An implementation of the Pay Attention when Required transformer

Overview

Pay Attention when Required (PAR) Transformer-XL

An implementation of the Pay Attention when Required transformer from the paper: https://arxiv.org/pdf/2009.04534.pdf

alt text [source: Jonathan Kernes]

Quick overview

The Pay Attention when Required Transformer (Mandava, et. al. 2020) is just a regular transformer-XL (Dai et. al. 2019)[https://arxiv.org/pdf/1901.02860.pdf] , but the ratio of attention and dense layers has been optimized. This optimization is performed by allowing the network to choose which types of layer it prefers in each block of the network. The present implementation is not an exact replica of the author's efforts. Instead, we perform a simultaneous optimization procedure on both the model architecture and model parameters. The search is performed using a SuperNet, which is a sequential neural network composed of stochastic blocks, as shown in the figure below (taken from the paper. Please don't sue me!)

alt text [Source: Mandava et. al. 2020]

The key component is a Gumbel-Softmax layer [(Jang et al., 2016) and (Maddison et al., 2016). jang link: https://arxiv.org/pdf/1611.01144.pdf]. This layer is a continuous representation of a discrete sampling from a Categorical distribution, thereby allowing us to use gradients to learn parameters of a discrete distribution. (Recall a categorical is a distrbution over K states with kth state having probability pi_k, and we must have the normalization condition \sum_{i=1}^K pi_i = 1)

As the model learns, it is free to adjust both the usual model parameters, as well as its architecture search parameters pi, indicating the probability of choosing either

  1. Attention

  2. Dense

  3. Identity

for any given stochastic block. We perform simulated annealing: since the categorical distribution is approximated by a continuous representation, we get some scores like (0.02, 0.98, 0.02) for the probability of say sampling that state 2 is picked. The sharpness of this is set by a parameter \tau (the temperature), with a categorical distribution the limit tau-->0. Simulated annealing means we begin with tau=1 to let the model figure out what it wants, then slowly decrease tau so the distribution approaches a categorical.

All of this is implemented on the freely available wiki-text2 dataset.

Explanation of the main GIF: The main gif is the result of our experiments. It shows the pi distribution for each stochastic block of a 6 block SuperNet, as a function of training iterations. The number indicates the probability of the most likely layer type (darker means more probable). As you can see, the model learns to put attention in the beginning, and dense layers at the end.

Requirements

Usual ML stuff, if you have a conda environment, python 3+, TensorFlow 2+ you should be ok. You will need TensorFlow Text as well to handle the SentencePiece Tokenization

If you choose to run your own tokenizer (a flag option in data_utils for handling new text data), you will also need to download the SentencePiece package: https://github.com/google/sentencepiece

Data

The dataset used is Wiki-text2. We have provided a copy of this in the data folder, along with some preprocessed data for training. In order to reproduce this from scratch, run the shell script

./create_tfrecords.sh

This will download the wiki-text2 dataset from its source, then proceed to clean, batch, and write the data to a tfrecords file. The shell script calls build_data.py which offers more control over what type of data to generate. The general parameters you will want to tune are:

*batch_size *seq_len.

You can also supply your own dataset instead of the one provided. The underlying tokenizer uses sentencepiece (Kudo): https://github.com/google/sentencepiece, which works at the byte level and can handle any kind of input. Simply change the --input_text flag to your file, and set the desired --vocab_size.

Why do we need to specify the batch size? Transformer XL uses memory states to form a recurrent, long range network. After analyzing a particular sequence say [A,B] of the sequence [A,B,C,D], the results of [A,B] are fed into the [C,D] calculation with a stop gradient. Therefore, we must be sure that each datapoint follows chronologically from the previous one.

This is achieved by context batching (see data_utils.py function) where we break the entire dataset into batch_size segments, then pull in order one sequence from each batch at a time to form the dataset. Because of this, note that adding more shards to the data could result in a large loss (order of batch_size*seq_len*shards), as each shard will drop the remaining datapoint of size (batch_size*seq_len) to keep the tensor shapes.

Addtional technical details

Per the original Transformer-XL, we also implement an adaptive softmax layer (Grave et. al. 2017, https://arxiv.org/abs/1609.04309) to deal with a potentially large number of outputs in the final dense layer. This implemenation is inspired by the TF 1.0 example at https://github.com/yangsaiyong/tf-adaptive-softmax-lstm-lm. To use the adaptive softmax, set the --cutoffs= flag in train.py. The cutoffs are the max values of each bin, and should NOT include the vocab size (i.e. the max cutoff of the final bin). If no cutoffs are specified, the model defaults to normal softmax.

For completeness, we have also provided a script optimal_cuts.py that determines the optimal cutoffs given a return space separated file of unigram probabilities (based on the assumptions of Grave et. al. regarding GPU computation complexity -- see the paper for details). The algorithm uses dynamic programming, but is quite slow at O(KN^2), for K cutoffs and N vocab words. In principle it's a one time cost to determine the cutoffs, but we are impatient and recommend to just play around with the cutoffs instead. See the script for flag details

Training and Benchmarks

The default model we use has memory length 16, feed-forward dimension 1024, attention dimension 128, and 6 stochastic blocks, with an adaptive softmax layer and 2 clusters. We trained on a colab GPU for 20 epochs, taking a total of 37 minutes. We use an Adam optimzer with cosine rate decay: an initial warmup of 4000 steps and a maximum learning rate of 1e-4, decaying to zero at the end of training. Our training benchmarks are:

Iteration (thousands) Train_perplexity Validation_perplexity Time
2.7k 163.9 114.4 1m 58s
8.5k 78.56 62.33 5m 37s
14.1k 65.71 51.88 9m 28s
28.3k 48.52 42.61 18m 40s
48.1k 41.85 39.57 31m 51s
56.5k 42.12 39.41 37m 14s

To train, simply run the shell script

./base_model.sh

adjusting the parameters as you see fit. The above model is the default configuration. To train in colab, simply open up the notebook "colab.ipynb" and follow the instructions. This is most easily done by going to [google.colab.com] and searching this repository in github. The benefit of colab, is it's easier to play around with the model after training.

While training, we have provided two ways to monitor the output

  1. A tensorboard log. The colab notebook takes care of running this for you. In the terminal, first create a 'logs' directory, then run the command tensorboard --logdir logs in a separate tab. This will open a port where you can view live plots of the learning rate, tau annealing, train/valid loss and perplexity.

  2. An output log saved to training_log.log. This will log the model summary, parameters, etc. as well as print out loss updates every 100 steps and save it to the log file.

Thanks for reading this far!

Enjoy! And thank you to the wonderful researchers that inspired this project.

If you would like to contribute, or have any comments questions concerns please open a pull request or email me directly.

BERN2: an advanced neural biomedical namedentity recognition and normalization tool

BERN2 We present BERN2 (Advanced Biomedical Entity Recognition and Normalization), a tool that improves the previous neural network-based NER tool by

DMIS Laboratory - Korea University 99 Jan 06, 2023
The following links explain a bit the idea of semantic search and how search mechanisms work by doing retrieve and rerank

Main Idea The following links explain a bit the idea of semantic search and how search mechanisms work by doing retrieve and rerank Semantic Search Re

Sergio Arnaud Gomez 2 Jan 28, 2022
Coreference resolution for English, German and Polish, optimised for limited training data and easily extensible for further languages

Coreferee Author: Richard Paul Hudson, msg systems ag 1. Introduction 1.1 The basic idea 1.2 Getting started 1.2.1 English 1.2.2 German 1.2.3 Polish 1

msg systems ag 169 Dec 21, 2022
The aim of this task is to predict someone's English proficiency based on a text input.

English_proficiency_prediction_NLP The aim of this task is to predict someone's English proficiency based on a text input. Using the The NICT JLE Corp

1 Dec 13, 2021
Korean Simple Contrastive Learning of Sentence Embeddings using SKT KoBERT and kakaobrain KorNLU dataset

KoSimCSE Korean Simple Contrastive Learning of Sentence Embeddings implementation using pytorch SimCSE Installation git clone https://github.com/BM-K/

34 Nov 24, 2022
Rhasspy 673 Dec 28, 2022
A multi-voice TTS system trained with an emphasis on quality

TorToiSe Tortoise is a text-to-speech program built with the following priorities: Strong multi-voice capabilities. Highly realistic prosody and inton

James Betker 2.1k Jan 01, 2023
jiant is an NLP toolkit

🚨 Update 🚨 : As of 2021/10/17, the jiant project is no longer being actively maintained. This means there will be no plans to add new models, tasks,

ML² AT CILVR 1.5k Dec 28, 2022
ProteinBERT is a universal protein language model pretrained on ~106M proteins from the UniRef90 dataset.

ProteinBERT is a universal protein language model pretrained on ~106M proteins from the UniRef90 dataset. Through its Python API, the pretrained model can be fine-tuned on any protein-related task in

241 Jan 04, 2023
Source code for AAAI20 "Generating Persona Consistent Dialogues by Exploiting Natural Language Inference".

Generating Persona Consistent Dialogues by Exploiting Natural Language Inference Source code for RCDG model in AAAI20 Generating Persona Consistent Di

16 Oct 08, 2022
Learning Spatio-Temporal Transformer for Visual Tracking

STARK The official implementation of the paper Learning Spatio-Temporal Transformer for Visual Tracking Highlights The strongest performances Tracker

Multimedia Research 485 Jan 04, 2023
BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese

Table of contents Introduction Using BARTpho with fairseq Using BARTpho with transformers Notes BARTpho: Pre-trained Sequence-to-Sequence Models for V

VinAI Research 58 Dec 23, 2022
Ceaser-Cipher - The Caesar Cipher technique is one of the earliest and simplest method of encryption technique

Ceaser-Cipher The Caesar Cipher technique is one of the earliest and simplest me

Lateefah Ajadi 2 May 12, 2022
Deal or No Deal? End-to-End Learning for Negotiation Dialogues

Introduction This is a PyTorch implementation of the following research papers: (1) Hierarchical Text Generation and Planning for Strategic Dialogue (

Facebook Research 1.4k Dec 29, 2022
This project is part of Eleuther AI's quest to create a massive repository of high quality text data for training language models.

This project is part of Eleuther AI's quest to create a massive repository of high quality text data for training language models.

EleutherAI 42 Dec 13, 2022
jel - Japanese Entity Linker - is Bi-encoder based entity linker for japanese.

jel: Japanese Entity Linker jel - Japanese Entity Linker - is Bi-encoder based entity linker for japanese. Usage Currently, link and question methods

izuna385 10 Jan 06, 2023
ProtFeat is protein feature extraction tool that utilizes POSSUM and iFeature.

Description: ProtFeat is designed to extract the protein features by employing POSSUM and iFeature python-based tools. ProtFeat includes a total of 39

GOKHAN OZSARI 5 Dec 16, 2022
Reproducing the Linear Multihead Attention introduced in Linformer paper (Linformer: Self-Attention with Linear Complexity)

Linear Multihead Attention (Linformer) PyTorch Implementation of reproducing the Linear Multihead Attention introduced in Linformer paper (Linformer:

Kui Xu 58 Dec 23, 2022
Paddle2.x version AI-Writer

Paddle2.x 版本AI-Writer 用魔改 GPT 生成网文。Tuned GPT for novel generation.

yujun 74 Jan 04, 2023
L3Cube-MahaCorpus a Marathi monolingual data set scraped from different internet sources.

L3Cube-MahaCorpus L3Cube-MahaCorpus a Marathi monolingual data set scraped from different internet sources. We expand the existing Marathi monolingual

21 Dec 17, 2022