An implementation of Fastformer: Additive Attention Can Be All You Need in TensorFlow

Last update: Dec 28, 2022

Overview

Fast Transformer

This repo implements Fastformer: Additive Attention Can Be All You Need by Wu et al. in TensorFlow. Fast Transformer is a Transformer variant based on additive attention that can handle long sequences efficiently with linear complexity. Fastformer is much more efficient than many existing Transformer models and can meanwhile achieve comparable or even better long text modeling performance.

Installation

Run the following to install:

pip install fast-transformer

Developing fast-transformer

To install fast-transformer, along with tools you need to develop and test, run the following in your virtualenv:

git clone https://github.com/Rishit-dagli/Fast-Transformer.git
# or clone your own fork

cd fast-transformer
pip install -e .[dev]

Usage

import tensorflow as tf
from fast_transformer import FastTransformer

mask = tf.ones([1, 4096], dtype=tf.bool)
model = FastTransformer(
    num_tokens = 20000,
    dim = 512,
    depth = 2,
    max_seq_len = 4096,
    absolute_pos_emb = True, # Absolute positional embeddings
    mask = mask
)
x = tf.experimental.numpy.random.randint(0, 20000, (1, 4096))

logits = model(x) # (1, 4096, 20000)

Want to Contribute 🙋‍♂️ ?

Awesome! If you want to contribute to this project, you're always welcome! See Contributing Guidelines. You can also take a look at open issues for getting more information about current or upcoming tasks.

Want to discuss? 💬

Have any questions, doubts or want to present your opinions, views? You're always welcome. You can start discussions.

Citation

@misc{wu2021fastformer,
    title   = {Fastformer: Additive Attention is All You Need}, 
    author  = {Chuhan Wu and Fangzhao Wu and Tao Qi and Yongfeng Huang},
    year    = {2021},
    eprint  = {2108.09084},
    archivePrefix = {arXiv},
    primaryClass = {cs.CL}
}

Yannic Kilcher's video was super helpful while building this.

License

Copyright 2020 Rishit Dagli

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

You might also like...

Unofficial PyTorch implementation of Neural Additive Models (NAM) by Agarwal, et al.

nam-pytorch Unofficial PyTorch implementation of Neural Additive Models (NAM) by Agarwal, et al. [abs, pdf] Installation You can access nam-pytorch vi

11 Mar 14, 2022

The tool under this branch fork can be used to crack devices above A12 and up to A15. After cracking, you can also use SSH channel strong opening tool to open SSH channel and activate it with Demo or Shell script. The file can be extracted from my Github homepage, and the SSH channel opening tool can be extracted from Dr238 account.

Welcome to C0xy-A12-A15-Attack-Tool The tool under this branch fork can be used to crack devices above A12 and up to A15. After cracking, you can also

13 Dec 23, 2022

PixelPick This is an official implementation of the paper "All you need are a few pixels: semantic segmentation with PixelPick."

PixelPick This is an official implementation of the paper "All you need are a few pixels: semantic segmentation with PixelPick." [Project page] [Paper

59 Sep 25, 2022

Unofficial implementation of HiFi-GAN+ from the paper "Bandwidth Extension is All You Need" by Su, et al.

HiFi-GAN+ This project is an unoffical implementation of the HiFi-GAN+ model for audio bandwidth extension, from the paper Bandwidth Extension is All

134 Dec 30, 2022

H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.

H2O H2O is an in-memory platform for distributed, scalable machine learning. H2O uses familiar interfaces like R, Python, Scala, Java, JSON and the Fl

6.1k Jan 5, 2023

Torch-based tool for quantizing high-dimensional vectors using additive codebooks

Trainable multi-codebook quantization This repository implements a utility for use with PyTorch, and ideally GPUs, for training an efficient quantizer

41 Jan 7, 2023

Neural network-based build time estimation for additive manufacturing

Neural network-based build time estimation for additive manufacturing Oh, Y., Sharp, M., Sprock, T., & Kwon, S. (2021). Neural network-based build tim

1 Nov 15, 2021

Load What You Need: Smaller Multilingual Transformers for Pytorch and TensorFlow 2.0.

Smaller Multilingual Transformers This repository shares smaller versions of multilingual transformers that keep the same representations offered by t

79 Dec 28, 2022

Code for "Diffusion is All You Need for Learning on Surfaces"

Source code for "Diffusion is All You Need for Learning on Surfaces", by Nicholas Sharp Souhaib Attaiki Keenan Crane Maks Ovsjanikov NOTE: the linked

247 Dec 28, 2022

Comments

Implement Additive Attention
Implement Additive Attention as a TensorFlow layer:

[x] Figure out using rotary embeddings

[x] Add masking functionality

[x] Relative Position embeddings

[x] Calculate query attention logits

[x] Calculate Global Query tokens

[x] Calculate key attention logits

[x] Calculate Global Key tokens

[x] Add queries as residuals
opened by Rishit-dagli 0

Releases(v0.2.0)

v0.2.0(Jan 16, 2022)
✅ Bug Fixes / Improvements

Unit Tests for output rank and shape

Looser dependency requirements (now supports all TensorFlow versions >= 2.5.0)

Source code(tar.gz)
Source code(zip)
v0.1.0(Sep 3, 2021)
This is the initial release of Fast Transformer and implements Fast Transformer as a subclassed TensorFlow model.

Classes

FastAttention: Implements additive attention as a TensorFlow Keras layer, and supports using relative positional encodings.

PreNorm: Normalize the activations of the previous layer for each given example in a batch independently and apply some function to it, implemented as a TensorFlow Keras Layer.

FeedForward: Create a FeedForward neural net with two Dense layers and GELU activation, implemented as a TensorFlow Keras Layer.

FastTransformer: Implements the FastTransformer model using all the other classes, allows using rotary embeddings, weight tie projections, and converts to logits. Implemented as a TensorFlow Keras Model.

Source code(tar.gz)
Source code(zip)

Owner

Rishit Dagli

High School,TEDx,2xTED-Ed speaker | International Speaker | Microsoft Student Ambassador | Mentor, @TFUGMumbai | Organize @KotlinMumbai

GitHub Repository

Tightness-aware Evaluation Protocol for Scene Text Detection

TIoU-metric Release on 27/03/2019. This repository is built on the ICDAR 2015 evaluation code. If you propose a better metric and require further eval

206 Nov 18, 2022

Fast sparse deep learning on CPUs

SPARSEDNN **If you want to use this repo, please send me an email: [email pro

44 Nov 30, 2022

This is the repo for the paper `SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization'. (published in Bioinformatics'21)

SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization This is the code for our paper ``SumGNN: Multi-typed Drug

58 Dec 21, 2022

Image Super-Resolution Using Very Deep Residual Channel Attention Networks

14 Oct 14, 2022

This repository provides the official implementation of 'Learning to ignore: rethinking attention in CNNs' accepted in BMVC 2021.

inverse_attention This repository provides the official implementation of 'Learning to ignore: rethinking attention in CNNs' accepted in BMVC 2021. Le

5 Jul 08, 2022

Implementation of ICCV2021(Oral) paper - VMNet: Voxel-Mesh Network for Geodesic-aware 3D Semantic Segmentation

VMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation Created by Zeyu HU Introduction This work is based on our paper VMNet: Voxel-Mes

82 Dec 27, 2022

Converting CPT to bert form for use

cpt-encoder 将CPT转成bert形式使用说明刚刚刷到又出了一种模型：CPT，看论文显示，在很多中文任务上性能比mac bert还好，就迫不及待想把它用起来。根据对源码的研究，发现该模型在做nlu建模时主要用的encoder部分，也就是bert，因此我将这部分权重转为bert权重类型

1 Oct 14, 2021

PyTorch Implementation for Deep Metric Learning Pipelines

Easily Extendable Basic Deep Metric Learning Pipeline Karsten Roth ([email

543 Jan 04, 2023

SUPERVISED-CONTRASTIVE-LEARNING-FOR-PRE-TRAINED-LANGUAGE-MODEL-FINE-TUNING - The Facebook paper about fine tuning RoBERTa with contrastive loss

"# SUPERVISED-CONTRASTIVE-LEARNING-FOR-PRE-TRAINED-LANGUAGE-MODEL-FINE-TUNING" i

28 Dec 12, 2022

An implementation of Fastformer: Additive Attention Can Be All You Need in TensorFlow

Related tags

Overview

Fast Transformer

Installation

Developing fast-transformer

Usage

Want to Contribute 🙋‍♂️ ?

Want to discuss? 💬

Citation

License

You might also like...

Unofficial PyTorch implementation of Neural Additive Models (NAM) by Agarwal, et al.

PixelPick This is an official implementation of the paper "All you need are a few pixels: semantic segmentation with PixelPick."

Unofficial implementation of HiFi-GAN+ from the paper "Bandwidth Extension is All You Need" by Su, et al.

Torch-based tool for quantizing high-dimensional vectors using additive codebooks

Neural network-based build time estimation for additive manufacturing

Load What You Need: Smaller Multilingual Transformers for Pytorch and TensorFlow 2.0.

Code for "Diffusion is All You Need for Learning on Surfaces"

Comments

Implement Additive Attention

Releases(v0.2.0)

v0.2.0(Jan 16, 2022)

✅ Bug Fixes / Improvements

v0.1.0(Sep 3, 2021)

Classes

Owner

Rishit Dagli

Tightness-aware Evaluation Protocol for Scene Text Detection

Fast sparse deep learning on CPUs

This is the repo for the paper `SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization'. (published in Bioinformatics'21)

Image Super-Resolution Using Very Deep Residual Channel Attention Networks

This repository provides the official implementation of 'Learning to ignore: rethinking attention in CNNs' accepted in BMVC 2021.

Implementation of ICCV2021(Oral) paper - VMNet: Voxel-Mesh Network for Geodesic-aware 3D Semantic Segmentation

Converting CPT to bert form for use

PyTorch Implementation for Deep Metric Learning Pipelines

SUPERVISED-CONTRASTIVE-LEARNING-FOR-PRE-TRAINED-LANGUAGE-MODEL-FINE-TUNING - The Facebook paper about fine tuning RoBERTa with contrastive loss

FCA: Learning a 3D Full-coverage Vehicle Camouflage for Multi-view Physical Adversarial Attack

CTRL-C: Camera calibration TRansformer with Line-Classification

Attention-guided gan for synthesizing IR images

The official code repository for examples in the O'Reilly book 'Generative Deep Learning'

Implementation of ConvMixer in TensorFlow and Keras

Single-Stage Instance Shadow Detection with Bidirectional Relation Learning (CVPR 2021 Oral)

Solving SMPL/MANO parameters from keypoint coordinates.

Simple (but Strong) Baselines for POMDPs

Video-Music Transformer

An example of Scatterbrain implementation (combining local attention and Performer)

Mmdetection3d Noted - MMDetection3D is an open source object detection toolbox based on PyTorch