shape = random_indexes.shape
axs[i, j].imshow(images[random_indexes[i * rows + j]])
axs[i, j].set_title(predictions[random_indexes[i * rows + j]])
for ax in axs.flat:
ax.label_outer()
plt.show()
print('Loading RBM...')
rbm = RBM()
paths = [
'.\\RBMs\\weights_1587597688.2092822.csv',
'.\\RBMs\\input_bias_1587597688.2092822.csv',
'.\\RBMs\\hidden_bias_1587597688.2092822.csv'
]
rbm.load_machine(paths)
rbm.load_data('.\\Dataset\\train-images')
train_data = rbm.generate_hidden_representation(rbm.samples)
train_labels = read_idx('train-labels')
rbm.load_data('.\\Dataset\\test-images')
test_data = rbm.generate_hidden_representation(rbm.samples)
test_labels = read_idx('test-labels')
print('Loading SVM classifier...')
classifier = SVMClassifier()
classifier.load_classifier([
'.\\Classifiers\\coef_1587589776.9200084.csv',
'.\\Classifiers\\intercept_1587589776.9218214.csv',
'.\\Classifiers\\classes_1587589776.9238167.csv'
