Code for our EMNLP 2021 paper "Learning Kernel-Smoothed Machine Translation with Retrieved Examples"

Last update: Nov 24, 2022

Related tags

Overview

KSTER

Code for our EMNLP 2021 paper "Learning Kernel-Smoothed Machine Translation with Retrieved Examples" [paper].

Usage

Download the processed datasets from this site. You can also download the built databases from this site and download the model checkpoints from this site.

Train a general-domain base model

Take English -> Germain translation for example.

export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m joeynmt train configs/transformer_base_wmt14_en2de.yaml

Finetuning trained base model on domain-specific datasets

Take English -> Germain translation in Koran domain for example.

export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m joeynmt train configs/transformer_base_koran_en2de.yaml

Build database

Take English -> Germain translation in Koran domain for example, wmt14_en_de.transformer.ckpt is the path of trained general-domain base model checkpoint.

mkdir database/koran_en_de_base
export CUDA_VISIBLE_DEVICES=0
python3 -m joeynmt build_database configs/transformer_base_koran_en2de.yaml \
        --ckpt wmt14_en_de.transformer.ckpt \
        --division train \
        --index_path database/koran_en_de_base/trained.index \
        --token_map_path database/koran_en_de_base/token_map \
        --embedding_path database/koran_en_de_base/embeddings.npy

Train the bandwidth estimator and weight estimator in KSTER

Take English -> Germain translation in Koran domain for example.

export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m joeynmt combiner_train configs/transformer_base_koran_en2de.yaml \
        --ckpt wmt14_en_de.transformer.ckpt \
        --combiner dynamic_combiner \
        --top_k 16 \
        --kernel laplacian \
        --index_path database/koran_en_de_base/trained.index \
        --token_map_path database/koran_en_de_base/token_map \
        --embedding_path database/koran_en_de_base/embeddings.npy \
        --in_memory True

Inference

We unify the inference of base model, finetuned or joint-trained model, kNN-MT and KSTER with a concept of combiner (see joeynmt/combiners.py).

Combiner type	Methods	Description
NoCombiner	Base, Finetuning, Joint-training	Directly inference without retrieval.
StaticCombiner	kNN-MT	Retrieve similar examples during inference. mixing_weight and bandwidth are pre-specified.
DynamicCombiner	KSTER	Retrieve similar examples during inference. mixing_weight and bandwidth are dynamically estimated.

Inference with NoCombiner for Base model

Take English -> Germain translation in Koran domain for example.

export CUDA_VISIBLE_DEVICES=0
python3 -m joeynmt test configs/transformer_base_koran_en2de.yaml \
        --ckpt wmt14_en_de.transformer.ckpt \
        --combiner no_combiner

Inference with StaticCombiner for kNN-MT

Take English -> Germain translation in Koran domain for example.

export CUDA_VISIBLE_DEVICES=0
python3 -m joeynmt test configs/transformer_base_koran_en2de.yaml \
        --ckpt wmt14_en_de.transformer.ckpt \
        --combiner static_combiner \
        --top_k 16 \
        --mixing_weight 0.7 \
        --bandwidth 10 \
        --kernel gaussian \
        --index_path database/koran_en_de_base/trained.index \
        --token_map_path database/koran_en_de_base/token_map

Inference with DynamicCombiner for KSTER

Take English -> Germain translation in Koran domain for example, koran_en_de.laplacian.combiner.ckpt is the path of trained bandwidth estimator and weight estimator for Koran domain.
--in_memory option specifies whether to load the example embeddings to memory. Set in_memory == True for faster inference, set in_memory == False for lower memory demand.

export CUDA_VISIBLE_DEVICES=0
python3 -m joeynmt test configs/transformer_base_koran_en2de.yaml \
        --ckpt wmt14_en_de.transformer.ckpt \
        --combiner dynamic_combiner \
        --combiner_path koran_en_de.laplacian.combiner.ckpt \
        --top_k 16 \
        --kernel laplacian \
        --index_path database/koran_en_de_base/trained.index \
        --token_map_path database/koran_en_de_base/token_map \
        --embedding_path database/koran_en_de_base/embeddings.npy \
        --in_memory True

See bash_scripts/test_*.sh for reproducing our results.
See logs/*.log for the logs of our results.

Acknowledgements

We build the models based on the joeynmt codebase.

Code for our EMNLP 2021 paper "Learning Kernel-Smoothed Machine Translation with Retrieved Examples"

Related tags

Overview

KSTER

Usage

Train a general-domain base model

Finetuning trained base model on domain-specific datasets

Build database

Train the bandwidth estimator and weight estimator in KSTER

Inference

Inference with NoCombiner for Base model

Inference with StaticCombiner for kNN-MT

Inference with DynamicCombiner for KSTER

Acknowledgements

Owner

jiangqn

Tutorials, assignments, and competitions for MIT Deep Learning related courses.

Pytorch implementation for A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose

In-Place Activated BatchNorm for Memory-Optimized Training of DNNs

Hippocampal segmentation using the UNet network for each axis

Incremental Transformer Structure Enhanced Image Inpainting with Masking Positional Encoding (CVPR2022)

Unofficial pytorch-lightning implement of Mip-NeRF

Official PyTorch implementation of CAPTRA: CAtegory-level Pose Tracking for Rigid and Articulated Objects from Point Clouds

Official Implementation of Neural Splines

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (ICCV, 2021) (PyTorch) - We released the training code!

Code for paper "Which Training Methods for GANs do actually Converge? (ICML 2018)"

Lipstick ain't enough: Beyond Color-Matching for In-the-Wild Makeup Transfer (CVPR 2021)

Parametric Contrastive Learning (ICCV2021)

Gapmm2: gapped alignment using minimap2 (align transcripts to genome)

This repo contains code to reproduce all experiments in Equivariant Neural Rendering

(ICCV 2021) PyTorch implementation of Paper "Progressive Correspondence Pruning by Consensus Learning"

YOLOv5 detection interface - PyQt5 implementation

Torch implementation of "Enhanced Deep Residual Networks for Single Image Super-Resolution"

source code the paper Fast and Robust Iterative Closet Point.

This code finds bounding box of a single human mouth.

Dictionary Learning with Uniform Sparse Representations for Anomaly Detection