RtZero-CSDN博客

原创 Tiny Tiny RSS ttrss 部署

#!/usr/bin/env bash# downloadgit clone https://git.tt-rss.org/fox/ttrss-docker-compose.git ttrss-dockercd ttrss-dockergit checkout static-dockerhub# installsudo apt install postgresql-client-commonsudo apt install postgresql-clientsudo apt insta.

2020-07-01 14:16:01 3707

原创 caffe之classification.cpp & cmake

classification.cpp#include #ifdef USE_OPENCV#include #include #include #endif // USE_OPENCV#include #include #include #include #include #include #ifdef USE_OPENCVusing namespace caf

2016-12-21 15:47:23 719

原创 caffe之小小的CNN网络跑起来

把前面做好的mdb跑起来,网络嘛就是简单的lenet_lr网络.呼呼网络设置:name: "LeNet"layer { name: "mnist" type: "Data" top: "data" top: "label" include { phase: TRAIN } transform_param { scale: 0.0039062

2016-12-18 15:14:42 621

原创 caffe之操蛋的create_imagenet

QAQ,鼓捣了三四天,反正我蠢我bbtrain0.JPEG 0train1.JPEG 0train2.JPEG 0train3.JPEG 0train4.JPEG 0train5.JPEG 0train6.JPEG 0train7.JPEG 0......train.txt应该这样写cdcd 文档/sh_learnbash val.shbash trai

2016-10-31 10:11:33 1254

原创 caffe之blob_demo.cpp正常运行大法

声明啊:我电脑名砸叫erkp,caffe放在caffe-master里.各位自己改咯写个blob_demo.cpp如下:#include #include #include using namespace std;using namespace caffe;int main(void){ Blob a; cout<<"size : "<<a.shape_str

2016-10-25 16:35:08 2834 3

转载 matconvnet源代码backPropLim

case 'conv' [res(i).dzdx, dzdw{1}, dzdw{2}] = ... vl_nnconv(res(i).x, l.weights{1}, l.weights{2}, res(i+1).dzdx, ... 'pad', l.pad, ... 'stride', l.stride, ...

2016-10-16 15:58:47 711

转载 matconvnet源代码:skipForward

case 'conv' res(i+1).x = vl_nnconv(res(i).x, l.weights{1}, l.weights{2}, ... 'pad', l.pad, ... 'stride', l.stride, ... 'dilate', l.dilate, ... l.opts{:}, ...

2016-10-16 15:57:33 993

原创 matconvnet分步注释(四)网络设置

网络设置:这东西好像很简单,只需要注意网络结构和网络的学习率什么的function net =cnn_plate_init()rng('default');rng(0) ;%随机数设置f=1/100 ;net.layers = {};net.layers{end+1} = struct('type', 'conv', ...

2016-10-16 15:45:59 1321

原创 matconvnet分步注释(三)运行matconvnet

运行matconvnet勒function [net, info] = cnn_plate()%初始化神经网络run('C:\Users\TAT\Desktop\matconvnet-1.0-beta22\matlab\vl_setupnn.m') ;%datadir='E:\MachineLearning\caffe\caffe-windows-master\platerecognit

2016-10-16 15:44:27 826

原创 matconvnet分步注释(二)主程序

主程序：预测clear;close all;clc;%训练网络[net_bn, info_bn] = cnn_plate();%设置验证图像img=imread('C:\Users\TAT\Desktop\matconvnet-1.0-beta22\mycode\plate\1.jpg');img=imresize(img,net_bn.meta.inputSize(1:2));

2016-10-16 15:40:51 659

原创 matconvnet分步注释(一)数据预处理

数据预处理clear;close all;clc;datadir='C:\Users\TAT\Desktop\matconvnet-1.0-beta22\mycode\helloworld\date';inputSize =[20,20,1];%输入设置subdir=dir(datadir);%文件夹结构imdb.images.data=[];%图像数据imdb.images

2016-10-16 14:45:42 1435 1

原创 cuda,day-13,tensorcalc张量计算

//--------------------------//Tensor calculate//--------------------------#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include #include

2016-05-18 12:00:48 557

原创 cuda-矩阵乘法

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include "iostream"using namespace std;#define Mx 64#define My 32#define Nx 32#define Ny 48#define Px 64#defi

2016-04-24 08:59:52 329

原创 Cuda,多层感知机-cuda框架

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;void f(float *dst, float src[], int h, int w);void df(float *dst, float src[], int h

2016-04-12 11:38:34 362

原创 day-12,多层感知机-矩阵计算 <C版>

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;void f(float *dst, float src[], int h, int w);void df(float *dst, float src[], int h

2016-04-11 09:42:33 439

原创 CUDA,day-9,大规模矩阵乘法

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;#define u32 unsigned int #define ARRAY_SIZE_X 8#define ARRAY_SIZE_Y 8u32

2016-04-04 16:30:37 324

原创 cuda,day-9,矩阵乘法

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;#define u32 unsigned int #define ARRAY_SIZE_X 2#define ARRAY_SIZE_Y 3u32

2016-04-04 16:17:42 213

原创 cuda,常量内存使用2

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include "assert.h"#include "conio.h"using namespace std;#define u16 unsigned short int#define u32 unsigned int

2016-03-24 16:53:35 13379

原创 cuda ,常量内存使用

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include "assert.h"#include "conio.h"using namespace std;#define CUDA_CALL(x){const cudaError_t a=(x);if(a!=cuda

2016-03-24 16:40:37 8748

原创 cuda,内存变量定义

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func1(int* data1, int* data2, int* data3);__dev

2016-03-24 16:13:57 695

原创 cuda 常量内存,头文件不知道有什么gui问题

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include "device_atomic_functions.h"#include #include #include "crt\func_macro.h"#include "asse

2016-03-24 15:43:32 1113

原创 CUDA 常数与指针的传递

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func3(int * const data, int num_elements);#define

2016-03-22 18:19:52 1808

原创 CUDA 数学函数

C.1 数学函数C.1.1 atomicAdd()int atomicAdd(int* address, int val);unsigned int atomicAdd(unsigned int* address, unsigned int val);unsigned long long int atomicAdd(un

2016-03-20 18:10:31 2958

原创 CUDA 内置变量

内置变量1.gridDim 此变量的类型为 dim3，包含网格的维度。2. blockIdx此变量的类型为 uint3，包含网格内的块索引。3.blockDim此变量的类型为 dim3，包含块的维度。4. threadIdx此变量的类型为 uint3，包含块内的线程索引。5.warpSize此变量的类型为 int，包含以线程为单位的 warp 块

2016-03-20 18:06:13 888

转载 day-3,cdpSimplePrint

2016-03-17 17:11:20 343

原创 CUDA 函数

CUDA里 A << B 是什么意思啊?相当于 A * 2^B那A >>B 呢?A / 2^B

2016-03-17 15:36:37 350

原创 CUDA,标准文件

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;#define u32 unsigned intint main(){return 0;}

2016-03-17 12:03:50 277

原创 CUDA,day-2,统计直方图

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include #include using namespace std;__global__ void func1(const unsig

2016-03-17 12:01:44 343

原创 CUDA,day-2,寄存器用法

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;__device__ static unsigned int d_tmp[NUM_ELEM];__global__ void test_gpu_gmem(unsigne

2016-03-17 09:34:07 355

原创 CUDA,day-2,合并列表

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include using namespace std;#define u32 unsigned int__global__ void gpu_sort_

2016-03-17 09:32:53 347

原创 CUDA,day-2,共享内存-排序算法

#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;#define u32 unsigned int__device__ void func1(u32 * const sort_tmp,const u32

2016-03-17 09:32:15 345

原创 CUDA,day-2,二维数组操作

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func2(int * block_x,int * block_y,int * th

2016-03-17 09:31:53 1276

原创 CUDA,day-1,一维数组

#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func1(int* block,int* thread,int* warp,int*

2016-03-17 09:30:51 944