Skip to content

Instantly share code, notes, and snippets.

from collections import Counter
import sys
import time
import ray
""" This script is meant to be run from a pod in the same Kubernetes namespace
as your Ray cluster.
"""
@ray.remote
# Job to submit a Ray program from a pod outside a running Ray cluster.
apiVersion: batch/v1
kind: Job
metadata:
name: ray-test-job
spec:
template:
spec:
restartPolicy: Never
containers:
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
make examples/bernoulli/bernoulli
cd examples/bernoulli/
./bernoulli sample data file=bernoulli.data.R
./bernoulli variational data file=bernoulli.data.R
> print(rats_fit)
Inference for Stan model: rats.
4 chains, each with iter=1000; warmup=500; thin=1;
post-warmup draws per chain=500, total post-warmup draws=2000.
mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat
alpha[1] 239.90 0.07 2.70 234.72 238.03 239.90 241.71 245.11 1595 1.00
alpha[2] 247.89 0.06 2.78 242.43 246.02 247.94 249.74 253.40 2000 1.00
alpha[3] 252.53 0.06 2.76 247.12 250.71 252.57 254.35 257.86 1838 1.00
alpha[4] 232.53 0.06 2.71 227.16 230.69 232.56 234.40 237.75 1797 1.00
@bakfoo
bakfoo / rats_stan_advi.R
Last active November 11, 2020 22:28
Rstan ADVI Test using rats examples.
y <- read.table('https://raw.github.com/wiki/stan-dev/rstan/rats.txt', header = TRUE)
x <- c(8, 15, 22, 29, 36)
xbar <- mean(x)
N <- nrow(y)
T <- ncol(y)
data <- list(y=y, x=x, xbar=xbar, N=N, T=T)
rats_model <- stan_model(file = "https://raw.githubusercontent.com/stan-dev/example-models/master/bugs_examples/vol1/rats/rats.stan")
#NUTS

PyCon Jp 2015「エンジニアのためのベイズ推定入門」要項

0 チュートリアル環境の構築前の注意

確率論的プログラミングはまだ若い分野ですので,計算環境の構築方法が成熟していません.チュートリアルではpymc3やpystanを利用しますが,それらの開発者は基本的にUbuntuにAnaconda Pythonを利用してるので,まともに動作する環境はどうしてもUbuntu + Anacondaが中心になってしまいます.以下に構築前の注意を列挙します.

  • Windowsで確率論的プログラミングを行うことは,pymc3のtheanoのGPUの問題,pystanのプロセス制限の問題等,制約が多すぎて困難な道となります.Windowsを利用する場合,VMWare, VirtualBox, Vagrant等の仮想環境またはDocker等のコンテナ技術を利用してLinuxを用意するほうが精神衛生上望ましいと考えています.
  • Python環境の構築はAnacondaディストリビューションを前提とします.素のPythonから確率論的プログラミングの環境を構築することは,数々の罠に嵌まる可能性が否定できませんので推奨できません.
  • もしも,チュートリアル当日までに環境を構築できなくても問題ありません.チュートリアルはJupyter Notebookで行いますので,Docker Imageを用意するか,JupyterHub等によりWebブラウザからチュートリアルマテリアルにアクセスし実行できるようにします.
  • 皆様には大変申し訳ございませんが,JupyterHubが技術的理由により用意できませんでした.お詫びを申しあげます.代わりに以下のnbviewerで代替させていただきます.