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
Annotate datasets with a semi-trained or fully trained YOLOv5 model

YOLOv5 Auto Annotator Annotate datasets with a semi-trained or fully trained YOLOv5 model Prerequisites Ubuntu =20.04 Python =3.7 System dependencie

Akash James 3 May 14, 2022
[CVPR'21] Projecting Your View Attentively: Monocular Road Scene Layout Estimation via Cross-view Transformation

Projecting Your View Attentively: Monocular Road Scene Layout Estimation via Cross-view Transformation Weixiang Yang, Qi Li, Wenxi Liu, Yuanlong Yu, Y

118 Dec 26, 2022
Implementation of GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation (ICLR 2022).

GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation [OpenReview] [arXiv] [Code] The official implementation of GeoDiff: A Geome

Minkai Xu 155 Dec 26, 2022
This repository provides code for "On Interaction Between Augmentations and Corruptions in Natural Corruption Robustness".

On Interaction Between Augmentations and Corruptions in Natural Corruption Robustness This repository provides the code for the paper On Interaction B

Meta Research 33 Dec 08, 2022
CodeContests is a competitive programming dataset for machine-learning

CodeContests CodeContests is a competitive programming dataset for machine-learning. This dataset was used when training AlphaCode. It consists of pro

DeepMind 1.6k Jan 08, 2023
Air Pollution Prediction System using Linear Regression and ANN

AirPollution Pollution Weather Prediction System: Smart Outdoor Pollution Monitoring and Prediction for Healthy Breathing and Living Publication Link:

Dr Sharnil Pandya, Associate Professor, Symbiosis International University 19 Feb 07, 2022
Implementation of CSRL from the AAAI2022 paper: Constraint Sampling Reinforcement Learning: Incorporating Expertise For Faster Learning

CSRL Implementation of CSRL from the AAAI2022 paper: Constraint Sampling Reinforcement Learning: Incorporating Expertise For Faster Learning Python: 3

4 Apr 14, 2022
Here is the implementation of our paper S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations.

S2VC Here is the implementation of our paper S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations. In thi

81 Dec 15, 2022
PSGAN running with ncnn⚡妆容迁移/仿妆⚡Imitation Makeup/Makeup Transfer⚡

PSGAN running with ncnn⚡妆容迁移/仿妆⚡Imitation Makeup/Makeup Transfer⚡

WuJinxuan 144 Dec 26, 2022
Implementation of UNet on the Joey ML framework

Independent Research Project - Code Joey can be cloned from here https://github.com/devitocodes/joey/. Devito and other dependencies such as PyTorch a

Navjot Kukreja 1 Oct 21, 2021
Fully Convolutional DenseNets for semantic segmentation.

Introduction This repo contains the code to train and evaluate FC-DenseNets as described in The One Hundred Layers Tiramisu: Fully Convolutional Dense

485 Nov 26, 2022
Spatially-Adaptive Pixelwise Networks for Fast Image Translation, CVPR 2021

Image Translation with ASAPNets Spatially-Adaptive Pixelwise Networks for Fast Image Translation, CVPR 2021 Webpage | Paper | Video Installation insta

Tamar Rott Shaham 100 Dec 28, 2022
Official implementation of VQ-Diffusion

Vector Quantized Diffusion Model for Text-to-Image Synthesis Overview This is the official repo for the paper: [Vector Quantized Diffusion Model for T

Microsoft 592 Jan 03, 2023
An efficient PyTorch implementation of the winning entry of the 2017 VQA Challenge.

Bottom-Up and Top-Down Attention for Visual Question Answering An efficient PyTorch implementation of the winning entry of the 2017 VQA Challenge. The

Hengyuan Hu 731 Jan 03, 2023
The codebase for our paper "Generative Occupancy Fields for 3D Surface-Aware Image Synthesis" (NeurIPS 2021)

Generative Occupancy Fields for 3D Surface-Aware Image Synthesis (NeurIPS 2021) Project Page | Paper Xudong Xu, Xingang Pan, Dahua Lin and Bo Dai GOF

xuxudong 97 Nov 10, 2022
A library for augmentation of a YOLO-formated dataset

YOLO Dataset Augmentation lib Инструкция по использованию этой библиотеки Запуск всех файлов осуществлять из консоли. GoogleCrawl_to_Dataset.py Это ск

Egor Orel 1 Dec 10, 2022
Learning from Guided Play: A Scheduled Hierarchical Approach for Improving Exploration in Adversarial Imitation Learning Source Code

Learning from Guided Play: A Scheduled Hierarchical Approach for Improving Exploration in Adversarial Imitation Learning Source Code

STARS Laboratory 8 Sep 14, 2022
Mix3D: Out-of-Context Data Augmentation for 3D Scenes (3DV 2021)

Mix3D: Out-of-Context Data Augmentation for 3D Scenes (3DV 2021) Alexey Nekrasov*, Jonas Schult*, Or Litany, Bastian Leibe, Francis Engelmann Mix3D is

Alexey Nekrasov 189 Dec 26, 2022
PAIRED in PyTorch 🔥

PAIRED This codebase provides a PyTorch implementation of Protagonist Antagonist Induced Regret Environment Design (PAIRED), which was first introduce

UCL DARK Lab 46 Dec 12, 2022
The repository forked from NVlabs uses our data. (Differentiable rasterization applied to 3D model simplification tasks)

nvdiffmodeling [origin_code] Differentiable rasterization applied to 3D model simplification tasks, as described in the paper: Appearance-Driven Autom

Qiujie (Jay) Dong 2 Oct 31, 2022