GithubHelp home page GithubHelp logo

liuchen1997 / rfaconv Goto Github PK

View Code? Open in Web Editor NEW
133.0 1.0 17.0 49.06 MB

RAFConv: Innovating Spatital Attention and Standard Convolutional Operation

Python 88.86% Shell 1.13% Jupyter Notebook 9.44% Dockerfile 0.57%

rfaconv's People

Contributors

cv-zhangxin avatar dr-yanzhiwei avatar li-ke666 avatar liuchen1997 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar

rfaconv's Issues

关于组数的问题

您好,感谢您的杰出工作!我想问一下您的resnet18和34中RFAConv这个组数是多少呢?

common文件中没有相关RFAConv的定义

hello,首先感谢您出色的工作,我想在yolov7的基础上复现一下相关实验,但是在common.py文件中没有找到相关的RFAConv的代码,在yolo.py文件中直接使用了,应该怎么解决?

YOLOv8

Why is this error occurring."RuntimeError: Given groups=1, weight of size [32, 16, 1, 1], expected input[1, 128, 64, 64] to have 16 channels, but got 128 channels instead"

CAConv模块

您好,请问CAConv模块您是参考的什么论文里边的呢,我想学习下CAConv模块理论部分,

关于adjust shape的问题

您好,非常感谢您的工作!
看到您的论文里关于adjust shape的描述是这样的:
image
但是我看到源码中对于shape的更改是这样的:

 generate_feature = rearrange(generate_feature, 'b c (n1 n2) h w -> b c (h n1) (w n2)', n1=self.kernel_size,
                         n2=self.kernel_size)

对于rearrange这个库我实践了一下是这样的结果:

from einops import rearrange
import torch


demo = torch.arange(36).view(9, 2, 2)
print('demo: ', demo)
out = rearrange(demo, '(k1 k2) w h -> 1 (k1 w) (k2 h)', k1=3, k2=3)
print('end=', out)

# output
demo:  tensor([[[ 0,  1],
         [ 2,  3]],

        [[ 4,  5],
         [ 6,  7]],

        [[ 8,  9],
         [10, 11]],

        [[12, 13],
         [14, 15]],

        [[16, 17],
         [18, 19]],

        [[20, 21],
         [22, 23]],

        [[24, 25],
         [26, 27]],

        [[28, 29],
         [30, 31]],

        [[32, 33],
         [34, 35]]])
end= tensor([[[ 0,  1,  4,  5,  8,  9],
         [ 2,  3,  6,  7, 10, 11],
         [12, 13, 16, 17, 20, 21],
         [14, 15, 18, 19, 22, 23],
         [24, 25, 28, 29, 32, 33],
         [26, 27, 30, 31, 34, 35]]])

这个结果与图中的adjust好像不一样,并不是小矩阵对应左上角的的元素全都集中在adjust后的左上区域,而是将9个小矩阵顺序排列的,没有改变矩阵内部的关系,这样的话并没有对图像做类似放大的操作,而只是将图像进行了排列,这样对于权重共享问题是为什么work的呢?
十分期待您的解答!!

关于resnet50中的问题

作者您好,我想问下:

  1. RFAConv是替换resnet50的Bottleneck中的33的卷积嘛
    2.在resnet34中为什么只替换掉BasicBlock第一个3
    3的卷积

About RFAConv3d

如果我想用在三维数据上,d,h,w, 我该修改哪些地方

关于CAconv

您好,非常感谢您和您团队的贡献,对于RFCAConv中kernel_size的大小的选取,3*3的效果是最好的吗?

not enough memory

我尝试复现论文中所提出的注意力模块,当我将复现的模块添加到yolov8中进行模型生成时,出现了以下错误
RuntimeError: [enforce fail at C:\cb\pytorch_1000000000000\work\c10\core\impl\alloc_cpu.cpp:81] data. DefaultCPUAllocator: not enough memory:
you tried to allocate 8589934592 bytes.

关于NovelConv模块与yolo.py文件的疑问?

作者您好,有两个疑问想请教一下:

(1)https://github.com/Liuchen1997/RFAConv/blob/main/Detection/YOLOv5/models/yolov5n.yaml%20(NovelConv) 中的NovelConv模块,是指 https://github.com/Liuchen1997/RFAConv/blob/main/Detection/YOLOv5/models/common.py 中的RFAConv的模块吗?

(2)https://github.com/Liuchen1997/RFAConv/tree/main/Detection/YOLOv5/models 文件夹下是否缺了yolo.py文件?能否分享一下该yolo.py文件?

Training the YOLOv5 and YOLOv8 with RFAConv

Dear Mr. @Liuchen1997,
I am trying to train the YOLOv5 and YOLOv8 architectures with your proposed modules but I got many errors.
Please provide the guide details to train the YOLOv5 and YOLOv8 architectures with RFAConv.
Thank you very much.
Best regards.

yolov8

hello,首先感谢您出色的工作,我想在yolov8的基础上复现一下相关实验,但是在文件中没有找到相关的RFAConv的代码

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.