperGLUE Results from P-Tuning and P-TuningV2 Paper

#33 theoqian closed 2 years ago
2
fewGLUE中采用的template format

#32 Sihan-Song opened 2 years ago
0
original, lama, or shared?

#31 One-punch24 closed 2 years ago
0
Can we add our custimized classification task (such as other GLUE tasks) into this package?

#30 CSerxy opened 3 years ago
0
what is the cudatoolkit you used to match the requirements.txt

#29 CSerxy closed 3 years ago
1
Any illustration for the installed package/environment?

#28 CSerxy closed 3 years ago
2
词表问题

#27 shaoyangxu closed 2 years ago
2
Code about the implement of gpts

#26 Deonmi opened 3 years ago
0
Question about the prompt encoder of PT-Fewshot

#25 ikm565 closed 3 years ago
2
BERT output is a tuple (in LAMA)

#24 olenmg closed 3 years ago
3
能否测试一下小样本学习基准-中文版(FewCLUE)的效果?

#23 brightmart closed 3 years ago
2
请问LAMA中每个关系都需要训练一个单独的PromptEncoder吗？

#22 superqing001 closed 3 years ago
1
Typo in code (will cause the prompt not use the warmup

#21 yhcc closed 3 years ago
2
Bug in loss recording for LAMA

#20 lxuechen closed 3 years ago
1
I have a question about prompt encoder's inputs.

#19 sooftware closed 3 years ago
1
P-tuuning的一些问题

#18 Ming-Qin-tech opened 3 years ago
4
gpt2-medium LAMA

#17 casually-PYlearner opened 3 years ago
2
AttributeError: 'PTuneForLAMA' object has no attribute 'prompt'

#16 puraminy opened 3 years ago
1
Few-shot SuperGLUE的部分数据集效果复现问题

#15 Riroaki opened 3 years ago
6
Where is the GPT-based implementation on SuperGLUE?

#14 shizhediao opened 3 years ago
2
Potential bug in learning rate schedule for few-shot

#13 lxuechen closed 3 years ago
1
few-shot实验encoder换成bert-base-cased效果差很多

#12 Life-0-1 closed 3 years ago
2
Prompt format selection

#11 lxuechen closed 3 years ago
1
Is LAMA p-tuning seeking a global prompt representation across relation?

#10 asahi417 closed 3 years ago
2
multi-label classification

#9 JaheimLee closed 3 years ago
1
Fully-supervised SuperGLUE

#8 zjujh1995 closed 3 years ago
2
Few-shot NLU: learning rate for model parameters vs. embedding parameters

#7 nelson-liu closed 3 years ago
2
Prompt Length in SuperGLUE

#6 Shamdan17 closed 3 years ago
1
关于论文的一些疑惑

#5 bojone closed 3 years ago
7
Question on the optimization process

#4 Robinysh closed 3 years ago
1
Nice work! when would you release the code?

#3 cooelf closed 3 years ago
2
A problem about the prompt

#2 logoutAgain closed 3 years ago
3