来碗拿铁️-CSDN博客

原创 CUDA initialization: Unexpected error from cudaGetDeviceCount(). Did you run some cuda functions bef

安装fabricmanager。

2024-06-12 20:48:40 194

原创 Transformers在自动驾驶领域的应用（High-level like in scene）

Transformers uses global attention mechanism to capture long-range interaction information. Two core modules in Transformers are Self-Attention and Cross-Attention, which defined asAttn(Q,K,V)=softmax(QK⊤dk)VSelfAttn(X)=Attn(WQX,WKX,WVX)CrossAttn(X,Y)=Attn

2023-04-26 10:56:00 1364

原创 Execution failed for task ... The tar might be corrupted or it is compressed in an unexpected way错误

最近在跑Android的程序，出现以下有关gradle的错误：The tar might be corrupted or it is compressed in an unexpected way.By default the tar tree tries to guess the compression based on the file extension.If you need to specify the compression explicitly please refer to the

2021-04-15 16:15:08 420

原创基于FCOS的航拍巡检图像绝缘子识别

本学期选修了一门CVPR课程，最后的大作业要求结合自身的专业进行一次物体识别的作业。我的专业是电气工程，因此选择了航拍图像中的绝缘子故障识别问题。基于FCOS的航拍巡检图像绝缘子识别引言应用背景高压线路输电距离长，沿线的自然环境多样，地理环境复杂，输电线路易在恶劣环境中受到损害。如果监控不到位，发生故障无法及时处理，必将导致重大事故发生，严重威胁和影响电力系统的正常稳定运行，甚至造成系统大规模瘫痪的严重后果。对输电线路进行定期巡视检查是保证电力系统稳定运行的重要前提。绝缘子类型多、数量大、分布广，是

2021-03-21 17:39:16 1141 4

原创 anaconda安装opencv3.4.2

conda menpo和pip的方法很容易网络中断，推荐使用清华的源，秒安装。conda install -c https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main opencv

2021-02-18 15:50:15 804

原创多模态情绪识别调研

多模态情绪识别调研问题定义情绪 Emotion多模态 Multi-modal融合算法决策融合特征融合多模态HMM受限/深度玻尔兹曼机（Restricted/Deep Boltzmann Machine）不同设计的网络结构Dense Multimodal FusionMultimodal Transfer ModuleGATED MULTIMODAL UNITSEmotiConLate Fusion of Multimodal InformationCross-Modal Self-Attention Net

2021-02-01 14:08:47 2755

原创 Inverse Geometric Transform； Gaussian and Laplacian Pyramid； Canny Filter； Harris Detection从零实现

Inverse Geometric Transform; Gaussian and Laplacian Pyramid; Canny Filter; Harris Detection从零实现Geometric TransformInterpolate functionTransform functionplot functionTransformsTranslationResizeRotationRigidSimilarityAffineProjectionImage PyramidConvolutiond

2020-12-09 11:48:21 810

原创 Convex Optimization 读书笔记 (10)

Chapter11: Interior-point methods11.1 Inequality constrained minimization problemsIn this chapter we discuss interior-point methods for solving convex optimization problems that include inequality constraints,minimize f0(x)subject

2020-11-18 18:03:28 291

原创 Convex Optimization 读书笔记 (9)

Chapter10: Equality constrained minimization10.1 Equality constrained minimization problemsConsider a optimization problem:minimize f(x)subject to Ax=b\begin{aligned}{\rm minimize} \ \ \ \ & f(x)

2020-11-18 11:23:53 380

原创 Convex Optimization 读书笔记 (8)

Chapter9: Unconstrained minimization9.1 Unconstrained minimization problemsThe unconstrained optimization problem isminimize f(x)\begin{aligned}{\rm minimize} \ \ \ \ & f(x) \\\end{aligned}minimize

2020-11-12 17:50:20 495

原创 Convex Optimization 读书笔记 (7)

Chapter8: Geometric problems8.1 Projection on a setThe distance of a point x0∈Rnx_0 ∈ \mathbf{R}^nx0∈Rn to a closed set C⊆RnC⊆\mathbf{R}^nC⊆Rn, in the norm ∥⋅∥∥·∥∥⋅∥, is defined asdist(x0,C)=inf⁡{∣∣x0−x∣∣∣x∈C}{\rm \bold{dist}}(x_0,C)=\inf\{ ||x_0-x||

2020-11-11 14:33:37 562

原创 Convex Optimization 读书笔记 (6)

Chapter7: Statistical estimation7.1 Parametric distribution estimation7.1.1 Maximum likelihood estimationDefine log-likelihood function, and denoted l:l(x)=log⁡px(y)l(x)=\log p_x(y)l(x)=logpx(y)A widely used method, called maximum likelihood (ML) es

2020-11-08 20:17:34 374

原创 Convex Optimization 读书笔记 (5)

Chapter 6: Approximation and fitting6.1 Norm approximation6.1.1 Basic norm approximation problemThe simplest norm approximation problem is an unconstrained problem of the formminimize ∣∣Ax−b∣∣{\rm minimize} \ \ \ \ ||Ax-b||min

2020-11-05 21:25:24 458

原创 Convex Optimization 读书笔记 (4)

Chapter5: Duality5.1 The Lagrange dual function5.1.1 The LagrangianConsider an optimization problem in the standard formminimize f0(x)subject to fi(x)≤0,i=1,...mhi(x)=0,i=1,...p\begin{aligned}{\rm mi

2020-11-02 11:34:24 657

原创 Convex Optimization 读书笔记 (3)

4 Convex optimization problems4.1 Optimization problems4.1.1 Basic terminologyWe use the notationminimize f0(x)subject to fi(x)≤0,i=1,...mhi(x)=0,i=1,...p\begin{aligned}{\rm minimize} \ \ \ \ & f

2020-10-29 14:56:35 556

原创 Convex Optimization 读书笔记 (2)

Chapter3: Convex Functions3.1 Basic properties and examples3.1.1 DefinitionA function f:Rn→Rf:\mathbf{R}^n\rightarrow\mathbf{R}f:Rn→R is a convex function if dom f\mathbf{dom}\space fdom f is convex set and for x,y∈dom f,θ∈[0,1]x,y\in \ma

2020-10-27 15:52:12 793

原创 Convex Optimization 读书笔记 (1)

Chapter2 Convex Set2.1 Affine, Convex Set2.1.1 Line, Line SegmentSuppose x1,x2x_1,x_2x1,x2 are two points in Rn\mathbf{R}^nRn,y=θx1+(1−θ)x2,θ∈Ry=\theta x_1+(1-\theta)x_2,\theta\in\mathbf{R}y=θx1+(1−θ)x2,θ∈Ris a line. If θ∈[0,1]\theta\in[0,1]θ∈[0,1

2020-10-24 14:16:19 564

原创 Designing Network Design Spaces论文导读

Designing Network Design Spaces论文导读AnyNetXAnyNetX~B~，AnyNetX~C~AnyNetX~D~，AnyNetX~E~RegNetXGeneralizationRegnetX AnalysisTrendComplexityRegNetX CKaiming He的力作，内容非常之丰富，且具有很大的指导意义。大多数的NAS都是只找到一个最优的结构，本文则通过不断的缩小假设空间，以找到一簇比较优秀的网络：通过采样进行超参数的选择，根据误差的经验分布函数（em

2020-09-30 17:18:50 333

原创 MINE复现[DV, fGAN, infoNCE]

Environmentpytorch 1.5.1torchvision 0.6.1numpy 1.18.5matplotlib 3.3.0Main usage of class# -*- coding: UTF-8 -*-"""=================================================@IDE :Pycharm

2020-09-25 20:11:33 1194 1

原创互信息的原理、计算和应用

互信息的原理、计算和应用Background熵 Entropy交叉熵 Cross Entropy条件熵 Conditional EntropyKL-散度 KL-divergence定义计算方法变分法 Variational approachMutual Information Neural Estimation, MINEDEEP INFOMAX应用迁移学习迁移强化学习自监督学习ReferencesAppendixA.B.Background熵 Entropy在信息论中，熵是给定概率下的最佳编码方式[

2020-09-14 18:00:01 5617

原创 [pytorch] WACA论文，photo2sketch的复现

WACA论文，photo2sketch的复现：让你的照片自动变成简笔画论文idea数据集网络结构损失函数代码复现networksUnet generatorunet modulebuild unetpatchGAN discriminatorbuild netnetwork utilitiesLoss function结果分析生成对抗网络（GAN）虽然近年来热度有所下降，不过它展现了神经网络所拥有的巨大创造潜力。这篇被WACA收录的论文Photo-Sketching:Inferring Contour

2020-07-28 14:20:32 866

qq_39337332的博客

原创 CUDA initialization: Unexpected error from cudaGetDeviceCount(). Did you run some cuda functions bef

原创 Transformers在自动驾驶领域的应用（High-level like in scene）

原创 Execution failed for task ... The tar might be corrupted or it is compressed in an unexpected way错误

原创基于FCOS的航拍巡检图像绝缘子识别

原创 anaconda安装opencv3.4.2

原创多模态情绪识别调研

原创 Inverse Geometric Transform； Gaussian and Laplacian Pyramid； Canny Filter； Harris Detection从零实现

原创 Convex Optimization 读书笔记 (10)

原创 Convex Optimization 读书笔记 (9)

原创 Convex Optimization 读书笔记 (8)

原创 Convex Optimization 读书笔记 (7)

原创 Convex Optimization 读书笔记 (6)

原创 Convex Optimization 读书笔记 (5)

原创 Convex Optimization 读书笔记 (4)

原创 Convex Optimization 读书笔记 (3)

原创 Convex Optimization 读书笔记 (2)

原创 Convex Optimization 读书笔记 (1)

原创 Designing Network Design Spaces论文导读

原创 MINE复现[DV, fGAN, infoNCE]

原创互信息的原理、计算和应用

原创 [pytorch] WACA论文，photo2sketch的复现

原创词向量--原理、word2vec、GloVe及其实现

原创 unet道路提取

原创 [fasi.ai] unet实现CamVid数据集预测

原创 module 'tensorflow.python._pywrap_tensorflow_internal' has no attribute 'version'

原创算法导论的C实现——优先队列

原创算法导论的C实现——堆排序

原创算法导论的C实现——画出d叉树

原创算法导论的C实现——画出二叉树

原创算法导论的C实现——分治法求最大子数组

原创微信小程序：chart.js的使用

原创微信小程序+Servlet后台开发（三）

原创阿里云短信服务

原创微信小程序+Servlet后台开发（二）

原创微信小程序+Servlet后台开发（一）

空空如也

空空如也