szfhy的博客

a = np.floor(20 * np.random.random((2,2)))b = np.floor(20 * np.random.random((2,2)))c = np.floor(20 * np.random.random((2,2)))print(a)print(b)print(c)[[ 8. 2.] [16. 15.]][[ 7. 0.] [10. 12.]]...

Ctrl + Shift + - #将一个 cell拆分为两个cellesc dd #删除一个cell, esc 按一下，切换到命令模式Ctrl+enter #执行本cellshift+enter #执行本cell且 向下建立一个新cellesc+m ...

http://www.machinelearningplus.com/plots/top-50-matplotlib-visualizations-the-master-plots-python/

pandas 是一个强大的时间序列数据处理工具包，由两个比较重要的数据结构，series和dataframe，可以简单的将其抽象为一个表格。series可以表达一行数据，理解为一维数组。s = pandas.Series([4,2,5,0,6,3])DataFrame表示二维数组， df = pandas.DataFrame(numpy.random.randn(6,4), columns...

坐标轴刻度import matplotlib.pyplot as pltimport numpy as npx=np.arange(1,11,1)plt.plot(x,x)ax = plt.gca()#ax.locator_params(nbins=5) #xy轴同时调整ax.locator_params('x',nbins=10) #只调整x轴ax.locator_para...

色彩配置import matplotlib.pyplot as pltimport numpy as npy = np.arange(5)plt.plot(y, color='g')plt.plot(y+1, color='0.5')plt.plot(y+2, color='#ff00ff')plt.plot(y+3, color=(0.1,0.2,0.3))plt.s...

热成像图%matplotlib inlinefrom matplotlib import pyplot as pltimport numpy as npdef f(x, y): return (1 - x / 2 + x ** 5 + y ** 3) * np.exp(-x ** 2 - y ** 2)n = 10x = np.linspace(-3, 3, 4 * n...

饼图import numpy as npimport matplotlib.pyplot as pltlabels='A','B','C','D'fracs = [15,30, 45, 10]explode = [0,0.1,0.05,0]#长宽比为1:1plt.axes(aspect=1)#autopct 控制百分比的设置，使用format字符串或者format funct...

单变量直方图import numpy as npimport matplotlib.pyplot as pltmu = 100 # mean of distributionsigma = 20 #standard deviatioin of distributionx = mu + sigma * np.random.randn(200000)#绘制单变量的直方图plt.his...

import numpy as npimport matplotlib.pyplot as pltN = 5y = [20, 10, 30, 25, 15]index = np.arange(N)plt.bar(left=index, height=y,color='red', width=0.5) #left横坐标，height纵坐标, color颜色 widthplt.show(...

import matplotlib.pyplot as pltimport numpy as npx= np.linspace(0,5*np.pi, 1000)y1 = np.sin(x)y2 = np.sin(2*x)#plt.plot(x,y1)#plt.plot(x,y2)plt.fill(x,y1,'b',alpha=0.5)plt.fill(x,y2,'r',a...

share一些python实现的code#!/usr/bin/env python#coding=utf-8import cv2img = cv2.imread("trace_border2.bmp")[img_h, img_w, img_channel] = img.shapetrace = []start_x = 0start_y = 0gray = img[:,...

关于混淆矩阵，之前看别人的一些教程，感觉作者很牛，图形化显示出来也很酷，最近深挖了一下文档，顿时感觉我们很牛人就间隔一张纸。详细的实现：confusion_matrix顺便添加自己的调试的代码，记个笔记。import numpy as npimport matplotlib.pyplot as pltfrom sklearn import svm, datasetsfrom ...

安装pip install ipywidgets配置jupyter nbextension enable --py widgetsnbextension使用from ipywidgets import interactdef f2(x): return xinteract(f2, x='')显示效果

xgboost需要单独安装pip install xgboost #安装xgboost库pip install --upgrade xgboost #更新xgboost库import xgboost as xgbfrom xgboost import XGBRegressor as XGBRfrom sklearn.ensemble import RandomForestRegr...

代码：from sklearn.linear_model import LinearRegression as LRfrom sklearn.model_selection import train_test_splitfrom sklearn.model_selection import cross_val_scorefrom sklearn.datasets import fetc...

代码笔记：from sklearn.datasets import make_blobsfrom sklearn.svm import SVCimport matplotlib.pyplot as pltimport numpy as npX,y = make_blobs(n_samples=50, centers=2, random_state=0, cluster_std=0....

代码笔记：from sklearn.datasets import load_breast_cancerfrom sklearn.svm import SVCfrom sklearn.model_selection import train_test_splitimport matplotlib.pyplot as pltimport numpy as npfrom time im...

代码笔记：from sklearn.linear_model import LogisticRegression as LRfrom sklearn.datasets import load_breast_cancerimport numpy as npimport matplotlib.pyplot as pltfrom sklearn.model_selection import...

代码笔记import matplotlib.pyplot as pltfrom sklearn.datasets import load_irisfrom sklearn.decomposition import PCAiris = load_iris()y = iris.targetX = iris.dataimport pandas as pdpd.DataFrame(X...

from sklearn.datasets import load_breast_cancerfrom sklearn.ensemble import RandomForestClassifierfrom sklearn.model_selection import GridSearchCVfrom sklearn.model_selection import cross_val_scor...

代码笔记%matplotlib inlinefrom sklearn.tree import DecisionTreeClassifierfrom sklearn.ensemble import RandomForestClassifierfrom sklearn.datasets import load_winewine = load_wine()from sklearn....

代码笔记1.导库from sklearn import treefrom sklearn.datasets import load_winefrom sklearn.model_selection import train_test_split2. 加载数据，拆分wine = load_wine()Xtrain, Xtest, Ytrain, Ytest = train_...

在机器学习中，当我们基于某个业务建立模型并训练后，接下来我们需要评判模型好坏的时候需要基于混淆矩阵，ROC和AUC等来进行辅助判断。混淆矩阵也叫精度矩阵，是用来表示精度评价，为N *N的矩阵，用来判别分类好坏的指标。混淆矩阵中有以下几个概念：TP(True Positive): 被判定为正样本，实际也为正样本FN(False Negative):伪阴性 ，被判定为负样本，实际为正样...

pandas:http://pandas.pydata.org/pandas-docs/stable/reference/index.htmlscikit-learn:https://scikit-learn.org/stable/seaborn:http://seaborn.pydata.org/numpyhttps://www.numpy.org/devdocs/

numpy.newaxis 为数组增加一个新维度，机器学习的常用框架中处理的数据大多是矩阵，经常需要将一个向量转换成列矩阵或者行矩阵；同样numpy.expand_dims也可以实现对应的功能，效果如下： ...

使用jupyter-notebook打开ipynb,如下图所示，可以转换为其他格式的文件。 命令行的方式：ipynb转换为pythonjupyter nbconvert --to python my_file.ipynbipynb转换为mdjupyter nbconvert --to md my_file.ipynbipynb转为htmljupyter nbco...

创建key的步骤如下：step1 生成一个 notebook 配置文件: jupyter notebook --generate-configstep2 生成密码：jupyter notebook passwordjupyter notebook passwordEnter password: ****Verify password: ****[NotebookPasswordApp] Wrot...

ORB： Oriented Fast and Rotated BRIEF它是OpenCV_Labs 在2011年提出的，首先它是免费的，而SIFT 和 SURF 是受专利保护，并且要收费的，ORB 是他们的一个替代品，在计算能力比有限的设备上适合这个东西。它的效果： sift > orb > surf, 计算速度 orb > surf > sift#!/usr/bi

BRIEF：Binary Robust Independent Elementary Rreatures占位

#!/usr/bin/env pythonimport cv2img = cv2.imread('chess.png')gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)sift = cv2.SIFT()print sift.shapekp=sift.detect(gray, None)img = cv2.drawKeypoints(gr

Fast： Feature from Accelerated Segment Test适合在计算能力有限的设备上使用，其速度比SIFT 和SURF快，但是精度有限。#!/usr/bin/env pythonimport cv2img = cv2.imread('chess.png')gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

SURF: speeded-up robust features是SIFT 的加速版#!/usr/bin/env pythonimport cv2img = cv2.imread('chess.png')gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)surf = cv2.SURF(400)print surf.shapekp,

占位

图像金字塔操作，又分为高斯金字塔和拉普拉斯金字塔，高斯金字塔简单的理解为图像的downsample和upsample. 拉普拉斯金字塔是图像A  - up(down(A))的操作，看上去全是边缘。#!/usr/bin/env pythonimport cv2img = cv2.imread('frame.jpg')cv2.imshow('src', img)low

鼠标操作：#!/usr/bin/env pythonimport numpy as npimport cv2events=[i for i in dir(cv2) if 'EVENT' in i]#events=[i for i in dir(cv2)]print eventsdef draw_circle(event, x, y, flags, param): if

#!/usr/bin/env pythonimport numpy as npimport cv2img = np.zeros((512,512,3), np.uint8)cv2.line(img, (0,0), (511, 511), (255,0,0), 5) #line color (BGR)cv2.rectangle(img, (384,0), (510, 128), (

直接上code吧，算是学习笔记。#!/usr/bin/env pythonimport numpy as npimport cv2#imread第二个参数为0,强制将图像转换为灰度图像，如果为1，强制转换为彩色图像，-1应该是根据图像自身的信息决定#img = cv2.imread("src.png", 0)img = cv2.imread("src.png", 1)cv2.im

Winpython 下载地址：http://sourceforge.net/projects/winpython/?source=typ_redirect注意，里面有python2和python3对应的版本，自己根据实际需要选择一下，默认是下载V3的。下载后一路next安装就行了。里面已经默认包括了matlab对应的libwinpython安装目录下有一个spader.ex

先上两个code吧#!/usr/bin/env pythonimport cv2import numpy as npimg1 = cv2.imread('box.png', 0)img2 = cv2.imread('box_in_scene.png', 0)orb = cv2.ORB_create()kp1, des1 = orb.detectAndCompute(img1