Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

Overview

Perceiver - Pytorch

Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

Install

$ pip install perceiver-pytorch

Usage

import torch
from perceiver_pytorch import Perceiver

model = Perceiver(
    num_fourier_features = 6,    # number of fourier features, with original value (2 * K + 1)
    depth = 48,                  # depth of net, in paper, they went deep, making up for lack of attention
    num_latents = 6,             # number of latents, or induced set points, or centroids. different papers giving it different names
    cross_dim = 512,             # cross attention dimension
    latent_dim = 512,            # latent dimension
    cross_heads = 1,             # number of heads for cross attention. paper said 1
    latent_heads = 8,            # number of heads for latent self attention, 8
    cross_dim_head = 64,
    latent_dim_head = 64,
    num_classes = 1000,          # output number of classes
    attn_dropout = 0.,
    ff_dropout = 0.,
    weight_tie_layers = False    # whether to weight tie layers (optional, as indicated in the diagram)
)

img = torch.randn(1, 224 * 224) # 1 imagenet image, pixelized

model(img) # (1, 1000)

Citations

@misc{jaegle2021perceiver,
    title   = {Perceiver: General Perception with Iterative Attention},
    author  = {Andrew Jaegle and Felix Gimeno and Andrew Brock and Andrew Zisserman and Oriol Vinyals and Joao Carreira},
    year    = {2021},
    eprint  = {2103.03206},
    archivePrefix = {arXiv},
    primaryClass = {cs.CV}
}
Comments
  • Latent averaging to the logits?

    Latent averaging to the logits?

    I read through the paper last night and came away confused about a few things. I looked through your code hoping for some clarity.

    One issue that doesn't seem to be explained in the paper (or I am missing it) is how the authors go from a set of latents to the logits used at the classification head. You implemented this by taking the mean of the latent set:

    https://github.com/lucidrains/perceiver-pytorch/blob/main/perceiver_pytorch/perceiver_pytorch.py#L203

    Is this actually how the authors convert to logits?

    opened by neonbjb 7
  • PerceiverAR?

    PerceiverAR?

    Hey @lucidrains - love this repo, and still trying to wrap my head around the various difference between Perceiver architectures; how hard would it be to extend PerceiverIO to PerceiverAR; what fundamentally needs to change?

    opened by siddk 5
  • Not using the classification head in Perceiver

    Not using the classification head in Perceiver

    Hi @lucidrains, thank you for your great job!

    I'd like to use the Perceiver (not PerceiverIO) without the classification head (average and projection). Do you think we could add an option to avoid using it? I can do a PR if you want.

    Thanks!

    opened by gegallego 4
  • Decoder Attention Module needs a FF network as well in perceiver_io.py script

    Decoder Attention Module needs a FF network as well in perceiver_io.py script

    Hi,

    According to perceiver io paper's (https://arxiv.org/abs/2107.14795) architectural details, they mention that the decoder attention block contains a cross attention block (4), which is already implemented in the perceiver_io.py script (Line 151), followed by a Feedforward network, given by equation (6) in the paper, which is not present in that script. I am not aware of the repercussions of not having FF in the decoder module but it might be a good idea to have it in the implementation. Something like self.decoder_ff = PreNorm(FeedForward(queries_dim)) would do the job. Experimentally, the authors had found that omitting equation (5) is helpful.

    opened by Hritikbansal 4
  • Positional encoding are already part of the input

    Positional encoding are already part of the input

    Hello! First of all, thank you for this implementation.

    My inputs already have the proper positional encoding as part of the channel axis. Would it be possible to add a feature to deactivate the default implementation of the positional encoding?

    Thank you!

    opened by Atlis 4
  • x = self.latents + self.pos_emb

    x = self.latents + self.pos_emb

    self.latents = nn.Parameter(torch.randn(num_latents, latent_dim))
    self.pos_emb = nn.Parameter(torch.randn(num_latents, latent_dim))
    ...
    x = self.latents + self.pos_emb
    

    I'm not very familiar with pytorch, but does this make sense? I mean, what's intended when 2 trainable weight matrices are simply summed and that's that's the only place where both latents and pos_emb appear. It looks like it can be replaced with only one matrix.

    opened by galchinsky 4
  • Fourier encoding is not similar to the paper

    Fourier encoding is not similar to the paper

    First of all, thanks for sharing the code !

    I have a follow up question to #4.

    In the paper, the authors mentioned about [sin(f_kπx_d), cos(f_kπx_d)], where f_k is a bank of frequencies spaced log-linearly between 1 and µ/2. Can you maybe point out how you came to the 1/2**i scaling in the code ?

    https://github.com/lucidrains/perceiver-pytorch/blob/6ae733773d29cb29383f3ac7b45af8cb6bd2c0dc/perceiver_pytorch/perceiver_pytorch.py#L28-L35

    Thanks!

    opened by cheneeheng 4
  • Fourier encoding should be for position coordinates instead of byte array

    Fourier encoding should be for position coordinates instead of byte array

    The fourier_encode function as implemented takes as input a byte array x and directly encodes it with sin/cos before concating with the input.

    As I understand the NeRF position encodings, they encode the x/y/etc. position coordinates, and not a transformation of the data itself. From the Perceiver paper:

    We parametrize the frequency encoding to take the values [sin(fkπxd), cos(fkπxd)], where the frequencies fk is the kth band of a bank of frequencies spaced log-linearly between 1 and µ/2... For example, by allowing the network to resolve the maximum frequency present in an input array, we can encourage it to learn to compare the values of bytes at any positions in the input array. xd is the value of the input position along the dth dimension (e.g. for images d = 2 and for video d = 3). xd takes values in [−1, 1] for each dimension. We concatenate the raw positional value xd to produce the final representation of position. This results in a positional encoding of size d(2K + 1).

    NeRF position encoding examples:

    • https://github.com/bmild/nerf/blob/20a91e764a28816ee2234fcadb73bd59a613a44c/run_nerf_helpers.py#L22
    • https://github.com/ankurhanda/nerf2D
    opened by eridgd 4
  • Positional encoding frequency bands should be linearly spaced

    Positional encoding frequency bands should be linearly spaced

    A small bug, but as alluded to in this comment by @marcdumon, it seems as though the frequency bands are indeed spaced linearly in the official JAX implementation.

    opened by djl11 2
  • Bug in fourier_encode (?)

    Bug in fourier_encode (?)

    Thank you for this great implementation. I'm learning a lot from it!

    I think I found a problem in the fourier_encode method. In this line: https://github.com/lucidrains/perceiver-pytorch/blob/b33aced4e1b266aeb1383e03ab63f0a9951f9126/perceiver_pytorch/perceiver_pytorch.py#L36

    the scales are always the same whatever value of parameter base. Example:

    max_freq = 10, num_bands=6, base = 2
    => scales = [1.0000, 1.3797, 1.9037, 2.6265, 3.6239, 5.0000]
    
    max_freq = 10, num_bands=6, base = 10
    => scales = [1.0000, 1.3797, 1.9037, 2.6265, 3.6239, 5.0000]
    
    opened by marcdumon 2
  • Attention softmax is applied to incorrect dimension?

    Attention softmax is applied to incorrect dimension?

    I am studying multi-head attention. When I was reading through [1], I found that the attenion softmax is applied over the last dimension of the similarity tensor sim:

            q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> (b h) n d', h = h), (q, k, v))
    
            sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
    
            if exists(mask):
                <removed>
    
            # attention, what we cannot get enough of
            attn = sim.softmax(dim = -1)
    

    If I understand correctly sim has the shape (b*h) n1 n2. The softmax is computed over the last dimension n2. Shouldn't the softmax be applied to matrices with all the similarity values of a single head (i.e. with shape n1, n2)?

    [1] https://github.com/lucidrains/perceiver-pytorch/blob/main/perceiver_pytorch/perceiver_io.py#L97

    opened by breuderink 2
  • Issue defining base in fourier_encode for experimental.py, gated.py, mixed_latents.py

    Issue defining base in fourier_encode for experimental.py, gated.py, mixed_latents.py

    Hey Lucid, love the work, it appears you deprecated base in fourier_encode at https://github.com/lucidrains/perceiver-pytorch/commit/144b0d9716a7212b5fd6d95a2267c4d4a08b56a7

    But experimental.py, gated.py, mixed_latents.py are still trying to define the base within the forward pass. https://github.com/lucidrains/perceiver-pytorch/blob/abbb5d5949d3509c57749bd134f5068f2761aac7/perceiver_pytorch/experimental.py#L122 https://github.com/lucidrains/perceiver-pytorch/blob/2d59df42ebb0b7538af77d584f5ae5b50759618b/perceiver_pytorch/mixed_latents.py#L85 https://github.com/lucidrains/perceiver-pytorch/blob/2d59df42ebb0b7538af77d584f5ae5b50759618b/perceiver_pytorch/gated.py#L103

    Thanks again, keep up the great work.

    opened by TannerLaBorde 0
  • Audio + Text data?

    Audio + Text data?

    Can someone please guide me on how you can process both audio and .txt data through perceiver simultaneously for multimodality learning?

    An example code would be nice.

    Thanks

    opened by Sidz1812 1
  • just a suggestion

    just a suggestion

    Hi I like to start with thanking you for such a great work with a lot of great implementations. I have a small suggestion. I suggest for all your codes/modules try to add if __name__ == "__main__": so that if someone just wants to use one file/module can easily try that without having going through whole implementations. for example I am trying to use the this, in case of having a if __name__ == "__main__": I can easily try to run a random input and see how it will work. This will increase the usability with a huge amount.

    Keep up the great work :)

    opened by seyeeet 4
  • What should I change if I want to use data with input size 720*184

    What should I change if I want to use data with input size 720*184

    thanks for sharing this code, I was wondering what should I change if I want to be able to use data that can be converted into images with an input size of 720*184? thanks in advance

    opened by Oussamab21 0
  • Question regarding queries dimensionality in Perceiver IO

    Question regarding queries dimensionality in Perceiver IO

    Hi @lucidrains,

    I think I may be missing something - why do we define the perceiver IO queries vector to have a batch dimension (i.e. queries = torch.randn(1, 128, 32))? Was this just to make the code work nicely? Shouldnt we be using queries = torch.randn(128, 32) ? I expect to use the same embedding for all of my batch elements, which is IIUC what your code is doing.

    opened by pcicales 3
Releases(0.8.6)
Owner
Phil Wang
Working with Attention. It's all we need.
Phil Wang
Trading Strategies for Freqtrade

Freqtrade Strategies Strategies for Freqtrade, developed primarily in a partnership between @werkkrew and @JimmyNixx from the Freqtrade Discord. Use t

Bryan Chain 242 Jan 07, 2023
B-cos Networks: Attention is All we Need for Interpretability

Convolutional Dynamic Alignment Networks for Interpretable Classifications M. Böhle, M. Fritz, B. Schiele. B-cos Networks: Alignment is All we Need fo

58 Dec 23, 2022
Analyzing basic network responses to novel classes

novelty-detection Analyzing how AlexNet responds to novel classes with varying degrees of similarity to pretrained classes from ImageNet. If you find

Noam Eshed 34 Oct 02, 2022
Spatial-Location-Constraint-Prototype-Loss-for-Open-Set-Recognition

Spatial Location Constraint Prototype Loss for Open Set Recognition Official PyTorch implementation of "Spatial Location Constraint Prototype Loss for

Xia Ziheng 12 Jun 24, 2022
noisy labels; missing labels; semi-supervised learning; entropy; uncertainty; robustness and generalisation.

ProSelfLC: CVPR 2021 ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks For any specific discussion or potential fu

amos_xwang 57 Dec 04, 2022
Code for our work "Activation to Saliency: Forming High-Quality Labels for Unsupervised Salient Object Detection".

A2S-USOD Code for our work "Activation to Saliency: Forming High-Quality Labels for Unsupervised Salient Object Detection". Code will be released upon

15 Dec 16, 2022
Deep Ensemble Learning with Jet-Like architecture

Ransomware analysis using DEL with jet-like architecture comprising two CNN wings, a sparse AE tail, a non-linear PCA to produce a diverse feature space, and an MLP nose

Ahsen Nazir 2 Feb 06, 2022
A PyTorch implementation of the continual learning experiments with deep neural networks

Brain-Inspired Replay A PyTorch implementation of the continual learning experiments with deep neural networks described in the following paper: Brain

182 Dec 27, 2022
The source codes for ACL 2021 paper 'BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data'

BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data This repository provides the implementation details for

124 Dec 27, 2022
The official repository for our paper "The Neural Data Router: Adaptive Control Flow in Transformers Improves Systematic Generalization".

Codebase for learning control flow in transformers The official repository for our paper "The Neural Data Router: Adaptive Control Flow in Transformer

Csordás Róbert 24 Oct 15, 2022
Implementations of the algorithms in the paper Approximative Algorithms for Multi-Marginal Optimal Transport and Free-Support Wasserstein Barycenters

Implementations of the algorithms in the paper Approximative Algorithms for Multi-Marginal Optimal Transport and Free-Support Wasserstein Barycenters

Johannes von Lindheim 3 Oct 29, 2022
2D&3D human pose estimation

Human Pose Estimation Papers [CVPR 2016] - 201511 [IJCAI 2016] - 201602 Other Action Recognition with Joints-Pooled 3D Deep Convolutional Descriptors

133 Jan 02, 2023
Official implementation of the paper "Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering"

Light Field Networks Project Page | Paper | Data | Pretrained Models Vincent Sitzmann*, Semon Rezchikov*, William Freeman, Joshua Tenenbaum, Frédo Dur

Vincent Sitzmann 130 Dec 29, 2022
Curating a dataset for bioimage transfer learning

CytoImageNet A large-scale pretraining dataset for bioimage transfer learning. Motivation In past few decades, the increase in speed of data collectio

Stanley Z. Hua 9 Jun 20, 2022
We provided a matlab implementation for an evolutionary multitasking AUC optimization framework (EMTAUC).

EMTAUC We provided a matlab implementation for an evolutionary multitasking AUC optimization framework (EMTAUC). In this code, SBGA is considered a ba

7 Nov 24, 2022
CT-Net: Channel Tensorization Network for Video Classification

[ICLR2021] CT-Net: Channel Tensorization Network for Video Classification @inproceedings{ li2021ctnet, title={{\{}CT{\}}-Net: Channel Tensorization Ne

33 Nov 15, 2022
Source code for Transformer-based Multi-task Learning for Disaster Tweet Categorisation (UCD's participation in TREC-IS 2020A, 2020B and 2021A).

Source code for "UCD participation in TREC-IS 2020A, 2020B and 2021A". *** update at: 2021/05/25 This repo so far relates to the following work: Trans

Congcong Wang 4 Oct 19, 2021
🐦 Opytimizer is a Python library consisting of meta-heuristic optimization techniques.

Opytimizer: A Nature-Inspired Python Optimizer Welcome to Opytimizer. Did you ever reach a bottleneck in your computational experiments? Are you tired

Gustavo Rosa 546 Dec 31, 2022
🔥RandLA-Net in Tensorflow (CVPR 2020, Oral & IEEE TPAMI 2021)

RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds (CVPR 2020) This is the official implementation of RandLA-Net (CVPR2020, Oral

Qingyong 1k Dec 30, 2022
This is the official repository for our paper: ''Pruning Self-attentions into Convolutional Layers in Single Path''.

Pruning Self-attentions into Convolutional Layers in Single Path This is the official repository for our paper: Pruning Self-attentions into Convoluti

Zhuang AI Group 77 Dec 26, 2022