machine_learning_KernelPCA.py

from sklearn.datasets import fetch_mldata
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import KernelPCA
import matplotlib.pyplot as plt

# download MNIST data
mnist = fetch_mldata('MNIST original', data_home='./data/minist')

print(mnist)
## {'DESCR': 'mldata.org dataset: mnist-original', 'COL_NAMES': ['label', 'data'], 'target': array([0., 0., 0., ..., 9., 9., 9.]), 'data': array([[0, 0, 0, ..., 0, 0, 0],
##        [0, 0, 0, ..., 0, 0, 0],
##        [0, 0, 0, ..., 0, 0, 0],
##        ...,
##        [0, 0, 0, ..., 0, 0, 0],
##        [0, 0, 0, ..., 0, 0, 0],
##        [0, 0, 0, ..., 0, 0, 0]], dtype=uint8)}

print(mnist.data.shape)
## (70000, 784)

print(mnist.target.shape)
## (70000,)

# only use the first 10,000 images (to reduce runtime)
img = mnist.data[1:10000]

# standardization
scaler = StandardScaler()
scaler.fit(img)
img_scaled = scaler.transform(img)

# kernel PCA
kpca = KernelPCA(n_components=5, kernel='rbf', gamma=15,
                 fit_inverse_transform=True)
kpca.fit(img_scaled)
img_scaled_kpca = kpca.transform(img_scaled)

# get standardized & kernel PCA-transformed data
img_lowdim = scaler.inverse_transform(kpca.inverse_transform(img_scaled_kpca))


# plot figures
plt.figure(figsize=(8,4));

# original image
plt.subplot(1, 2, 1);
plt.imshow(img[1].reshape(28,28),
           cmap=plt.cm.gray, interpolation='nearest',
           clim=(0, 255));
plt.title('original image', fontsize = 20);

# standardized & PCA-transformed image
plt.subplot(1, 2, 2);
plt.imshow(img_lowdim[1].reshape(28, 28),
           cmap=plt.cm.gray, interpolation='nearest',
           clim=(0, 255));
plt.title('5 explained variances', fontsize = 20);