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
Human pose estimation from video plays a critical role in various applications such as quantifying physical exercises, sign language recognition, and full-body gesture control.

Pose Detection Project Description: Human pose estimation from video plays a critical role in various applications such as quantifying physical exerci

Hassan Shahzad 2 Jan 17, 2022
PyTorch implementation of the Value Iteration Networks (VIN) (NIPS '16 best paper)

Value Iteration Networks in PyTorch Tamar, A., Wu, Y., Thomas, G., Levine, S., and Abbeel, P. Value Iteration Networks. Neural Information Processing

LEI TAI 75 Nov 24, 2022
I-BERT: Integer-only BERT Quantization

I-BERT: Integer-only BERT Quantization HuggingFace Implementation I-BERT is also available in the master branch of HuggingFace! Visit the following li

Sehoon Kim 139 Dec 27, 2022
AI Based Smart Exam Proctoring Package

AI Based Smart Exam Proctoring Package It takes image (base64) as input: Provide Output as: Detection of Mobile phone. Detection of More than 1 person

NARENDER KESWANI 3 Sep 09, 2022
Setup and customize deep learning environment in seconds.

Deepo is a series of Docker images that allows you to quickly set up your deep learning research environment supports almost all commonly used deep le

Ming 6.3k Jan 06, 2023
Directed Greybox Fuzzing with AFL

AFLGo: Directed Greybox Fuzzing AFLGo is an extension of American Fuzzy Lop (AFL). Given a set of target locations (e.g., folder/file.c:582), AFLGo ge

380 Nov 24, 2022
Generative Modelling of BRDF Textures from Flash Images [SIGGRAPH Asia, 2021]

Neural Material Official code repository for the paper: Generative Modelling of BRDF Textures from Flash Images [SIGGRAPH Asia, 2021] Henzler, Deschai

Philipp Henzler 80 Dec 20, 2022
Differentiable Quantum Chemistry (only Differentiable Density Functional Theory and Hartree Fock at the moment)

DQC: Differentiable Quantum Chemistry Differentiable quantum chemistry package. Currently only support differentiable density functional theory (DFT)

75 Dec 02, 2022
Density-aware Single Image De-raining using a Multi-stream Dense Network (CVPR 2018)

DID-MDN Density-aware Single Image De-raining using a Multi-stream Dense Network He Zhang, Vishal M. Patel [Paper Link] (CVPR'18) We present a novel d

He Zhang 224 Dec 12, 2022
Scales, Chords, and Cadences: Practical Music Theory for MIR Researchers

ISMIR-musicTheoryTutorial This repository has slides and Jupyter notebooks for the ISMIR 2021 tutorial Scales, Chords, and Cadences: Practical Music T

Johanna Devaney 58 Oct 11, 2022
SANet: A Slice-Aware Network for Pulmonary Nodule Detection

SANet: A Slice-Aware Network for Pulmonary Nodule Detection This paper (SANet) has been accepted and early accessed in IEEE TPAMI 2021. This code and

Jie Mei 39 Dec 17, 2022
YuNetのPythonでのONNX、TensorFlow-Lite推論サンプル

YuNet-ONNX-TFLite-Sample YuNetのPythonでのONNX、TensorFlow-Lite推論サンプルです。 TensorFlow-LiteモデルはPINTO0309/PINTO_model_zoo/144_YuNetのものを使用しています。 Requirement Op

KazuhitoTakahashi 8 Nov 17, 2021
This repository contains the DendroMap implementation for scalable and interactive exploration of image datasets in machine learning.

DendroMap DendroMap is an interactive tool to explore large-scale image datasets used for machine learning. A deep understanding of your data can be v

DIV Lab 33 Dec 30, 2022
AI-UPV at IberLEF-2021 EXIST task: Sexism Prediction in Spanish and English Tweets Using Monolingual and Multilingual BERT and Ensemble Models

AI-UPV at IberLEF-2021 EXIST task: Sexism Prediction in Spanish and English Tweets Using Monolingual and Multilingual BERT and Ensemble Models Descrip

Angel de Paula 1 Jun 08, 2022
Detection of PCBA defect

Detection_of_PCBA_defect Detection_of_PCBA_defect Use yolov5 to train. $pip install -r requirements.txt Detect.py will detect file(jpg,mp4...) in cu

6 Nov 28, 2022
Code for paper "A Critical Assessment of State-of-the-Art in Entity Alignment" (https://arxiv.org/abs/2010.16314)

A Critical Assessment of State-of-the-Art in Entity Alignment This repository contains the source code for the paper A Critical Assessment of State-of

Max Berrendorf 16 Oct 14, 2022
Unofficial PyTorch implementation of Google AI's VoiceFilter system

VoiceFilter Note from Seung-won (2020.10.25) Hi everyone! It's Seung-won from MINDs Lab, Inc. It's been a long time since I've released this open-sour

MINDs Lab 883 Jan 07, 2023
Repo for the ACMMM20 submission: "Personalized breath based biometric authentication with wearable multimodality".

personalized-breath Repo for the ACMMM20 submission: "Personalized breath based biometric authentication with wearable multimodality". Guideline To ex

Manh-Ha Bui 2 Nov 15, 2021
The project covers common metrics for super-resolution performance evaluation.

Super-Resolution Performance Evaluation Code The project covers common metrics for super-resolution performance evaluation. Metrics support The script

xmy 10 Aug 03, 2022
Implementation of Hire-MLP: Vision MLP via Hierarchical Rearrangement and An Image Patch is a Wave: Phase-Aware Vision MLP.

Hire-Wave-MLP.pytorch Implementation of Hire-MLP: Vision MLP via Hierarchical Rearrangement and An Image Patch is a Wave: Phase-Aware Vision MLP Resul

Nevermore 29 Oct 28, 2022