sklearn SVM机器学习_数字识别

import os
from random import randrange
import numpy as np
from PIL import  Image, ImageDraw, ImageFont
from sklearn import svm
import cv2 as cv
from matplotlib import pyplot as plt
import pandas as pd


width, height = 20, 40
fontSize = 40
noiseRate = 0.01

def generateDigits(dstDir='temp', num=40000):
    dstDir = 'temp'
    if not os.path.isdir(dstDir):
        os.mkdir(dstDir)

    with open(dstDir+'\\digits.txt', 'w') as fp:
        for i in range(num):
            digit = str(randrange(10))
            im = Image.new('L', (width, height), 255)
            imDraw = ImageDraw.Draw(im)
            font = ImageFont.truetype(r"C:\Windows\Fonts\simsun.ttc", fontSize)
            imDraw.text((0,0), digit, font=font)

            for j in range(int(noiseRate*width*height)):
                w, h = randrange(1, width-1), randrange(height)
                c1 = im.getpixel((w, h))