Sequence-to-Sequence learning using PyTorch

Last update: Nov 17, 2022

Overview

Seq2Seq in PyTorch

This is a complete suite for training sequence-to-sequence models in PyTorch. It consists of several models and code to both train and infer using them.

Using this code you can train:

Neural-machine-translation (NMT) models
Language models
Image to caption generation
Skip-thought sentence representations
And more...

Installation

git clone --recursive https://github.com/eladhoffer/seq2seq.pytorch
cd seq2seq.pytorch; python setup.py develop

Models

Models currently available:

Simple Seq2Seq recurrent model
Recurrent Seq2Seq with attentional decoder
Google neural machine translation (GNMT) recurrent model
Transformer - attention-only model from "Attention Is All You Need"

Datasets

Datasets currently available:

WMT16
WMT17
OpenSubtitles 2016
COCO image captions
Conceptual captions

All datasets can be tokenized using 3 available segmentation methods:

Character based segmentation
Word based segmentation
Byte-pair-encoding (BPE) as suggested by bpe with selectable number of tokens.

After choosing a tokenization method, a vocabulary will be generated and saved for future inference.

Training methods

The models can be trained using several methods:

Basic Seq2Seq - given encoded sequence, generate (decode) output sequence. Training is done with teacher-forcing.
Multi Seq2Seq - where several tasks (such as multiple languages) are trained simultaneously by using the data sequences as both input to the encoder and output for decoder.
Image2Seq - used to train image to caption generators.

Usage

Example training scripts are available in scripts folder. Inference examples are available in examples folder.

example for training a transformer on WMT16 according to original paper regime:

DATASET=${1:-"WMT16_de_en"}
DATASET_DIR=${2:-"./data/wmt16_de_en"}
OUTPUT_DIR=${3:-"./results"}

WARMUP="4000"
LR0="512**(-0.5)"

python main.py \
  --save transformer \
  --dataset ${DATASET} \
  --dataset-dir ${DATASET_DIR} \
  --results-dir ${OUTPUT_DIR} \
  --model Transformer \
  --model-config "{'num_layers': 6, 'hidden_size': 512, 'num_heads': 8, 'inner_linear': 2048}" \
  --data-config "{'moses_pretok': True, 'tokenization':'bpe', 'num_symbols':32000, 'shared_vocab':True}" \
  --b 128 \
  --max-length 100 \
  --device-ids 0 \
  --label-smoothing 0.1 \
  --trainer Seq2SeqTrainer \
  --optimization-config "[{'step_lambda':
                          \"lambda t: { \
                              'optimizer': 'Adam', \
                              'lr': ${LR0} * min(t ** -0.5, t * ${WARMUP} ** -1.5), \
                              'betas': (0.9, 0.98), 'eps':1e-9}\"
                          }]"

example for training attentional LSTM based model with 3 layers in both encoder and decoder:

python main.py \
  --save de_en_wmt17 \
  --dataset ${DATASET} \
  --dataset-dir ${DATASET_DIR} \
  --results-dir ${OUTPUT_DIR} \
  --model RecurrentAttentionSeq2Seq \
  --model-config "{'hidden_size': 512, 'dropout': 0.2, \
                   'tie_embedding': True, 'transfer_hidden': False, \
                   'encoder': {'num_layers': 3, 'bidirectional': True, 'num_bidirectional': 1, 'context_transform': 512}, \
                   'decoder': {'num_layers': 3, 'concat_attention': True,\
                               'attention': {'mode': 'dot_prod', 'dropout': 0, 'output_transform': True, 'output_nonlinearity': 'relu'}}}" \
  --data-config "{'moses_pretok': True, 'tokenization':'bpe', 'num_symbols':32000, 'shared_vocab':True}" \
  --b 128 \
  --max-length 80 \
  --device-ids 0 \
  --trainer Seq2SeqTrainer \
  --optimization-config "[{'epoch': 0, 'optimizer': 'Adam', 'lr': 1e-3},
                          {'epoch': 6, 'lr': 5e-4},
                          {'epoch': 8, 'lr':1e-4},
                          {'epoch': 10, 'lr': 5e-5},
                          {'epoch': 12, 'lr': 1e-5}]" \

Sequence-to-Sequence learning using PyTorch

Related tags

Overview

Seq2Seq in PyTorch

Installation

Models

Datasets

Training methods

Usage

Owner

Elad Hoffer

Collective Multi-type Entity Alignment Between Knowledge Graphs (WWW'20)

Implementation of Uformer, Attention-based Unet, in Pytorch

Deep motion generator collections

Pytorch implementation for the EMNLP 2020 (Findings) paper: Connecting the Dots: A Knowledgeable Path Generator for Commonsense Question Answering

Nightmare-Writeup - Writeup for the Nightmare CTF Challenge from 2022 DiceCTF

Pretrained language model and its related optimization techniques developed by Huawei Noah's Ark Lab.

Learning Features with Parameter-Free Layers (ICLR 2022)

PyTorch DepthNet Training on Still Box dataset

This is the winning solution of the Endocv-2021 grand challange.

Train Dense Passage Retriever (DPR) with a single GPU

TensorRT examples (Jetson, Python/C++)(object detection)

Space Ship Simulator using python

Full Transformer Framework for Robust Point Cloud Registration with Deep Information Interaction

Pocsploit is a lightweight, flexible and novel open source poc verification framework

Parallel and High-Fidelity Text-to-Lip Generation; AAAI 2022 ; Official code

Image restoration with neural networks but without learning.

Vision transformers (ViTs) have found only limited practical use in processing images

3D cascade RCNN for object detection on point cloud

Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy" (ICLR 2022 Spotlight)

Predicting a person's gender based on their weight and height