Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"

Related tags

Deep Learningdeep-learningartificial-intelligencememory-efficientattention-mechanism
Overview

Memory Efficient Attention Pytorch

Implementation of a memory efficient multi-head attention as proposed in the paper, Self-attention Does Not Need O(n²) Memory. In addition, the module will take care of masking, causal masking, as well as cross attention.

Install

$ pip install memory-efficient-attention-pytorch

Usage

For autoregressive language model

import torch
from memory_efficient_attention_pytorch import Attention

attn = Attention(
    dim = 512,
    dim_head = 64,                # dimension per head
    heads = 8,                    # number of attention heads
    causal = True,                # autoregressive or not
    memory_efficient = True,      # whether to use memory efficient attention (can be turned off to test against normal attention)
    q_bucket_size = 1024,         # bucket size along queries dimension
    k_bucket_size = 2048          # bucket size along key / values dimension
).cuda()

x = torch.randn(1, 65536, 512).cuda()
out = attn(x) # (1, 65536, 512)

Cross attention

import torch
from memory_efficient_attention_pytorch import Attention

cross_attn = Attention(
    dim = 512,
    dim_head = 64,
    heads = 8,
    memory_efficient = True,
    q_bucket_size = 1024,
    k_bucket_size = 2048
).cuda()

x = torch.randn(1, 65536, 512).cuda()
context = torch.randn(1, 65536, 512).cuda()
mask = torch.ones(1, 65536).bool().cuda()

out = cross_attn(x, context = context, mask = mask) # (1, 65536, 512)
  • benchmark and see how much torch jit helps
  • look at Triton and Keops and see if either can be a fit

Citations

@misc{rabe2021selfattention,
    title   = {Self-attention Does Not Need $O(n^2)$ Memory}, 
    author  = {Markus N. Rabe and Charles Staats},
    year    = {2021},
    eprint  = {2112.05682},
    archivePrefix = {arXiv},
    primaryClass = {cs.LG}
}
@misc{liu2021swin,
    title   = {Swin Transformer V2: Scaling Up Capacity and Resolution},
    author  = {Ze Liu and Han Hu and Yutong Lin and Zhuliang Yao and Zhenda Xie and Yixuan Wei and Jia Ning and Yue Cao and Zheng Zhang and Li Dong and Furu Wei and Baining Guo},
    year    = {2021},
    eprint  = {2111.09883},
    archivePrefix = {arXiv},
    primaryClass = {cs.CV}
}
Comments
  • [feature request] Combining with flash attention?

    [feature request] Combining with flash attention?

    There is a new algorithm to optimize the qkv attention, https://github.com/HazyResearch/flash-attention https://arxiv.org/abs/2205.14135 It optimises the qkv attention part. Maybe you can look into integrating it with this.

    opened by Vbansal21 15
  • i did this, we could build on top

    i did this, we could build on top

    Hi there!

    It seems I did already some of the code... https://github.com/CHARM-Tx/linear_mem_attention_pytorch could we build on top of this? I talked to https://github.com/Chillee about an experimental functionality from functorch: https://github.com/pytorch/functorch that would allow for increased speed (mainly i want to match jax perofmance but its just difficult w/ pytorch imperative style).

    I would love to collaborate on this if you want!

    opened by hypnopump 5
  • Added dropout support to memory efficient variant

    Added dropout support to memory efficient variant

    Hey Phil,

    I have been using this repository for a project and I wanted to add dropout for completeness. I checked consistency with perceiver-ar impl.. I hope this is helpful.

    -Matt

    opened by usryokousha 2
  • Making this work with relative position bias from XTransformers

    Making this work with relative position bias from XTransformers

    Is there a way to make this work with RelativePositionBias. Currently this produces an attention bias of size $BHN^2$ where B is batch size, H is number of heads and N is input size. Can this be chunked and computed per chunk?

    opened by pfeatherstone 5
  • save_for_backward can only save variables, but argument 5 is of type bool

    save_for_backward can only save variables, but argument 5 is of type bool

    Hi,

    Thank you for your indescribable work. I was trying to test your method specifically for cross-attention but It seems I get the error " save_for_backward can only save variables, but argument 5 is of type bool". I am not sure what I am doing wrong. I tried your own examples too but get the same error.

    Can you please help me out?

    Code:

    import torch from memory_efficient_attention_pytorch import Attention

    cross_attn = Attention( dim = 512, dim_head = 64, heads = 8, memory_efficient = True, q_bucket_size = 1024, k_bucket_size = 2048 ).cuda() (# out = sm_mod(inp1)) did this to avoid being a header x = torch.randn(1, 65536, 512).cuda() context = torch.randn(1, 65536, 512).cuda() (# mask = torch.ones(1, 65536).bool().cuda()) did this to avoid being a heading out = cross_attn(x

    ERROR:

    File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/home/abali/.vscode-server/extensions/ms-python.python-2022.8.1/pythonFiles/lib/python/debugpy/main.py", line 45, in cli.main() File "/home/abali/.vscode-server/extensions/ms-python.python-2022.8.1/pythonFiles/lib/python/debugpy/../debugpy/server/cli.py", line 444, in main run() File "/home/abali/.vscode-server/extensions/ms-python.python-2022.8.1/pythonFiles/lib/python/debugpy/../debugpy/server/cli.py", line 285, in run_file runpy.run_path(target_as_str, run_name=compat.force_str("main")) File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/runpy.py", line 265, in run_path return _run_module_code(code, init_globals, run_name, File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/runpy.py", line 97, in _run_module_code _run_code(code, mod_globals, init_globals, File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/data/stars/user/abali/Phd_work/ISBI2023/X3D-Multigrid/CrossAttn_X3d_v2.py", line 872, in out = cross_attn(x, context = context, mask = mask) # (1, 65536, 512) print(out) File "/home/abali/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/site-packages/memory_efficient_attention_pytorch/memory_efficient_attention.py", line 215, in forward out = attn_fn(q, k, v, mask = mask, attn_bias = attn_bias, causal = self.causal, q_bucket_size = q_bucket_size, k_bucket_size = k_bucket_size) File "/home/abali/.conda/envs/py38_ydp5/lib/python3.8/site-packages/memory_efficient_attention_pytorch/memory_efficient_attention.py", line 127, in memory_efficient_attention exp_weight_chunk, weighted_value_chunk, weight_max_chunk = summarize_qkv_fn( File "/home/abali/.local/lib/python3.8/site-packages/torch/utils/checkpoint.py", line 163, in checkpoint return CheckpointFunction.apply(function, preserve, *args) TypeError: save_for_backward can only save variables, but argument 5 is of type bool

    opened by aliabid2243 1
  • Checkpointing is not compatible with .grad() or when an `inputs` parameter is passed to .backward()

    Checkpointing is not compatible with .grad() or when an `inputs` parameter is passed to .backward()

    https://github.com/lucidrains/memory-efficient-attention-pytorch/blob/35559a05572f9d4eb982a8e2e399b40a2d61b85c/memory_efficient_attention_pytorch/memory_efficient_attention.py#L95

    Should this be: summarize_qkv_fn = summarize_qkv_chunk if needs_backwards else checkpointed_summarize_qkv_chunk instead of: summarize_qkv_fn = checkpointed_summarize_qkv_chunk if needs_backwards else summarize_qkv_chunk

    opened by vrobot 0
Releases(0.1.1)
Owner
Phil Wang
Working with Attention. It's all we need
Phil Wang
GitHub Repository
face property detection pytorch

This is the face property train code of project face-detection-project

i am x 2 Oct 18, 2021
Dense matching library based on PyTorch

Dense Matching A general dense matching library based on PyTorch. For any questions, issues or recommendations, please contact Prune at

Prune Truong 399 Dec 28, 2022
Implementation of "Large Steps in Inverse Rendering of Geometry"

Large Steps in Inverse Rendering of Geometry ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia), December 2021. Baptiste Nicolet · Alec Jacob

RGL: Realistic Graphics Lab 274 Jan 06, 2023
MOpt-AFL provided by the paper "MOPT: Optimized Mutation Scheduling for Fuzzers"

MOpt-AFL 1. Description MOpt-AFL is a AFL-based fuzzer that utilizes a customized Particle Swarm Optimization (PSO) algorithm to find the optimal sele

172 Dec 18, 2022
CL-Gym: Full-Featured PyTorch Library for Continual Learning

CL-Gym: Full-Featured PyTorch Library for Continual Learning CL-Gym is a small yet very flexible library for continual learning research and developme

Iman Mirzadeh 36 Dec 25, 2022
The implementation for "Comprehensive Knowledge Distillation with Causal Intervention".

Comprehensive Knowledge Distillation with Causal Intervention This repository is a PyTorch implementation of "Comprehensive Knowledge Distillation wit

Xiang Deng 10 Nov 03, 2022
This is an official repository of CLGo: Learning to Predict 3D Lane Shape and Camera Pose from a Single Image via Geometry Constraints

CLGo This is an official repository of CLGo: Learning to Predict 3D Lane Shape and Camera Pose from a Single Image via Geometry Constraints An earlier

刘芮金 32 Dec 20, 2022
A python library to artfully visualize Factorio Blueprints and an interactive web demo for using it.

Factorio Blueprint Visualizer I love the game Factorio and I really like the look of factories after growing for many hours or blueprints after tweaki

Piet Brömmel 124 Jan 07, 2023
Rank1 Conversation Emotion Detection Task

Rank1-Conversation_Emotion_Detection_Task accuracy macro-f1 recall 0.826 0.7544 0.719 基于预训练模型和时序预测模型的对话情感探测任务 1 摘要 针对对话情感探测任务，本文将其分为文本分类和时间序列预测两个子任务，分

Yuchen Han 2 Nov 28, 2021
Implementation of paper "DeepTag: A General Framework for Fiducial Marker Design and Detection"

Implementation of paper DeepTag: A General Framework for Fiducial Marker Design and Detection. Project page: https://herohuyongtao.github.io/research/

Yongtao Hu 46 Dec 12, 2022
The challenge for Quantum Coalition Hackathon 2021

Qchack 2021 Google Challenge This is a challenge for the brave 2021 qchack.io participants. Instructions Hello, intrepid qchacker, welcome to the G|o

quantumlib 18 May 04, 2022
System-oriented IR evaluations are limited to rather abstract understandings of real user behavior

Validating Simulations of User Query Variants This repository contains the scripts of the experiments and evaluations, simulated queries, as well as t

IR Group at Technische Hochschule Köln 2 Nov 23, 2022
Pre-trained Deep Learning models and demos (high quality and extremely fast)

OpenVINO™ Toolkit - Open Model Zoo repository This repository includes optimized deep learning models and a set of demos to expedite development of hi

OpenVINO Toolkit 3.4k Dec 31, 2022
The mini-AlphaStar (mini-AS, or mAS) - mini-scale version (non-official) of the AlphaStar (AS)

A mini-scale reproduction code of the AlphaStar program. Note: the original AlphaStar is the AI proposed by DeepMind to play StarCraft II.

Ruo-Ze Liu 216 Jan 04, 2023
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Super Resolution Examples We run this script under TensorFlow 2.0 and the TensorLayer2.0+. For TensorLayer 1.4 version, please check release. 🚀 🚀 🚀

TensorLayer Community 2.9k Jan 08, 2023
In this work, we will implement some basic but important algorithm of machine learning step by step.

WoRkS continued English 中文 Français Probability Density Estimation-Non-Parametric Methods(概率密度估计-非参数方法) 1. Kernel / k-Nearest Neighborhood Density Est

liziyu0104 1 Dec 30, 2021
PFLD pytorch Implementation

PFLD-pytorch Implementation of PFLD A Practical Facial Landmark Detector by pytorch. 1. install requirements pip3 install -r requirements.txt 2. Datas

zhaozhichao 669 Jan 02, 2023
Learning Representations that Support Robust Transfer of Predictors

Transfer Risk Minimization (TRM) Code for Learning Representations that Support Robust Transfer of Predictors Prepare the Datasets Preprocess the Scen

Yilun Xu 15 Dec 07, 2022
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection

CLOCs is a novel Camera-LiDAR Object Candidates fusion network. It provides a low-complexity multi-modal fusion framework that improves the performance of single-modality detectors. CLOCs operates on

Su Pang 254 Dec 16, 2022
MoveNet Single Pose on OpenVINO

MoveNet Single Pose tracking on OpenVINO Running Google MoveNet Single Pose models on OpenVINO. A convolutional neural network model that runs on RGB

35 Nov 11, 2022