I want to learn deep learning from scratch, I watch the "standford Feifeili" video to start. here to record some of the code.

These answer are genius, anyhow I type my own code in a very simple way, it saves a lot of unuseful information and easier for me to grab the main points.
https://blog.csdn.net/zhangxb35/article/details/55223825
https://github.com/lightaime/cs231n/blob/master/assignment1/knn.ipynb

import random
import numpy as np
from cs231n.data_utils import load_CIFAR10
import matplotlib.pyplot as plt
cifar10_dir = 'cs231n/cifar-10-batches-py'
X_train, y_train, X_test, y_test = load_CIFAR10(cifar10_dir)

plt.imshow(X_train[10000].astype('uint8'))

Subsample the data for more efficient code execution in this exercise

num_training = 5000
mask = range(num_training)
X_train = X_train[mask]
y_train = y_train[mask]

num_test = 500
mask = range(num_test)
X_test = X_test[mask]
y_test = y_test[mask]

X_train = np.reshape(X_train, (X_train.shape[0], -1))
X_test = np.reshape(X_test, (X_test.shape[0], -1))
print (X_train.shape, X_test.shape)

from cs231n.classifiers import KNearestNeighbor
classifier = KNearestNeighbor()
classifier.train(X_train, y_train)

dists = classifier.compute_distances_two_loops(X_test)
#k = 1 (which is Nearest Neighbor).
y_test_pred=classifier.predict_labels(dists,k=5)
y_test_pred.shape

num_test=len(y_test)
num_correct=np.sum(y_test_pred==y_test)
accuracy=float(num_correct)/num_test

print('%d/%d=accuracy=%f'%(num_correct,num_test, accuracy))