Using the KNearest code modify it to use the features temperature, luminosity, and absolute magnitude to predict the spectral class of the stars FEATURE FEATURE FEATURE LABEL Temperature luminosity absolute magnitude spectral type 3068 0.0024 16.12 M 2800 0.0002 16.65 M 16500 0.013 11.89 B 12990 0.000085 12.23 F 8500 0.0005 14.5 A 39000 204000 -4.7 O 4015 282000 -11.39 K 6850 229000 -10.07 G import math from sklearn.datasets import load_iris from sklearn.model_selection import train_test_split def knearest(X_train, y_train, predict): min_dist = 10000 min_sample_idx = 0 for i1 in range(0, len(X_train)): squares = 0 for i2 in range(0, len(X_train[i1])): squares = squares + math.pow((X_train[i1][i2] - predict[i2]), 2) if math.sqrt(squares) < min_dist: min_dist = math.sqrt(squares) min_sample_idx = i1 return y_train[min_sample_idx] def trainAndScore(randomState): iris_dataset = load_iris() X_train, X_test, y_train, y_test = train_test_split(iris_dataset['data'], iris_dataset['target'], random_state = randomState) total = 0 correct = 0 for i in range(0, len(X_test)): ans = knearest(X_train, y_train, X_test[i]) total = total + 1 if ans == y_test[i]: correct = correct + 1 return correct/total score = trainAndScore(0) print(round(score, 2))
Using the KNearest code modify it to use the features temperature, luminosity, and absolute magnitude to predict the spectral class of the stars
FEATURE FEATURE FEATURE LABEL
Temperature luminosity absolute magnitude spectral type
3068 0.0024 16.12 M
2800 0.0002 16.65 M
16500 0.013 11.89 B
12990 0.000085 12.23 F
8500 0.0005 14.5 A
39000 204000 -4.7 O
4015 282000 -11.39 K
6850 229000 -10.07 G
import math
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
def knearest(X_train, y_train, predict):
min_dist = 10000
min_sample_idx = 0
for i1 in range(0, len(X_train)):
squares = 0
for i2 in range(0, len(X_train[i1])):
squares = squares + math.pow((X_train[i1][i2] - predict[i2]), 2)
if math.sqrt(squares) < min_dist:
min_dist = math.sqrt(squares)
min_sample_idx = i1
return y_train[min_sample_idx]
def trainAndScore(randomState):
iris_dataset = load_iris()
X_train, X_test, y_train, y_test = train_test_split(iris_dataset['data'], iris_dataset['target'], random_state = randomState)
total = 0
correct = 0
for i in range(0, len(X_test)):
ans = knearest(X_train, y_train, X_test[i])
total = total + 1
if ans == y_test[i]:
correct = correct + 1
return correct/total
score = trainAndScore(0)
print(round(score, 2))
Step by step
Solved in 3 steps with 1 images
