2024年最新GitHub 7，6年老网络安全面经总结

给大家的福利

零基础入门

对于从来没有接触过网络安全的同学，我们帮你准备了详细的学习成长路线图。可以说是最科学最系统的学习路线，大家跟着这个大的方向学习准没问题。

同时每个成长路线对应的板块都有配套的视频提供：

因篇幅有限，仅展示部分资料

网络安全面试题

绿盟护网行动

还有大家最喜欢的黑客技术

网络安全源码合集+工具包

所有资料共282G，朋友们如果有需要全套《网络安全入门+黑客进阶学习资源包》，可以扫描下方二维码领取（如遇扫码问题，可以在评论区留言领取哦）~

网上学习资料一大堆，但如果学到的知识不成体系，遇到问题时只是浅尝辄止，不再深入研究，那么很难做到真正的技术提升。

需要这份系统化资料的朋友，可以点击这里获取

一个人可以走的很快，但一群人才能走的更远！不论你是正从事IT行业的老鸟或是对IT行业感兴趣的新人，都欢迎加入我们的的圈子（技术交流、学习资源、职场吐槽、大厂内推、面试辅导），让我们一起学习成长！

emb_dropout = 0.1

)

img = torch.randn(1, 3, 256, 256)

preds = v(img) # (1, 1000)

## 门槛值


[本文](https://bbs.csdn.net/topics/618540462)还指出了在更深层次上训练视觉转换器的困难，并提出了两种解决方案。首先，它建议对残差块的输出进行每通道乘法。其次，它建议让补丁互相关注，并且只允许 CLS 令牌关注最后几层的补丁。


他们还添加了[Talking Heads](https://bbs.csdn.net/topics/618540462)，注意到改进。


您可以按如下方式使用此方案：

import torch
from vit_pytorch.cait import CaiT

v = CaiT(
image_size = 256,
patch_size = 32,
num_classes = 1000,
dim = 1024,
depth = 12, # depth of transformer for patch to patch attention only
cls_depth = 2, # depth of cross attention of CLS tokens to patch
heads = 16,
mlp_dim = 2048,
dropout = 0.1,
emb_dropout = 0.1,
layer_dropout = 0.05 # randomly dropout 5% of the layers
)

img = torch.randn(1, 3, 256, 256)

preds = v(img) # (1, 1000)

## 代币对代币 ViT


![在这里插入图片描述](https://img-blog.csdnimg.cn/c2606b7c84b84c598198d907dc2e80ed.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_12,color_FFFFFF,t_70,g_se,x_16)  
 [本文](https://bbs.csdn.net/topics/618540462)提出前几层应该通过展开对图像序列进行下采样，导致每个标记中的图像数据重叠，如上图所示。您可以ViT按如下方式使用此变体。

import torch
from vit_pytorch.t2t import T2TViT

v = T2TViT(
dim = 512,
image_size = 224,
depth = 5,
heads = 8,
mlp_dim = 512,
num_classes = 1000,
t2t_layers = ((7, 4), (3, 2), (3, 2)) # tuples of the kernel size and stride of each consecutive layers of the initial token to token module
)

img = torch.randn(1, 3, 224, 224)

preds = v(img) # (1, 1000)

## CCT


![在这里插入图片描述](https://img-blog.csdnimg.cn/2d2c95f71ca245a9bdaa752251b8506d.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_20,color_FFFFFF,t_70,g_se,x_16)  
 CCT通过使用卷积而不是修补和执行序列池来提出紧凑型转换器。这使得 CCT 具有高精度和少量参数。


您可以通过两种方法使用它：

import torch
from vit_pytorch.cct import CCT

model = CCT(
img_size=224,
embedding_dim=384,
n_conv_layers=2,
kernel_size=7,
stride=2,
padding=3,
pooling_kernel_size=3,
pooling_stride=2,
pooling_padding=1,
num_layers=14,
num_heads=6,
mlp_radio=3.,
num_classes=1000,
positional_embedding=‘learnable’, # [‘sine’, ‘learnable’, ‘none’]
)

或者，您可以使用多个预定义模型之一，这些模型`[2,4,6,7,8,14,16]` 预定义了层数、注意力头数、mlp 比率和嵌入维度。

import torch
from vit_pytorch.cct import cct_14

model = cct_14(
img_size=224,
n_conv_layers=1,
kernel_size=7,
stride=2,
padding=3,
pooling_kernel_size=3,
pooling_stride=2,
pooling_padding=1,
num_classes=1000,
positional_embedding=‘learnable’, # [‘sine’, ‘learnable’, ‘none’]
)

[官方存储库](https://bbs.csdn.net/topics/618540462)包括到经过训练的模型检查点的链接。


## 交叉 ViT


![在这里插入图片描述](https://img-blog.csdnimg.cn/78c0e85947bb4a3681b0acc2bc33c28f.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_11,color_FFFFFF,t_70,g_se,x_16)


[本文](https://bbs.csdn.net/topics/618540462)建议让两个视觉转换器处理不同尺度的图像，每隔一段时间交叉处理一个。他们展示了在基础视觉转换器之上的改进。

import torch
from vit_pytorch.cross_vit import CrossViT

v = CrossViT(
image_size = 256,
num_classes = 1000,
depth = 4, # number of multi-scale encoding blocks
sm_dim = 192, # high res dimension
sm_patch_size = 16, # high res patch size (should be smaller than lg_patch_size)
sm_enc_depth = 2, # high res depth
sm_enc_heads = 8, # high res heads
sm_enc_mlp_dim = 2048, # high res feedforward dimension
lg_dim = 384, # low res dimension
lg_patch_size = 64, # low res patch size
lg_enc_depth = 3, # low res depth
lg_enc_heads = 8, # low res heads
lg_enc_mlp_dim = 2048, # low res feedforward dimensions
cross_attn_depth = 2, # cross attention rounds
cross_attn_heads = 8, # cross attention heads
dropout = 0.1,
emb_dropout = 0.1
)

img = torch.randn(1, 3, 256, 256)

pred = v(img) # (1, 1000)

## PiT


![在这里插入图片描述](https://img-blog.csdnimg.cn/9875bc66e64f4551b091dcafd44cddca.png)


[本文](https://bbs.csdn.net/topics/618540462)建议通过使用深度卷积的池化过程对令牌进行下采样。

import torch
from vit_pytorch.pit import PiT

v = PiT(
image_size = 224,
patch_size = 14,
dim = 256,
num_classes = 1000,
depth = (3, 3, 3), # list of depths, indicating the number of rounds of each stage before a downsample
heads = 16,
mlp_dim = 2048,
dropout = 0.1,
emb_dropout = 0.1
)

forward pass now returns predictions and the attention maps

img = torch.randn(1, 3, 224, 224)

preds = v(img) # (1, 1000)

## LeViT


![在这里插入图片描述](https://img-blog.csdnimg.cn/46b99248415b4fb7a7b52ee24b1f1a26.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_8,color_FFFFFF,t_70,g_se,x_16)  
 这篇[论文](https://bbs.csdn.net/topics/618540462)提出了一些变化，包括（1）卷积嵌入而不是逐块投影（2）阶段中的下采样（3）注意力中的额外非线性（4）二维相对位置偏差而不是初始绝对位置偏差（5 ) 批范数代替层范数。


[官方仓库](https://bbs.csdn.net/topics/618540462)

import torch
from vit_pytorch.levit import LeViT

levit = LeViT(
image_size = 224,
num_classes = 1000,
stages = 3, # number of stages
dim = (256, 384, 512), # dimensions at each stage
depth = 4, # transformer of depth 4 at each stage
heads = (4, 6, 8), # heads at each stage
mlp_mult = 2,
dropout = 0.1
)

img = torch.randn(1, 3, 224, 224)

levit(img) # (1, 1000)

## CvT


![在这里插入图片描述](https://img-blog.csdnimg.cn/4380d5ac2b5d42a383170721dbbd14fb.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_18,color_FFFFFF,t_70,g_se,x_16)  
 [本文](https://bbs.csdn.net/topics/618540462)提出混合卷积和注意力。具体来说，卷积用于分三个阶段嵌入和下采样图像/特征图。深度卷积还用于投影查询、键和值以引起注意。

import torch
from vit_pytorch.cvt import CvT

v = CvT(
num_classes = 1000,
s1_emb_dim = 64, # stage 1 - dimension
s1_emb_kernel = 7, # stage 1 - conv kernel
s1_emb_stride = 4, # stage 1 - conv stride
s1_proj_kernel = 3, # stage 1 - attention ds-conv kernel size
s1_kv_proj_stride = 2, # stage 1 - attention key / value projection stride
s1_heads = 1, # stage 1 - heads
s1_depth = 1, # stage 1 - depth
s1_mlp_mult = 4, # stage 1 - feedforward expansion factor
s2_emb_dim = 192, # stage 2 - (same as above)
s2_emb_kernel = 3,
s2_emb_stride = 2,
s2_proj_kernel = 3,
s2_kv_proj_stride = 2,
s2_heads = 3,
s2_depth = 2,
s2_mlp_mult = 4,
s3_emb_dim = 384, # stage 3 - (same as above)
s3_emb_kernel = 3,
s3_emb_stride = 2,
s3_proj_kernel = 3,
s3_kv_proj_stride = 2,
s3_heads = 4,
s3_depth = 10,
s3_mlp_mult = 4,
dropout = 0.
)

img = torch.randn(1, 3, 224, 224)

pred = v(img) # (1, 1000)

## Twins SVT


![在这里插入图片描述](https://img-blog.csdnimg.cn/8b3adc6d376947e2b3858f0e45fde6ec.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_11,color_FFFFFF,t_70,g_se,x_16)  
 该[文](https://bbs.csdn.net/topics/618540462)提出了混合本地和全球的关注，与位置编码发生器（中提出沿[CPVT](https://bbs.csdn.net/topics/618540462)）和全球平均水平池，以达到相同的结果[斯文](https://bbs.csdn.net/topics/618540462)，没有转移的窗户，CLS令牌，也不是位置的嵌入的额外的复杂性。

import torch
from vit_pytorch.twins_svt import TwinsSVT

model = TwinsSVT(
num_classes = 1000, # number of output classes
s1_emb_dim = 64, # stage 1 - patch embedding projected dimension
s1_patch_size = 4, # stage 1 - patch size for patch embedding
s1_local_patch_size = 7, # stage 1 - patch size for local attention
s1_global_k = 7, # stage 1 - global attention key / value reduction factor, defaults to 7 as specified in paper
s1_depth = 1, # stage 1 - number of transformer blocks (local attn -> ff -> global attn -> ff)
s2_emb_dim = 128, # stage 2 (same as above)
s2_patch_size = 2,
s2_local_patch_size = 7,
s2_global_k = 7,
s2_depth = 1,
s3_emb_dim = 256, # stage 3 (same as above)
s3_patch_size = 2,
s3_local_patch_size = 7,
s3_global_k = 7,
s3_depth = 5,
s4_emb_dim = 512, # stage 4 (same as above)
s4_patch_size = 2,
s4_local_patch_size = 7,
s4_global_k = 7,
s4_depth = 4,
peg_kernel_size = 3, # positional encoding generator kernel size
dropout = 0. # dropout
)

img = torch.randn(1, 3, 224, 224)

pred = model(img) # (1, 1000)

## RegionViT


![在这里插入图片描述](https://img-blog.csdnimg.cn/75fe7fd9b7ba4583868c919ebd290b71.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_18,color_FFFFFF,t_70,g_se,x_16)


![在这里插入图片描述](https://img-blog.csdnimg.cn/e5b5a906d2ff452892f61bdd0056d6eb.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_14,color_FFFFFF,t_70,g_se,x_16)  
 本[文](https://bbs.csdn.net/topics/618540462)提出将特征图划分为局部区域，从而使局部标记相互协调。每个本地区域都有自己的区域令牌，然后处理其所有本地令牌以及其他区域令牌。


您可以按如下方式使用它：

import torch
from vit_pytorch.regionvit import RegionViT

model = RegionViT(
dim = (64, 128, 256, 512), # tuple of size 4, indicating dimension at each stage
depth = (2, 2, 8, 2), # depth of the region to local transformer at each stage
window_size = 7, # window size, which should be either 7 or 14
num_classes = 1000, # number of output classes
tokenize_local_3_conv = False, # whether to use a 3 layer convolution to encode the local tokens from the image. the paper uses this for the smaller models, but uses only 1 conv (set to False) for the larger models
use_peg = False, # whether to use positional generating module. they used this for object detection for a boost in performance
)

img = torch.randn(1, 3, 224, 224)

pred = model(img) # (1, 1000)

## CrossFormer


![在这里插入图片描述](https://img-blog.csdnimg.cn/1416ccf0c2274ec4a6027837111b6a3f.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_20,color_FFFFFF,t_70,g_se,x_16)  
 ![在这里插入图片描述](https://img-blog.csdnimg.cn/f2554f918b464aa89348ea0b557ff706.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_20,color_FFFFFF,t_70,g_se,x_16)  
 这[纸](https://bbs.csdn.net/topics/618540462)PVT和欧亚交替使用本地和全球的关注甘拜下风。全局注意力是跨窗口维度完成的，以降低复杂性，就像用于轴向注意力的方案一样。


他们还有跨尺度嵌入层，他们证明这是一个可以改进所有视觉转换器的通用层。还制定了动态相对位置偏差，以允许网络推广到更高分辨率的图像。

import torch
from vit_pytorch.crossformer import CrossFormer

model = CrossFormer(
num_classes = 1000, # number of output classes
dim = (64, 128, 256, 512), # dimension at each stage
depth = (2, 2, 8, 2), # depth of transformer at each stage
global_window_size = (8, 4, 2, 1), # global window sizes at each stage
local_window_size = 7, # local window size (can be customized for each stage, but in paper, held constant at 7 for all stages)
)

img = torch.randn(1, 3, 224, 224)

pred = model(img) # (1, 1000)

## NesT


![在这里插入图片描述](https://img-blog.csdnimg.cn/633f2eaf28c2425aaccfeeb3bbee79e2.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_16,color_FFFFFF,t_70,g_se,x_16)


这[纸](https://bbs.csdn.net/topics/618540462)决定来处理分层级的图像，注意力只在局部块，其中聚集因为它移动了层次结构的令牌。聚合是在图像平面中完成的，并包含一个卷积和后续的 maxpool，以允许它跨边界传递信息。


您可以使用以下代码（例如 NesT-T）

import torch
from vit_pytorch.nest import NesT

nest = NesT(
image_size = 224,
patch_size = 4,
dim = 96,
heads = 3,
num_hierarchies = 3, # number of hierarchies
block_repeats = (8, 4, 1), # the number of transformer blocks at each heirarchy, starting from the bottom
num_classes = 1000
)

img = torch.randn(1, 3, 224, 224)

pred = nest(img) # (1, 1000)

## MobileViT


![在这里插入图片描述](https://img-blog.csdnimg.cn/1b846690d42e48edbd30a19aa1377013.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_20,color_FFFFFF,t_70,g_se,x_16)


这个纸介绍MobileViT，重量轻的和通用的视觉变压器用于移动设备。MobileViT 为使用转换器对信息进行全局处理提供了不同的视角。


您可以使用以下代码（例如 mobilevit\_xs）

import torch
from vit_pytorch.mobile_vit import MobileViT

mbvit_xs = MobileViT(
image_size = (256, 256),
dims = [96, 120, 144],
channels = [16, 32, 48, 48, 64, 64, 80, 80, 96, 96, 384],
num_classes = 1000
)

img = torch.randn(1, 3, 256, 256)

pred = mbvit_xs(img) # (1, 1000)

## 简单的蒙版图像建模


![在这里插入图片描述](https://img-blog.csdnimg.cn/4e34e571c46f4009a6a6e1d168a7084b.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA6IuP5bee56iL5bqP5aSn55m9,size_20,color_FFFFFF,t_70,g_se,x_16)  
 这个纸提出了一种简单的掩蔽图像的建模（SimMIM）方案，仅使用一个线性投影断掩蔽令牌为像素空间后跟一个L1损失与掩蔽贴片的像素值。结果与其他更复杂的方法相比具有竞争力。


您可以按如下方式使用它：

import torch
from vit_pytorch import ViT
from vit_pytorch.simmim import SimMIM

v = ViT(
image_size = 256,
patch_size = 32,
num_classes = 1000,
dim = 1024,
depth = 6,
heads = 8,
mlp_dim = 2048
)

mim = SimMIM(
encoder = v,
masking_ratio = 0.5 # they found 50% to yield the best results
)

images = torch.randn(8, 3, 256, 256)

loss = mim(images)
loss.backward()

that’s all!

do the above in a for loop many times with a lot of images and your vision transformer will learn

torch.save(v.state_dict(), ‘./trained-vit.pt’)

## 屏蔽自编码器




### 给大家的福利


**零基础入门**


对于从来没有接触过网络安全的同学，我们帮你准备了详细的学习成长路线图。可以说是最科学最系统的学习路线，大家跟着这个大的方向学习准没问题。


![](https://img-blog.csdnimg.cn/img_convert/95608e9062782d28f4f04f821405d99a.png)


同时每个成长路线对应的板块都有配套的视频提供：


![在这里插入图片描述](https://img-blog.csdnimg.cn/direct/a91b9e8100834e9291cfcf1695d8cd42.png#pic_center)


因篇幅有限，仅展示部分资料

**网上学习资料一大堆，但如果学到的知识不成体系，遇到问题时只是浅尝辄止，不再深入研究，那么很难做到真正的技术提升。**

**[需要这份系统化资料的朋友，可以点击这里获取](https://bbs.csdn.net/topics/618540462)**

**一个人可以走的很快，但一群人才能走的更远！不论你是正从事IT行业的老鸟或是对IT行业感兴趣的新人，都欢迎加入我们的的圈子（技术交流、学习资源、职场吐槽、大厂内推、面试辅导），让我们一起学习成长！**