Pemilihan Komponen Utama

Pengurangan Dimensi dengan Python

Jeroen Boeye

Head of Machine Learning, Faktion

Menetapkan ambang ragam terserap

pipe = Pipeline([
        ('scaler', StandardScaler()),
        ('reducer', PCA(n_components=0.9))])

# Fit the pipe to the data
pipe.fit(poke_df)

print(len(pipe['reducer'].components_))

Jumlah komponen optimal

pipe.fit(poke_df)

var = pipe['reducer'].explained_variance_ratio_

plt.plot(var)

plt.xlabel('Principal component index')
plt.ylabel('Explained variance ratio')
plt.show()

rasio ragam terserap

Jumlah komponen optimal

pipe.fit(poke_df)

var = pipe['reducer'].explained_variance_ratio_

plt.plot(var)

plt.xlabel('Principal component index')
plt.ylabel('Explained variance ratio')
plt.show()

siku rasio ragam terserap

Operasi PCA

Skema fit-transform PCA 1

Operasi PCA

Skema fit-transform PCA 2

Operasi PCA

Skema fit-transform PCA 3

Kompresi gambar

wajah asli

Kompresi gambar

print(X_test.shape)

(15, 2914)

62 x 47 piksel = 2914 nilai grayscale

print(X_train.shape)

(1333, 2914)

Kompresi gambar

pipe = Pipeline([
        ('scaler', StandardScaler()),
        ('reducer', PCA(n_components=290))])

pipe.fit(X_train)

pc = pipe.fit_transform(X_test)

print(pc.shape)

(15, 290)

Membangun ulang gambar

pc = pipe.transform(X_test)

print(pc.shape)

(15, 290)

X_rebuilt = pipe.inverse_transform(pc)

print(X_rebuilt.shape)

(15, 2914)

img_plotter(X_rebuilt)

wajah terkompresi

Membangun ulang gambar

wajah asli

wajah terkompresi

Ayo berlatih!

Pengurangan Dimensi dengan Python