Project - Noise removal from images

#1

I did not understand the whole code.

def sort_by_target(mnist):
reorder_train = np.array(sorted([(target, i) for i, target in enumerate(mnist.target[:60000])]))[:, 1]
reorder_test = np.array(sorted([(target, i) for i, target in enumerate(mnist.target[60000:])]))[:, 1]
mnist.data[:60000] = mnist.data[reorder_train]
mnist.target[:60000] = mnist.target[reorder_train]
mnist.data[60000:] = mnist.data[reorder_test + 60000]
mnist.target[60000:] = mnist.target[reorder_test + 60000]

from sklearn.datasets import fetch_openml
import numpy as np

mnist = fetch_openml(‘mnist_784’, version=1, cache=True)

mnist.target = mnist.target.astype(np.int8)
sort_by_target(mnist)

X, y = mnist[“data”], mnist[“target”]
X.shape
y.shape

Please help me to understand.
Please reply

#2

I did not understand.

shuffle our training data so that it ensures that we don’t miss out any digit in a cross validation fold.

np.random.seed(42)
shuffle_index = np.random.permutation(60000)
X_train, y_train = X_train[shuffle_index], y_train[shuffle_index]

please help me to understand.
Please reply.

#3

I did not understand the whole code.

Since KNN take a lot of time, we are trimming it here

X_train = X_train[:30000]
y_train = y_train[:30000]

X_test = X_test[:5000]
y_test = y_test[:5000]

import numpy.random as rnd

noise_train = rnd.randint(0, 100, (len(X_train), 784))
X_train_mod = X_train + noise_train
noise_test = rnd.randint(0, 100, (len(X_test), 784))
X_test_mod = X_test + noise_test
y_train_mod = X_train
y_test_mod = X_test

Please help me.
Please reply.

#4

Hi Nirav, As mentioned earlier, please provide a time for a Hangout call so that we can reachout to you so that we can resolve your queries.