Machine Learning in Action
Peter Harrington
一.分類
U1.機(jī)器學(xué)習(xí)基礎(chǔ)
U2.k-近鄰算法
U3.決策樹
U4.基于概率論的分類方法：樸素貝葉斯
U5.Logistic回歸
U6.支持向量機(jī)
U7.利用AdaBoost元算法提高分類性能
二.利用回歸預(yù)測數(shù)值型數(shù)據(jù)
U8.預(yù)測數(shù)值型數(shù)據(jù)：回歸
U9.樹回歸
三. 無監(jiān)督學(xué)習(xí)
U10.利用K-均值聚類算法對(duì)未標(biāo)注數(shù)據(jù)分組
U11.使用Apriori算法進(jìn)行關(guān)聯(lián)分析
U12.使用FP-growth算法來高效發(fā)現(xiàn)頻繁項(xiàng)集
四. 其他工具
U13.利用PCA來簡化數(shù)據(jù)
U14.利用SVD簡化數(shù)據(jù)
U15.大數(shù)據(jù)與MapReduce
U1.機(jī)器學(xué)習(xí)基礎(chǔ)
1.7 Numpy 函數(shù)基礎(chǔ)庫
from numpy import * #將Numpy函數(shù)庫的所有模塊引入當(dāng)前空間
random.rand(4,4) #4*4的隨機(jī)數(shù)組
array([[0.97471811, 0.17179221, 0.86431691, 0.61410609],
[0.45551856, 0.01440638, 0.19788693, 0.3447664 ],
[0.50013188, 0.48419994, 0.06638319, 0.15177441],
[0.5132277 , 0.45546545, 0.8121785 , 0.62891273]])
randMat = mat(random.rand(4,4)) #mat():將數(shù)組轉(zhuǎn)換成矩陣
randMat.I #.I操作符實(shí)現(xiàn)了矩陣求逆的運(yùn)算
matrix([[ 1.26422879, 0.33153666, -1.48110659, 1.02540337],
[ 0.36639991, -1.65434382, 1.37766836, 0.02631806],
[ 0.51142747, 0.5036215 , 0.5738588 , -1.40227222],
[-1.54037058, 0.80239577, 0.31063159, 0.92973999]])
invRandMat = randMat.I #存儲(chǔ)逆矩陣
randMat * invRandMat #矩陣與其逆矩陣相乘柿菩，結(jié)果應(yīng)為單位矩陣，但是出現(xiàn)偏差，原因?yàn)橛?jì)算機(jī)處理誤差
matrix([[ 1.00000000e+00, 1.11022302e-16, -2.22044605e-16,
-1.11022302e-16],
[-1.11022302e-16, 1.00000000e+00, -1.38777878e-16,
-1.11022302e-16],
[ 0.00000000e+00, 0.00000000e+00, 1.00000000e+00,
0.00000000e+00],
[ 1.11022302e-16, 1.11022302e-16, -1.11022302e-16,
1.00000000e+00]])
myEye = randMat * invRandMat #存儲(chǔ)相乘結(jié)果
myEye - eye(4) #查看誤差值豆挽，eye(4):為$4*4$單位矩陣
matrix([[ 0.00000000e+00, 1.11022302e-16, -2.22044605e-16,
-1.11022302e-16],
[-1.11022302e-16, 0.00000000e+00, -1.38777878e-16,
-1.11022302e-16],
[ 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,
0.00000000e+00],
[ 1.11022302e-16, 1.11022302e-16, -1.11022302e-16,
0.00000000e+00]])
在這里插入圖片描述
