Skip to content

Instantly share code, notes, and snippets.

@sencagri
Created October 26, 2017 09:48
Show Gist options
  • Save sencagri/919fb15bb6d01a38583ad5ac9a604d9d to your computer and use it in GitHub Desktop.
Save sencagri/919fb15bb6d01a38583ad5ac9a604d9d to your computer and use it in GitHub Desktop.
Pattern Recognition P1 Q2
import numpy as np
# deviance ve mean buradan giriliyor
dev = 10
var = dev ** 2
mean = 3
# örnek kümesinin boyutu
size = 150
nor_dis = np.random.normal(mean, dev, size)
mean_c = 0
for item in nor_dis:
mean_c+=item
mean_c = mean_c / size
var_c = 0
for item in nor_dis:
var_c += (item - mean_c) ** 2
var_c = var_c / (size - 1)
# girilen değerleri yazdır
print("prompt mean %s" %mean)
print("prompt var %s" %var)
print()
# standart hesaplama yöntemleri ile hesaplanan değerleri yazdır
print("calc mean %s" %mean_c)
print("calc var %s" %var_c)
# numpy kütüphanesini kullanarak karşılaştırma amaçlı mean ve varyans hesabı yap
lib_mean = np.mean(nor_dis)
lib_var = np.var(nor_dis)
print("lib calc mean %s" %lib_mean)
print("lib calc var %s" %lib_var)
print()
# prior bilgisini buradan giriyoruz
prior = 3
bayes_mean = 0
sov = size / var
oovo = 1 / var_c
bayes_mean = (mean_c * sov / (sov + oovo)) + (oovo * prior / (sov + oovo))
# bayes estimator sonucunda bulunan mean değerini ekranda göster
print();
print("bayes mean %s" %bayes_mean)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment