
# distance 함수 구현하기
def distance(movie1, movie2):
    length_difference = (movie1[0] - movie2[0]) ** 2
    year_difference = (movie1[1] - movie2[1]) ** 2
    distance = (length_difference + year_difference) ** 0.5
    return distance


star_wars = [125, 1977]
raiders = [115, 1981]
mean_girls = [97, 2004]


# star_wars와 raiders의 거리
print(distance(star_wars, raiders))

# star_wars와 mean_girls의 거리
print(distance(star_wars, mean_girls))

# 스타워즈는 어느 영화와 더 비슷합니까?
# 정답:

10.770329614269007
38.897300677553446


# distance 함수 수정
def distance(movie1, movie2):
    squared_difference = 0
    for i in range(len(movie1)):
        squared_difference += (movie1[i] - movie2[i]) ** 2
    final_distance = squared_difference ** 0.5
    return final_distance


star_wars = [125, 1977, 11000000]
raiders = [115, 1981, 18000000]
mean_girls = [97, 2004, 17000000]


# star_wars와 raiders의 거리
print(distance(star_wars, raiders))

# star_wars와 mean_girls의 거리
print(distance(star_wars, mean_girls))

# 스타워즈는 어느 영화와 더 비슷합니까?
# 정답: mean girls

7000000.000008286
6000000.000126083


# min_max_normalize 함수 구현
def min_max_normalize(lst):
    # 1.
    minimum = min(lst)
    maximum = max(lst)
    # 2.
    normalized = []
  
    for value in lst:
        normalized_num = (value - minimum) / (maximum - minimum)
        normalized.append(normalized_num)
  
    return normalized


release_dates = [1897, 1998, 2000, 1948, 1962, 1950, 1975, 1960,
                 2017, 1937, 1968, 1996, 1944, 1891, 1995, 1948, 
                 2011, 1965, 1891, 1978]


from pprint import pprint

# 결과값 출력하기
pprint(min_max_normalize(release_dates))

[0.047619047619047616,
 0.8492063492063492,
 0.8650793650793651,
 0.4523809523809524,
 0.5634920634920635,
 0.46825396825396826,
 0.6666666666666666,
 0.5476190476190477,
 1.0,
 0.36507936507936506,
 0.6111111111111112,
 0.8333333333333334,
 0.42063492063492064,
 0.0,
 0.8253968253968254,
 0.4523809523809524,
 0.9523809523809523,
 0.5873015873015873,
 0.0,
 0.6904761904761905]

[
  [0.30, 'Superman II'],
  [0.31, 'Finding Nemo'],
  ...
  ...
  [0.38, 'Blazing Saddles']
]


# 영화 데이터 받아오기
import urllib.request
import ast

movie_dataset_url = "https://raw.githubusercontent.com/inikoreaackr/ml_datasets/main/movie_dataset.txt"
movie_datasets = []
for line in urllib.request.urlopen(movie_dataset_url):
    movie_datasets.append(line.decode('utf-8'))

movie_dataset = ast.literal_eval(movie_datasets[0][16:])
movie_labels = ast.literal_eval(movie_datasets[1][15:])


print(movie_dataset['The Avengers'])
print(movie_labels['The Avengers'])

[0.018009887923225047, 0.4641638225255973, 0.9550561797752809]
1


# classify 함수를 구현
def classify(unknown, dataset, k):
    distances = []
    #Looping through all points in the dataset
    for title in dataset:
        movie = dataset[title]
        distance_to_point = distance(movie, unknown)
        #Adding the distance and point associated with that distance
        distances.append([distance_to_point, title])
    distances.sort()
    #Taking only the k closest points
    neighbors = distances[0:k]
    
    return neighbors


# `classify` 함수를 테스트하고 결과를 출력합니다.
print(classify([.4, .2, .9], movie_dataset, 5))

[[0.08273614694606074, 'Lady Vengeance'], [0.22989623153818367, 'Steamboy'], [0.23641372358159884, 'Fateless'], [0.26735445689589943, 'Princess Mononoke'], [0.3311022951533416, 'Godzilla 2000']]

[
  [0.083, 'Lady Vengeance'],
  [0.236, 'Steamboy'],
  ...
  ...
  [0.331, 'Godzilla 2000']
]


# classify 함수 수정하기
# 1.
def classify(unknown, dataset, labels, k):
    distances = []
    for title in dataset:
        movie = dataset[title]
        distance_to_point = distance(movie, unknown)
        #Adding the distance and point associated with that distance
        distances.append([distance_to_point, title])
    distances.sort()
    #Taking only the k closest points
    neighbors = distances[0:k]
    # 2.
    num_good = 0
    num_bad = 0
    # 3.
    for neighbor in neighbors:
        title = neighbor[1]
        if labels[title] == 0:
            num_bad += 1
        elif labels[title] == 1:
            num_good += 1
    # 4.        
    if num_good > num_bad:
        return 1
    else:
        return 0


# `classify` 함수를 테스트하고 결과를 출력합니다.
print(classify([.4, .2, .9], movie_dataset, movie_labels, 5))

1


# "Call Me By Your Name"이 훈련 데이터에 있는지 확인하기
print("Call Me By Your Name" in movie_dataset)

False


# my_movie 변수 생성하기
my_movie = [3500000, 132, 2017]


# x, y, z points array 받아오기
import urllib.request
import ast
import numpy as np

points_url = "https://raw.githubusercontent.com/inikoreaackr/ml_datasets/main/points_xyz.txt"
points = []
for line in urllib.request.urlopen(points_url):
    points.append(line.decode('utf-8'))

training_x = np.array(ast.literal_eval(points[0]))
training_y = np.array(ast.literal_eval(points[1]))
training_z = np.array(ast.literal_eval(points[2]))


def normalize_dimension(value, lst):
    minimum = min(lst)
    maximum = max(lst)
    return (value - minimum) / (maximum - minimum)

def normalize_point(pt):
    global training_x
    global training_y
    global training_z
    newx = normalize_dimension(pt[0], training_x)
    newy = normalize_dimension(pt[1], training_y)
    newz = normalize_dimension(pt[2], training_z)
    return [newx, newy, newz]


# my_movie 변수를 정규화합니다.
normalized_my_movie = normalize_point(my_movie)
print(normalized_my_movie)

[0.00028650338197026213, 0.3242320819112628, 1.0112359550561798]


# classify 함수를 호출합니다.
classify(normalized_my_movie, movie_dataset, movie_labels, 5)

# TODO: Call Me By Your Name 영화의 분류된 lable은 무엇인가요?
# 정답:

1


import pandas as pd

data_url = 'https://raw.githubusercontent.com/inikoreaackr/ml_datasets/master/breastcancer.csv'

df = pd.read_csv(data_url)
df.head()


# id 와 Unnamed: 32 column을 삭제한 후 column들의 리스트를 출력합니다.
df.drop(columns=['id', 'Unnamed: 32'], inplace=True)
df.columns

Index(['diagnosis', 'radius_mean', 'texture_mean', 'perimeter_mean',
       'area_mean', 'smoothness_mean', 'compactness_mean', 'concavity_mean',
       'concave points_mean', 'symmetry_mean', 'fractal_dimension_mean',
       'radius_se', 'texture_se', 'perimeter_se', 'area_se', 'smoothness_se',
       'compactness_se', 'concavity_se', 'concave points_se', 'symmetry_se',
       'fractal_dimension_se', 'radius_worst', 'texture_worst',
       'perimeter_worst', 'area_worst', 'smoothness_worst',
       'compactness_worst', 'concavity_worst', 'concave points_worst',
       'symmetry_worst', 'fractal_dimension_worst'],
      dtype='object')


# 각 column에 na 값이 있지는 않은지 확인합니다.
df.isna().any()

diagnosis                  False
radius_mean                False
texture_mean               False
perimeter_mean             False
area_mean                  False
smoothness_mean            False
compactness_mean           False
concavity_mean             False
concave points_mean        False
symmetry_mean              False
fractal_dimension_mean     False
radius_se                  False
texture_se                 False
perimeter_se               False
area_se                    False
smoothness_se              False
compactness_se             False
concavity_se               False
concave points_se          False
symmetry_se                False
fractal_dimension_se       False
radius_worst               False
texture_worst              False
perimeter_worst            False
area_worst                 False
smoothness_worst           False
compactness_worst          False
concavity_worst            False
concave points_worst       False
symmetry_worst             False
fractal_dimension_worst    False
dtype: bool


# label이 어떻게 구성되어있나 봅니다.
df['diagnosis'].value_counts()

B    357
M    212
Name: diagnosis, dtype: int64


# 데이터를 X와 y로 나누고 train, test 셋으로 나눕니다.
from sklearn.model_selection import train_test_split

y = df['diagnosis']
X = df.drop(columns=['diagnosis'])

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42, shuffle=True)


# scikit-learn의 KNN classifier을 생성하고 훈련시킵니다.
from sklearn.neighbors import KNeighborsClassifier

knn = KNeighborsClassifier()
knn.fit(X_train, y_train)

KNeighborsClassifier()


# test 셋에 대하여 정확도를 출력해봅니다.
knn.score(X_test, y_test)

0.9590643274853801


n_neighbors_list = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
score = []

# 가장 적합한 k를 찾아봅니다.
for k in n_neighbors_list:
  knn = KNeighborsClassifier(n_neighbors=k)
  knn.fit(X_train, y_train)
  score.append(knn.score(X_test, y_test))


n_neighbors_list[np.argmax(score)]

9


import matplotlib.pyplot as plt

plt.plot(n_neighbors_list, score)

[<matplotlib.lines.Line2D at 0x7f0adc331190>]

	id	diagnosis	radius_mean	texture_mean	perimeter_mean	area_mean	smoothness_mean	compactness_mean	concavity_mean	concave points_mean	...	texture_worst	perimeter_worst	area_worst	smoothness_worst	compactness_worst	concavity_worst	concave points_worst	symmetry_worst	fractal_dimension_worst	Unnamed: 32
0	842302	M	17.99	10.38	122.80	1001.0	0.11840	0.27760	0.3001	0.14710	...	17.33	184.60	2019.0	0.1622	0.6656	0.7119	0.2654	0.4601	0.11890	NaN
1	842517	M	20.57	17.77	132.90	1326.0	0.08474	0.07864	0.0869	0.07017	...	23.41	158.80	1956.0	0.1238	0.1866	0.2416	0.1860	0.2750	0.08902	NaN
2	84300903	M	19.69	21.25	130.00	1203.0	0.10960	0.15990	0.1974	0.12790	...	25.53	152.50	1709.0	0.1444	0.4245	0.4504	0.2430	0.3613	0.08758	NaN
3	84348301	M	11.42	20.38	77.58	386.1	0.14250	0.28390	0.2414	0.10520	...	26.50	98.87	567.7	0.2098	0.8663	0.6869	0.2575	0.6638	0.17300	NaN
4	84358402	M	20.29	14.34	135.10	1297.0	0.10030	0.13280	0.1980	0.10430	...	16.67	152.20	1575.0	0.1374	0.2050	0.4000	0.1625	0.2364	0.07678	NaN

K-Nearest Neighbor Classifier (KNN)¶

1-1. 데이터 포인트 간의 거리 - 2D¶

Practice 1¶

1-2. 데이터 포인트 간의 거리 - 3D¶

Practice 2¶

1-3. 척도가 다른 데이터: 정규화¶

1-3-1. Normalization vs Standardization¶

Practice 3¶

1-4. Nearest Neighbors 찾기¶

Practice 4¶

1-5. Neighbors 세기¶

Practice 5¶

1-6. 하나의 데이터 포인트 분류하기¶

1-7. scikit-learn으로 KNN 구현하기¶

1-8. 가장 적합한 `k` 찾기¶

TAEWON KIM

Recent posts

Comments

K-Nearest Neighbor Classifier (KNN)¶

1-1. 데이터 포인트 간의 거리 - 2D¶

Practice 1¶

1-2. 데이터 포인트 간의 거리 - 3D¶

Practice 2¶

1-3. 척도가 다른 데이터: 정규화¶

1-3-1. Normalization vs Standardization¶

Practice 3¶

1-4. Nearest Neighbors 찾기¶

Practice 4¶

1-5. Neighbors 세기¶

Practice 5¶

1-6. 하나의 데이터 포인트 분류하기¶

1-7. scikit-learn으로 KNN 구현하기¶

1-8. 가장 적합한 k 찾기¶

TAEWON KIM

Recent posts

Newsletter

Comments

1-8. 가장 적합한 `k` 찾기¶