codeQin-优快云博客

原创 tesseract OCR 和Python pytesseract下载与安装

(一) tesseract OCR1.下载地址：https://digi.bib.uni-mannheim.de/tesseract/2.环境变量PATH配置“电脑-属性-高级系统设置-环境变量-系统变量-PATH”①软件安装路径：D:\vscode\tesseract-OCR②PATH新增3.验证安装是否成功(二)Python pytesseract1.下载：执行命令pip install pytesseract或者至https://github.co...

2021-08-29 22:56:28 1706 1

原创差分进化算法DE优化BPNN

(一)DE的基本原理主要操作：变异 - 交叉 - 选择 (二)DE优化BPNN%载入数据data = csvread('D:\matlab2016a\OA\sample.csv');%划分数据集rv = rand(1,size(data,1));[value,index] = sort(rv);x_train = data(index(1:150),1:2);y_tra...

2018-12-03 13:00:11 2182 5

原创 NOGM(1,1)

NOGM(1,1)基本思想: PSO优化初始条件, 即： data = csvread('D:\matlab2016a\demo\data\air.csv');num = size(data,1);predictnum = 6;train = data(1:num-predictnum,:);test = data((num-predictnum+1):end,:);...

2018-11-26 11:04:28 1079 4

原创改进版GM(1,1)

优化GM(1,1)主要的改进手段：优化微分方程的初始条件 data = csvread('D:\matlab2016a\demo\data\air.csv');num = size(data,1);predictnum = 6;train = data(1:num-predictnum,:);test = data((num-predictnum+1):end,:);...

2018-11-25 11:26:28 2165 3

原创基本GM(1,1)

data = csvread('D:\matlab2016a\demo\data\air.csv');%导入数据num = size(data,1);%样本数predictnum = 6;%设置预测记录数train = data(1:num-predictnum,:);%训练样本数test = data((num-predictnum+1):end,:);%测试样本数%做累加for ...

2018-11-25 11:15:57 790

原创回声状态网络ESN

dataset = csvread('D:\matlab2016a\data.csv');x = dataset(:,1:3);y = dataset(:,4);x_train = x(1:145,:);y_train = y(1:145);x_test = x(146:end,:);y_test = y(146:end);[samples,inputnum] = size(x_...

2018-10-16 14:04:29 1288 2

原创 Keras搭建神经网络LSTM(回归)

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import preprocessingfrom keras.models import Sequentialfrom keras.layers import Densefrom keras.layers import LST...

2018-10-06 22:21:22 5389

原创 Keras参数设置说明

(一)keras参数:loss定义:loss是目标函数(损失函数)可用目标函数: -回归问题(loss='mse' | 'mae' | 'mape') -分类问题(loss='binary_crossentropy' | 'categorical_crossentropy') tip:1.binary_crossentropy作为目标函数用于二分类 ...

2018-10-06 15:16:25 11306 1

原创 Keras搭建神经网络BPNN(回归问题)

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import datasets,preprocessingfrom sklearn.model_selection import train_test_splitfrom keras.models import Sequenti...

2018-10-06 13:35:43 5033 2

原创 Keras搭建神经网络BPNN(分类问题)

(一)BPNN(二分类)使用软件:AnacondaTip: 输出层使用Sigmoid函数import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import datasets,preprocessingfrom sklearn.model_selection import ...

2018-10-06 11:14:04 2566 2

原创 Python(pybrain模块)搭建神经网络BPNN

(一)Python环境下安装pybrain(开源神经网络工具包)pybrain模块(URL):https://github.com/pybrain/pybrainpybrain官方文档:http://pybrain.org/docs/git下载pybrain:git clone https://github.com/pybrain/pybrainTip:1.在安装pybrain前,...

2018-10-05 14:51:40 5365

原创 Python(Classification)

Keras搭建NN(classification)import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import datasetsfrom sklearn import preprocessingfrom keras.models import Sequential,Mod...

2018-10-04 20:01:14 1999

原创 Python(流水线)

sklearn-Pipelineimport numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import preprocessingfrom sklearn import datasetsfrom sklearn.model_selection import train_test_s...

2018-10-04 13:58:53 1687

原创 Python(Cluster)

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import preprocessingfrom sklearn.cluster import KMeansfrom sklearn.cluster import MeanShift,estimate_bandwidthfro...

2018-10-04 11:41:54 3857

原创公共数据网站

世界卫生组织数据:http://www.who.int/gho/database/en联合国数据:http://data.un.org世界银行数据:https://data.worldbank.org/Datahub:https://datahub.io/人口资料局:http://www.prb.org/data加拿大开放数据:http://open.canada.ca香港...

2018-09-28 13:37:25 567

原创 Latex(Elsevier)模板

Elsevier(Letex)模板下载及使用(一)模板下载URL:https://www.elsevier.com/authors/author-schemas/latex-instructions1.下载压缩文件:elsarticle-ecrc.zip2.解压缩后包含的文件有:elsdoc.pdf(简易使用手册) ...

2018-09-27 20:40:26 29791 1

原创 LSSVM分类和回归

LSSVM-回归gamma = 100;sig2 = 13;type = 'function estimation';%训练模型[alpha,b] = trainlssvm({x_train,y_train,type,gamma,sig2,'RBF_kernel'});%模型预测predict_train = simlssvm({x_train,y_train,type,gam...

2018-09-04 19:45:10 12146 1

原创 PSO优化ELM

data = csvread('D:\article\demo\seasonal\stl.csv');sample = size(data,1);timespan = 6;imfindex = 3;for i = 1:(sample-timespan) x(i,:) = data(i:i+timespan-1,imfindex)';endy = data(timespan+1:...

2018-09-04 19:40:19 10161 8

原创极限学习机ELM

BP网络的特点(优点)非线性映射能力；自适应性；(缺点)梯度下降法需要多次迭代，从而修正权值和阈值，所以训练速度慢；容易陷入局部最优，无法达到全局最小；学习率选择敏感(太小收敛速度慢，太大训练过程不稳定)；极限学习机(ELM)特点在训练过程中无需调整权值和阈值，只需要调整隐含层神经元个数，可获得唯一最优解；学习速度快，泛化性能好；data = csvread...

2018-09-04 19:20:31 1450

原创图形数字化工具 - getData

2018-08-31 11:12:02 12035

原创 PSO-LSSVM

粒子群算法优化LSSVMload regressionData.mat;output = output';%划分数据集[value,index] = sort(rand(1,2000));x_train = input(index(1:1800),:);x_test = input(index(1801:2000),:);y_train = output(index(1:1800...

2018-08-30 16:11:39 8018 7

原创最小二乘支持向量机LSSVM

LSSVM - Regression(注意：将LSSVM工具箱添加到toolbox会造成BPNN网络无法建立])解决办法：将LSSVM路径删去load regressionData.mat;output = output';%划分数据集[value,index] = sort(rand(1,2000));x_train = input(index(1:1800),:);x_...

2018-08-30 14:12:05 3968 4

原创回声状态网络ESN(原理)

回声状态网络ESN(原理)结构特点网络有3层：输入层 - 隐含层 - 输出层 ...

2018-08-26 17:04:11 8983

原创长短期记忆网络LSTM(Python)

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom keras.models import Sequentialfrom keras.layers import Densefrom keras.layers import LSTMfrom sklearn import preprocessin...

2018-08-25 11:25:55 2313 1

原创遗传算法GA优化BPNN

x_train = x(1:100,:);y_train = y(1:100,:);x_test = x(101:end,:);y_test = y(101:end,:);%数据归一化[train_input,inputps] = mapminmax(x_train');[train_output,outputps] = mapminmax(y_train');%节点设置inp...

2018-08-21 23:02:52 2919 6

原创集成经验模态分解EEMD

EEMD使用手册1.将eemd工具箱放置在toolbox文件夹下2.在matlab主界面，点击"设置路径" - "添加文件夹" - 将eemd路径加入 - 保存函数调用方法imfs = eemd(ts,Nstd,NE);注: ts(要分解的时间序列) Nstd(高斯白噪声标准差,0.01-0.4) NE(设置添加噪声的次数,取50或100次...

2018-08-21 17:18:51 17906 13

原创长短期记忆网络LSTM(matlab)

load datasetstop = 0; %控制预测的序列索引(0-倒数第1 1-倒数第2 2-倒数第3,..以此类推)series = dataset(1:end-stop,1); %导入时间序列数据timespan = 5; %时间跨度(即历史数据条数)%构造...

2018-08-17 16:11:50 7130 3

原创小波神经网络(时间序列预测)

Matlab(小波神经网络时间序列预测)时间序列数据66 64 60 58 17 49 34 17 8 53 15 15 45 57 33 42 8 52 24 29 -13 37 37 4 32 38 68 77 95 119 161 184 276 247 251 193 226 213 195 182 166 188 1...

2018-08-11 16:16:15 18319 16

原创概率神经网络PNN

%加载数据data = csvread('D:\python\anaconda\kaggle\wine.csv',1);x = data(:,1:13);y = data(:,14);y = y+1; %标签值必须为正(含有label=0需要进行此步操作)%划分数据集[m,n] = size(x);index = randsample(m,m);x_train = ...

2018-08-10 23:11:27 3120

原创 R-时间序列分析

ARIMA模型读取数据data<-read.table(file='D:/python/anaconda/ARIMA/airPassenger.csv',header=T,sep=',',row.names='time')数据格式如下: num1949/1/1 112...

2018-08-09 22:50:44 1264

原创粒子群优化算法PSO优化BPMN

load data input output%划分数据集random_num = rand(1,2000);[value,index] = sort(random_num);x_train = input(index(1:1800),:)';y_train = output(index(1:1800));x_test = input(index(1801:2000),:)';y_t...

2018-08-08 14:39:25 2279 4

原创果蝇优化算法FOA优化BPNN

load data input output%划分训练集random_num = rand(1,2000);[value,index] = sort(random_num);x_train = input(index(1:1800),:)';y_train = output(index(1:1800));x_test = input(index(1801:2000),:)';y_t...

2018-08-08 14:27:33 3082 8

原创 ML-ARIMA

import pandas as pdimport numpy as npimport matplotlib.pyplot as plt#判定时序数据稳定性from statsmodels.tsa.stattools import adfuller#将时序数据中的趋势和周期性分离出来from statsmodels.tsa.seasonal import seasonal_decomp...

2018-08-04 21:43:47 273

原创经验模态分解-EMD

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom pyhht.emd import EMDfrom pyhht.visualization import plot_imfsdataset = pd.read_csv('D:/python/anaconda/ARIMA/airPassenger...

2018-08-04 21:07:06 4330

原创 LIBSVM-(基本使用)

(一)下载安装LIBSVM模块(python)LIBSVM安装包下载地址: https://www.csie.ntu.edu.tw/~cjlin/libsvm/oldfiles/安装流程:1.将下载好的压缩包解压至Python目录(具体:python\Lib\site-packages),目的是方便使用时导入模块, 将解压后的文件夹重命名为libsvm2.在libsv...

2018-07-21 18:30:28 608

原创 ML-LSSVM(classifier:FOA优化参数)

import timeimport numpy as npimport matplotlib.pyplot as pltfrom sklearn import datasetsfrom sklearn import preprocessingfrom sklearn.model_selection import train_test_splitfrom sklearn.feature_...

2018-07-21 10:53:11 1327 1

原创 ML-LSSVM(regression:FOA优化参数)

import numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import datasetsfrom sklearn import preprocessingfrom sklearn.model_selection import train_test_splitframe = pd.r...

2018-07-21 10:44:10 1136

原创粒子群优化算法-PSO

import numpy as npimport matplotlib.pyplot as pltN = 10dim = 2iter_times = 100def function(vec): return vec[0]**2+vec[1]**2-5def initSwarm(N,dim): x = np.random.random((N,dim)) v ...

2018-07-21 10:34:40 1302

原创果蝇优化算法-FOA

import numpy as npimport matplotlib.pyplot as pltsample = 50iter_times = 100def initParams(bestX,bestY,sample): position = [] fitness = [] for num in range(sample): x,y = be...

2018-07-21 10:32:49 3454

原创 ML-KNeighborsRegressor

# -*- coding: utf-8 -*-"""Created on Wed Jul 18 15:00:32 2018@author: qinlan主要思想:选取样本的K个近邻样本，用近邻样本的响应值(y)的均值作为该样本的预测值"""import numpy as npimport pandas as pdimport matplotlib.pyplot as pltf...

2018-07-18 17:00:07 1125

百度音乐下载

空空如也