unaoya

library(MASS)
library(rstan)
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

waic <- function(loglik){
  te <- -mean(log(colMeans(exp(log_lik))))
  fv <- mean(colMeans(log_lik^2) - colMeans(log_lik)^2)
  waic <- te + fv
  return(waic)

unaoya

def issol(x,y,z,p,k):
    return (pow(x,k,p) + pow(y,k,p)) % p == pow(z,k,p)

def count(p,k):
    c = 0
    for x in range(1,p):
        for y in range(1,p):
            for z in range(1,p):
                if issol(x,y,z,p,k):
                    c += 1

unaoya

import numpy as np
import tensorflow as tf
import edward as ed
from edward.models import Bernoulli, Beta, PointMass

ed.set_seed(42)

# DATA

n = 4

unaoya

import numpy as np
import math
import matplotlib.pyplot as plt

def primes(n):
    result = set(range(2,n))
    for i in range(2,math.floor(math.sqrt(n))+1):
        for j in range(2,n//i+1):
            result.discard(i*j)
    return list(result)

unaoya

unaoya

library(ggplot2)
library(tidyr)

n <- 10^6
N <- 1000
p <- 0.3
t <- 2

result <- rep(0, n)
Mmean <- rep(0, n)

unaoya

data {
  int N; //学習期間の長さ
  int N_pred; //予測期間の長さ
  vector[N] Y; //販売台数データ
}

parameters {
  vector[N] alpha; //状態のトレンド成分
  vector[N] season; //状態の季節成分
  real<lower=0> s_Y; //観測誤差の分散

unaoya

library(dplyr)
library(ggplot2)

compare <- function(N,n,p,dist){
  rng <- get(dist)
  sample.med <- c()
  sample.mea <- c()
  for(i in 1:N){
    sample <- rng(n)
    sample.med[i] <- sort(sample)[p]

unaoya

library(dplyr)
library(ggplot2)

set.seed(123)
runif(10)
runif(10)

ord_sampling <- function(N,n,p){
  x <- c()
  for(i in 1:N){

unaoya

import numpy as np
import matplotlib.pyplot as plt

mu = 0
sigma = 1

n = 5
sample = np.random.normal(size=n, loc=mu, scale=sigma)
print(np.mean(sample))
print(np.var(sample))