在本次“数据挖掘”比赛中感觉收获最大的还是对于神经网络的认识，在接近一周的时间里，研究了进40种神经网络模型，虽然在持续一周的挖掘比赛把自己折磨的惨不忍睹，但是收获颇丰。现在想想也挺欣慰自己在这段时间里接受新知识的能力。关于神经网络方面的理解会在后续博文中补充（刚提交完论文，还没来得及整理），先分享一下高次拟合方面的知识。
# coding=utf-8import matplotlib.pyplot as pltimport numpy as npimport scipy as spimport csvfrom scipy.stats import normfrom sklearn.pipeline import Pipelinefrom sklearn.linear_model import LinearRegressionfrom sklearn.preprocessing import PolynomialFeaturesfrom sklearn import linear_model''''' 数据导入 '''def loadDataSet(fileName): dataMat = [] labelMat = [] csvfile = file(fileName, 'rb') reader = csv.reader(csvfile) b = 0 for line in reader: if line[50] is '': b += 1 else: dataMat.append(float(line[41])/100*20+30) labelMat.append(float(line[25])*100) csvfile.close() print "absence time number: %d" % b return dataMat,labelMatxArr,yArr = loadDataSet('data.csv')x = np.array(xArr)y = np.array(yArr)# x = np.arange(0, 1, 0.002)# y = norm.rvs(0, size=500, scale=0.1)# y = y + x ** 2def rmse(y_test, y): return sp.sqrt(sp.mean((y_test - y) ** 2))def R2(y_test, y_true): return 1 - ((y_test - y_true) ** 2).sum() / ((y_true - y_true.mean()) ** 2).sum()def R22(y_test, y_true): y_mean = np.array(y_true) y_mean[:] = y_mean.mean() return 1 - rmse(y_test, y_true) / rmse(y_mean, y_true)plt.scatter(x, y, s=5)#分别进行1,2,3,6次拟合degree = [1, 2,3, 6]y_test = []y_test = np.array(y_test)for d in degree: #普通 # clf = Pipeline([('poly', PolynomialFeatures(degree=d)), # ('linear', LinearRegression(fit_intercept=False))]) # clf.fit(x[:, np.newaxis], y) # 岭回归 clf = Pipeline([('poly', PolynomialFeatures(degree=d)), ('linear', linear_model.Ridge())]) clf.fit(x[:, np.newaxis], y) y_test = clf.predict(x[:, np.newaxis]) print('多项式参数%s' %clf.named_steps['linear'].coef_) print('rmse=%.2f, R2=%.2f, R22=%.2f, clf.score=%.2f' % (rmse(y_test, y), R2(y_test, y), R22(y_test, y), clf.score(x[:, np.newaxis], y))) plt.plot(x, y_test, linewidth=2)plt.grid()plt.legend(['1', '2','3', '6'], loc='upper left')plt.show() 
