Prefix Tuning
Note:
作者在论文中提到使用真实的word去初始化prefix的操作（Initializing the prefix with activations of real words，significantly improves generation）。我在使用作者提供的代码时遇到了一些问题，因此按照代码的思路添加了利用真实词汇进行初始化的内容。
可以采用以下的方式运行：
Train
cd seq2seq;
python train_bart.py --mode xsum --preseqlen 200 --do_train yes --fp16 yes --bsz 16 --epoch 30 --gradient_accumulation_step 3 --learning_rate 0.00005 --mid_dim 800 --use_lowdata_token 'yes' --lowdata_token 'summarize'
其中use_lowdata_token表示是否采用real word初始化的方式；lowdata_token表示传入的real word.
Decode
cd seq2seq;
python train_bart.py --mode xsum --do_train no --prefix_model_path {checkpoint_path} --preseqlen {same as training} --mid_dim {same as training} --use_lowdata_token 'yes' --lowdata_token 'summarize'
Files:
.
├── gpt2 # Code for GPT2 style autoregressive LM
│ ├── train_e2e.py # high-level scripts to train.
│ ├── train_control.py # code that implements prefix-tuning.
│ ├── trainer_prefix.py # trainer code for the training loop.
│ ├── run_language_modeling.py # training code (contains data loading, model loading, and calls trainer)
│ ├── gen.py # high-level scripts to decode.
│ └── run_generation.py # decoding code.
│
├── seq2seq # Code for encoder-decoder architecture
│ ├── train_bart.py # high-level scripts to train.
│ ├── prefixTuning.py # code that implements prefix-tuning.
│ ├── finetune.py # training code (contains data loading, model loading, and calls trainer)
│ ├── lightning_base.py # helper code
│ ├── utils.py # helper code
│ └── callbacks.py # helper code
└── ...
To run the code for GPT2 style autoregressive LM, the code is in gpt2/. This corresponds to the table-to-text experiments in the paper.
To run the code for encoder-decoder architecture like BART, the code is in seq2seq. This corresponds to the summarization experiments in the paper.
The two primary scripts I used to run my codes are gpt2/train_e2e.py (for table-to-text) and seq2seq/train_bart.py(for summarization). they are set to default of good hyperparameters, and can be used to tune hyperparameter :)
Setup:
cd transformer; pip install -e .
Train via prefix-tuning:
cd gpt2;
python train_e2e.py --optim_prefix yes --preseqlen 5 --epoch 5 --learning_rate 0.00005 --mode webnlg --bsz 5 --seed 101
cd seq2seq;
python train_bart.py --mode xsum --preseqlen 200 --do_train yes --fp16 yes --bsz 16 --epoch 30 --gradient_accumulation_step 3 --learning_rate 0.00005 --mid_dim 800
Other baseline approaches
cd gpt2;
python train_e2e.py --tuning_mode {finetune/adaptertune} --epoch 5 --learning_rate 0.00005 --mode webnlg --bsz 5 --seed 101
cd seq2seq;
python train_e2e.py --tuning_mode finetune --epoch 5 --learning_rate 0.00005 --mode webnlg --bsz 5 --seed 101
Decode:
cd gpt2;
python gen.py {data2text/webnlg/...} yes test {checkpoint_path} no
cd seq2seq;
python train_bart.py --mode xsum --do_train no --prefix_model_path {checkpoint_path} --preseqlen {same as training} --mid_dim {same as training}
For details of the methods and results, please refer to our paper.
@misc{li2021prefixtuning,
title={Prefix-Tuning: Optimizing Continuous Prompts for Generation},
author={Xiang Lisa Li and Percy Liang},
year={2021},
eprint={2101.00190},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
172 Dec 27, 2022
2.9k Jan 08, 2023
205 Dec 20, 2022
10 Dec 20, 2021
34 Jul 22, 2022
2k Jan 02, 2023
6.3k Jan 09, 2023
145 Dec 24, 2022
1.9k Dec 28, 2022
44 Dec 31, 2022
1 Jan 12, 2022
406 Dec 24, 2022
172 Dec 22, 2022
1.4k Jan 04, 2023
0 Mar 25, 2022
2.7k Dec 26, 2022
394 Dec 23, 2022
29 Nov 09, 2022
1 Jan 14, 2022
224 Dec 28, 2022