microsoft
/
GRTr
зеркало из https://github.com/microsoft/GRTr.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
Generative Retrieval Transformer
21 коммит 3 Ветки 0 Теги 233 KiB
Python 92.9%
Shell 5.3%
Dockerfile 1.8%
Перейти к файлу
Igor Shalyminov 00035052da
Merge pull request #5 from ishalyminov/retrieval_window 
Retrieval window
 		2020-09-23 19:31:18 +01:00
grtr encodings cache fix 2020-09-23 18:29:49 +00:00
img README.md updated 2020-03-04 19:09:52 +00:00
scripts fix 2020-09-18 19:44:45 +00:00
.gitignore Initial commit 2020-01-20 22:06:17 +00:00
CODE_OF_CONDUCT.md Initial CODE_OF_CONDUCT.md commit 2020-01-20 14:06:21 -08:00
Dockerfile Initial project commit 2020-01-20 17:15:09 -05:00
LICENSE Initial LICENSE commit 2020-01-20 14:06:22 -08:00
README.md ICASSP 2020 paper linked 2020-05-18 01:07:33 +01:00
SECURITY.md Initial SECURITY.md commit 2020-01-20 14:06:23 -08:00
environment.yml Initial project commit 2020-01-20 17:15:09 -05:00
metalwoz_dataspec.json More updates for release 2020-02-18 13:16:15 -05:00
requirements.txt Initial project commit 2020-01-20 17:15:09 -05:00
setup.cfg Initial project commit 2020-01-20 17:15:09 -05:00
setup.py Initial project commit 2020-01-20 17:15:09 -05:00

README.md

GRTr: Generative-Retrieval Transformers

Codebase for the papers:

Hybrid Generative-Retrieval Transformers for Dialogue Domain Adaptation, DSTC8@AAAI 2020

Fast Domain Adaptation for Goal-Oriented Dialogue Using a Hybrid Generative-Retrieval Transformer, ICASSP 2020 [virtual presentation]

By Igor Shalyminov, Alessandro Sordoni, Adam Atkinson, Hannes Schulz.

Model diagram

Installation

$ cd code-directory
$ conda create -n hybrid_retgen python=3.7
$ conda activate hybrid_retgen
$ conda install cython
$ pip install -e .

For mixed precision training:

$ pip install git+https://github.com/nvidia/apex

Datasets and experimental setup

Training a base GPT-2 model on MetaLWOz

$ python scripts/train <MetaLWOz zipfile> metalwoz_dataspec.json --dataset_cache cache exp/grtr --train_batch_size 4 --valid_batch_size 4 --early_stopping_after -1 --n_epochs 25

Add --fp16 O1 to use mixed precision training.

Predictions

generate-and-rank

python scripts/predict_generate_and_rank <MetaLWOz/MultiWoz zipfile> <testspec json> <output dir> <base GPT-2 model dir> --fine-tune --dataset_cache cache exp/grtr --train_batch_size 4 --valid_batch_size 4

generate only

python scripts/predict <MetaLWOz/MultiWoz zipfile> <testspec json> <output dir> <base GPT-2 model dir> --fine-tune --dataset_cache cache exp/grtr --train_batch_size 4 --valid_batch_size 4

Convenience bash scripts are provided in scripts/ to produce predictions for each of the three test specs.

Evaluation can be done using the evaluate script in the competition baseline repository.

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.