Python - Machine Learning

### PRINCIPAL COMPONENTS ANALYSIS + LINEAR REGRESSION

import pandas as pd
import numpy as np
import statsmodels.formula.api as sm
from sklearn import decomposition
from matplotlib import pyplot as plt
a0=pd.read_csv('DadosTeseLogit.csv',sep=',',header=0)
a=a0
a=a.ix[:,0:30]
a.skew()
a.corr()
k=29
pca=decomposition.PCA(n_components=k)
fa=decomposition.FactorAnalysis(a)
fa.score
u=pca.fit_transform(a,y=None)
pca.fit(a)
pca.explained_variance_ratio_
z=[float(i) for i in pca.explained_variance_ratio_]
z
pca.explained_variance_ratio_
for i in range(0,4):
    print("%.2f" % z[i])
pca.score_samples(a)
pca.fit_transform(a)
pca.fit(a)
pca.components_
pca.transform(a)
plt.plot(pca.components_[0])
z2=pca.inverse_transform(u)
z2.shape
x=np.transpose(z2)
x=z2[:,0:4]
x_test=x[10:20,:]

y=np.array(a0[[29]])
y=[item for sublist in y for item in sublist]

plt.plot(y)
pca.explained_variance_ratio_
linear=sm.OLS(y,x).fit()

pca.explained_variance_ratio_
plt.plot(z,marker='o',linestyle='--',color='r')
plt.xlabel('Components')
plt.ylabel('Variance Explained')
plt.title('Principal Components Analysis')
plt.show()
linear.summary()

linear.predict(x_test)
np.array(y[10:20])

accuracy=1-np.mean(abs(linear.predict(x_test)-np.array(y[10:20])))