Official PyTorch implementation of Time-aware Large Kernel (TaLK) Convolutions (ICML 2020)

Last update: Dec 07, 2022

Overview

Time-aware Large Kernel (TaLK) Convolutions (Lioutas et al., 2020)

This repository contains the source code, pre-trained models, as well as instructions to reproduce results for our paper Time-aware Large Kernel Convolutions (ICML 2020).

TaLK Convolutions is a sequence modeling method that uses an adaptive convolution operation that learns to predict the size of a summation kernel instead of using a fixed-sized learnable kernel matrix. It utilizes a fast parallelized implementation of the summed-area table, also known as the integral image operation, to efficiently calculate the convolution output that uses the summation kernel. We generate relative offsets for each timestep of the input sequence, which are used to adaptively expand the size of the summation kernel conditioned on the input. This method yields a time complexity of O(n), effectively making the sequence encoding process linear to the number of tokens.

Video Presentation:

Citation:

@inproceedings{lioutas2020timeaware,
    author={Vasileios Lioutas and Yuhong Guo},
    title={Time-aware Large Kernel Convolutions},
    booktitle={Proceedings of the 37th International Conference on Machine Learning (ICML)},
    year={2020}
}

Setup

Requirements

PyTorch version >= 1.3.1
fairseq version >= 0.10.1
Python version >= 3.6
CUDA >= 10.1
NVIDIA's apex library (for mixed-precision training)

Clone this repository

git clone https://github.com/lioutasb/TaLKConvolutions.git
cd TaLKConvolutions

Efficient CUDA Kernels

In order to support the parallelization of TaLK Convolutions, we have developed our own CUDA primitives. To install the kernels, use the commands below. We tested compiling the kernels using CUDA 10.1 but if a future CUDA release does not work, please feel free to open an issue.

cd talkconv/talkconv_module/
python setup.py install

We are welcoming contributions from experienced CUDA developers regarding making the CUDA kernels more efficient.

Translation

Pre-trained models

Dataset	Model	Prepared test set
IWSLT14 German-English	download (.pt)	IWSLT14 test: download (.zip)
WMT16 English-German	download (.pt)	newstest2014: download (.zip)
WMT14 English-French	download (.pt)	newstest2014: download (.zip)

Preprocessing the training datasets

Please follow the instructions https://github.com/pytorch/fairseq/blob/master/examples/translation/README.md to preprocess the data.

IWSLT14 De-En

Training and evaluating TaLK Convolutions on a single GPU:

# Training
SAVE="checkpoints/talkconv_iwslt_deen"
mkdir -p $SAVE

CUDA_VISIBLE_DEVICES=0 \
fairseq-train data-bin/iwslt14.tokenized.de-en \
    --user-dir talkconv/talkconv_fairseq \
    --arch talkconv_iwslt_de_en \
    --optimizer adam  --fp16 --lr 0.0005 \
    --source-lang de --target-lang en --max-tokens 4000 \
    --min-lr '1e-09' --weight-decay 0.0001 \
    --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
    --lr-scheduler inverse_sqrt \
    --dropout 0.3 --attention-dropout 0.1 --weight-dropout 0.1  \
    --max-update 85000 --warmup-updates 4000 --warmup-init-lr '1e-07' \
    --adam-betas '(0.9, 0.98)' --left-pad-source "False" --max-epoch 52 --seed 1024 \
    --save-dir $SAVE 

python utils/average_checkpoints.py --inputs $SAVE \
    --num-epoch-checkpoints 10 --output "${SAVE}/model.pt"

# Evaluation
fairseq-generate data-bin/iwslt14.tokenized.de-en --user-dir talkconv/talkconv_fairseq \
    --path "${SAVE}/model.pt" \
    --batch-size 128 --beam 5 --remove-bpe --lenpen 1.6 --gen-subset test --quiet

WMT16 En-De

Training and evaluating TaLK Convolutions on WMT16 En-De using cosine scheduler on one machine with 8 NVIDIA GPUs:

# Training
SAVE="checkpoints/talkconv_wmt_ende_big"
mkdir -p $SAVE

python -m torch.distributed.launch --nproc_per_node 8 fairseq-train \
    data-bin/wmt16_en_de_bpe32k --fp16 --log-interval 100 --no-progress-bar --distributed-no-spawn \
    --user-dir talkconv/talkconv_fairseq \
    --max-update 30243 --share-all-embeddings --optimizer adam \
    --adam-betas '(0.9, 0.98)' --clip-norm 0.0 --weight-decay 0.0 \
    --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
    --min-lr 1e-09 --update-freq 16 \
    --ddp-backend=no_c10d --max-tokens 3584 \
    --lr-scheduler cosine --warmup-init-lr 1e-7 --warmup-updates 10000 \
    --lr-shrink 1 --max-lr 0.001 --lr 1e-7 --min-lr 1e-9 --warmup-init-lr 1e-07 \
    --t-mult 1 --lr-period-updates 20000 \
    --arch talkconv_wmt_en_de_big \
    --save-dir $SAVE

# Checkpoint averaging
python utilss/average_checkpoints.py --inputs $SAVE \
    --num-epoch-checkpoints 10 --output "${SAVE}/model.pt"

# Evaluation on newstest2014
CUDA_VISIBLE_DEVICES=0 \
fairseq-generate data-bin/wmt16_en_de_bpe32k --user-dir talkconv/talkconv_fairseq \
  --path "${SAVE}/model.pt" \
  --batch-size 128 --beam 4 --remove-bpe --lenpen 0.35 --gen-subset test > wmt14_gen_ende.txt 

bash utils/compound_split_bleu.sh wmt14_gen_ende.txt

WMT14 En-Fr

Training and evaluating TaLK Convolutions on WMT14 En-Fr using cosine scheduler on one machine with 8 NVIDIA GPUs:

# Training
SAVE="checkpoints/talkconv_wmt_enfr_big"
mkdir -p $SAVE
python -m torch.distributed.launch --nproc_per_node 8 fairseq-train \
    data-bin/wmt14_en_fr --fp16 --log-interval 100 --no-progress-bar --distributed-no-spawn \
    --user-dir talkconv/talkconv_fairseq \
    --max-update 80000 --share-all-embeddings --optimizer adam \
    --adam-betas '(0.9, 0.98)' --clip-norm 0.0 --weight-decay 0.0 \
    --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
    --min-lr 1e-09 --update-freq 32 \
    --ddp-backend=no_c10d --max-tokens 1800 \
    --lr-scheduler cosine --warmup-init-lr 1e-7 --warmup-updates 10000 \
    --lr-shrink 1 --max-lr 0.001 --lr 1e-7 --min-lr 1e-9 --warmup-init-lr 1e-07 \
    --t-mult 1 --lr-period-updates 70000 \
    --arch talkconv_wmt_en_fr_big \
    --save-dir $SAVE

# Checkpoint averaging
python utils/average_checkpoints.py --inputs $SAVE \
    --num-epoch-checkpoints 10 --output "${SAVE}/model.pt"

# Evaluation
CUDA_VISIBLE_DEVICES=0 \
fairseq-generate data-bin/wmt14_en_fr --user-dir talkconv/talkconv_fairseq \
    --path "${SAVE}/model.pt" \
    --batch-size 128 --beam 6 --remove-bpe --lenpen 0.65 --gen-subset test --quiet

License

This project is MIT-licensed. The license applies to the pre-trained models as well.

Official PyTorch implementation of Time-aware Large Kernel (TaLK) Convolutions (ICML 2020)

Related tags

Overview

Time-aware Large Kernel (TaLK) Convolutions (Lioutas et al., 2020)

Video Presentation:

Citation:

Setup

Requirements

Clone this repository

Efficient CUDA Kernels

Translation

Pre-trained models

Preprocessing the training datasets

IWSLT14 De-En

WMT16 En-De

WMT14 En-Fr

License

Owner

Vasileios Lioutas

Implementation of COCO-LM, Correcting and Contrasting Text Sequences for Language Model Pretraining, in Pytorch

天池中药说明书实体识别挑战冠军方案；中文命名实体识别；NER; BERT-CRF & BERT-SPAN & BERT-MRC；Pytorch

Ongoing research training transformer language models at scale, including: BERT & GPT-2

VampiresVsWerewolves - Our Implementation of a MiniMax algorithm with alpha beta pruning in the context of an in-class competition

TLA - Twitter Linguistic Analysis

Translation for Trilium Notes. Trilium Notes 中文版.

VoiceFixer VoiceFixer is a framework for general speech restoration.

结巴中文分词

An A-SOUL Text Generator Based on CPM-Distill.

Official code for "Parser-Free Virtual Try-on via Distilling Appearance Flows", CVPR 2021

NLP techniques such as named entity recognition, sentiment analysis, topic modeling, text classification with Python to predict sentiment and rating of drug from user reviews.

GPT-Code-Clippy (GPT-CC) is an open source version of GitHub Copilot, a language model

An implementation of WaveNet with fast generation

Code for the Findings of NAACL 2022(Long Paper): AdapterBias: Parameter-efficient Token-dependent Representation Shift for Adapters in NLP Tasks

SentimentArcs: a large ensemble of dozens of sentiment analysis models to analyze emotion in text over time

Yet Another Neural Machine Translation Toolkit

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning

A method to generate speech across multiple speakers

Datasets of Automatic Keyphrase Extraction

A Transformer Implementation that is easy to understand and customizable.