XiangLi1999 PrefixTuning issues

XiangLi1999 / PrefixTuning

Prefix-Tuning: Optimizing Continuous Prompts for Generation

887 stars 161 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

The paper mentions the ability to train multiple prefixes in a single batch.

#53 phalexo opened 1 month ago
0
这个代码有没有集成到流行的代码仓里面啊

#52 jinec opened 3 months ago
0
Need to update many things in the repository.

#51 kevinscaria opened 4 months ago
0
pushing for colab

#50 andrebiehl closed 5 months ago
0
which paper is the first one to propose soft prompt

#49 lihuiliullh opened 8 months ago
0
Data issue and file location query

#48 JamZhangW opened 1 year ago
1
License

#47 Timo-Schrader opened 1 year ago
0
This is a relatively bad project because some strange packages were used

#46 mshmoon opened 1 year ago
0
Question: Do we need a supervised dataset with X & Y pairs?

#45 karrtikiyer opened 1 year ago
0
How to do inference with prefix tuning?

#44 JiaxiLi001 closed 5 months ago
0
Which version of the Transformers did you modify? v3.2.0?

#43 Ant0082 opened 2 years ago
4
Data preparation step

#42 mmderakhshani closed 2 years ago
1
control code is not used in PrefixTuning.get_prompt()

#41 XinyuGuan01 opened 2 years ago
3
Is it necessary to arrange position ids between [prefix_len, prefix_len+seq_len) ?

#40 baiyuting opened 2 years ago
2
question about the initialization experiment

#39 Tsingularity opened 2 years ago
0
notation typo in the paper

#38 Tsingularity opened 2 years ago
0
Should've mentioned about "CRITICAL" modifications done in transformers source code

#37 sonsus opened 2 years ago
0
test

#36 ArrowOoops closed 2 years ago
0
AttributeError: 'tuple' object has no attribute 'detach'

#35 YuffieHuang opened 2 years ago
0
Hi, What do function names get_prompt_p3,5,7 ... mean?

#34 LeeJiangWei opened 2 years ago
0
fix path name error

#33 yahoo17 opened 2 years ago
0
Use --init_shallow_word for seq2seq model

#32 JaniceXiong opened 2 years ago
1
Understanding the Seq2Seq Encoder-Decoder Prefix Implementation

#31 rajaswa opened 2 years ago
11
About the evaluation scripts

#30 Jiuzhouh opened 2 years ago
4
The version of pytorch_lightning

#29 Jamesswang opened 2 years ago
4
About the results on lowdata table2text task.

#28 insomnia1996 opened 2 years ago
0
OSError: [Errno 30] Read-only file system: '/u'

#27 wanglec closed 2 years ago
1
TypeError: setup() got an unexpected keyword argument 'stage'

#26 YahooHu closed 2 years ago
1
A piece of GPU can train this model?

#25 lrongzheni opened 2 years ago
4
GPT 2 prefix tuning. Input data format.

#24 ManasiPat opened 2 years ago
1
Why DeepCopy automatically remove the token id before <endoftext> token?

#23 14H034160212 closed 2 years ago
1
python: can't open file '/u/scr/xlisali/e2e-metrics/measure_scores.py': [Errno 2] No such file or directory

#22 14H034160212 closed 2 years ago
3
FileNotFoundError for e2e_results_conv2/data2textprefixtune_y_5_act_cat_b=10-e=5_d=0.0_u=no_lr=8e-05_w=0.0_s=101_r=n_m=512_o=1_o=1_test_gold

#21 14H034160212 closed 2 years ago
1
XSUM dataset differences with original

#20 jpilaul closed 2 years ago
16
PyTorch Lightning Version?

#19 ekoenitz closed 3 years ago
2
hyper-parameters setting in low-data scenario

#18 HiXiaochen opened 3 years ago
4
Possible mistake in prefix model parameter count? I am getting 15% not 2% like in the paper

#17 jpilaul opened 3 years ago
2
IndexError: list index out of range

#16 super-buster opened 3 years ago
2
Applying PrefixTuning with T5ForConditionalGeneration model

#15 yssjtu opened 3 years ago
3
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED

#14 super-buster closed 3 years ago
1
RuntimeError: Input, output and indices must be on the current device

#13 super-buster closed 3 years ago
4
How to evaluate DART ? The test set may be changed ?

#12 JinliangLu96 closed 3 years ago
3
About prefix tuning input

#11 danilproger opened 3 years ago
0
How to full train the model?

#10 StevenTang1998 opened 3 years ago
5
Could you share `get_prompt_p5_infix`?

#9 taineleau opened 3 years ago
1
OOM error

#8 taineleau closed 3 years ago
9
About the training speed verification

#7 Timothyxxx closed 2 years ago
6
The evaluation code of train.bart

#6 probe2 closed 3 years ago
0
About the e2e version

#5 Timothyxxx closed 3 years ago
1
Can you share the hyperparameter you used for GPT2-Large?

#4 drxmy opened 3 years ago
0