pymc3-fix_dirichlet_shape.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Applied stickbreaking-transform to p and added transformed p_stickbreaking_ to model.\n",
      "Applied interval-transform to sd and added transformed sd_interval_ to model.\n",
      " [-----------------100%-----------------] 100 of 100 complete in 0.2 sec[[  9.99747864e-01   2.52136138e-04]\n",
      " [  9.51298297e-01   4.87017033e-02]\n",
      " [  9.99573099e-01   4.26900692e-04]\n",
      " [  7.42235726e-01   2.57764274e-01]\n",
      " [  9.01139410e-01   9.88605896e-02]\n",
      " [  2.95292867e-02   9.70470713e-01]\n",
      " [  9.49719240e-02   9.05028076e-01]\n",
      " [  6.70368286e-04   9.99329632e-01]\n",
      " [  5.84711404e-04   9.99415289e-01]\n",
      " [  3.64180996e-02   9.63581900e-01]]\n",
      "[0 0 0 0 0 1 1 1 1 1]\n"
     ]
    }
   ],
   "source": [
    "import os, sys\n",
    "sys.path.insert(0, os.path.expanduser('~/work/git/github/taku-y/pymc3'))\n",
    "\n",
    "import numpy as np\n",
    "import pymc3 as pm\n",
    "\n",
    "# artificial data\n",
    "k = 2\n",
    "ndata = 10\n",
    "data = np.array([1, 2.2, 2.5, 2.8, 4, 6, 7.2, 7.5, 7.8, 9])\n",
    "\n",
    "# model\n",
    "alpha = 0.1 * np.ones((ndata, k))\n",
    "with pm.Model() as model:\n",
    "    p = pm.Dirichlet('p', alpha, shape=(ndata, k))\n",
    "    mu = pm.Normal('mu', mu=5, sd=3, shape=k)\n",
    "    sd = pm.Uniform('sd', lower=0.1, upper=0.5, shape=k)\n",
    "    categ = pm.Categorical('categ', p=p, shape=ndata)\n",
    "    obs = pm.Normal('obs',\n",
    "                    mu=mu[categ],\n",
    "                    sd=sd[categ],\n",
    "                    observed=data)\n",
    "\n",
    "    step1 = pm.Metropolis(vars=[p, sd, mu, obs])\n",
    "    step2 = pm.ElemwiseCategorical(vars=[categ])\n",
    "\n",
    "    tr = pm.sample(100, step=[step1, step2])\n",
    "    print(tr['p'][-1, :])\n",
    "    print(tr['categ'][-1, :])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  },
  "nav_menu": {},
  "toc": {
   "navigate_menu": true,
   "number_sections": true,
   "sideBar": true,
   "threshold": 6,
   "toc_cell": false,
   "toc_section_display": "block",
   "toc_window_display": false
  },
  "widgets": {
   "state": {},
   "version": "1.1.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}