lectures/Lecture_12.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Bayesian Statistical Inference: V\n",
    "\n",
    "*S. R. Taylor (2021)*\n",
    "\n",
    "##### Reading:\n",
    "\n",
    "- [Textbook](http://press.princeton.edu/titles/10159.html) Chapter 5.\n",
    "\n",
    "---\n",
    "\n",
    "## Contents\n",
    "* [Gibbs Sampling](#one)\n",
    "    * [Benefits Of Gibbs Sampling](#onea)\n",
    "    * [Drawbacks Of Gibbs Sampling](#oneb)\n",
    "    * [Conjugate Priors (Again)](#onec)\n",
    "* [Practical Evidence Evaluation & Model Comparison](#two)\n",
    "    * [Dynesty](#twoa)\n",
    "    * [UltraNest](#twob)\n",
    "\n",
    "---\n",
    "\n",
    "***<font color='red'>Notebook credit will be given in this lecture for successfully producing plots and output from running the code examples.</font>.***\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "***BEFORE YOU START:*** Please pause for a few minutes and install these packages `dynesty`, `UltraNest` before going through today's notebook. Make sure this notebook is in the correct Python kernel for the class conda environment before executing each of the following cells in turn. *(The second package you'll install below is just cython, which wraps C code in Python structures.)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "!pip3 install dynesty"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip3 install cython"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip3 install ultranest"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Gibbs Sampling <a class=\"anchor\" id=\"one\"></a>\n",
    "\n",
    "**Gibbs sampling** is an MCMC method that avoids acceptance-rejection techniques. Instead, one sweeps through each parameter (or block of parameters) to draw from their ***conditional probability distributions***, with all other parameters fixed to their current values. \n",
    "\n",
    "After sampling for a sufficiently large number of Gibbs steps, the principles of MCMC guarantee that **this process of sequential conditional probability drawing will converge to the joint posterior distribution of the overall model parameter space**. A standard Gibbs algorithm is shown below.\n",
    "\n",
    "---\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "---\n",
    "\n",
    "To break this algorithm up into words, we\n",
    "\n",
    "- We initialize the sampler at some random point in parameter space.\n",
    "\n",
    "\n",
    "- Fix all parameters except the first one. Draw a random value from the conditional posterior probability distribution of this first parameter **given** the fixed values of all other parameters.\n",
    "\n",
    "\n",
    "- Fix all parameters except the second one. Draw a random value from the conditional posterior probability distribution of this second parameter **given** the fixed values of all other parameters **and the updated value of the first parameter**.\n",
    "\n",
    "\n",
    "- *et cetera* for all parameters.\n",
    "\n",
    "\n",
    "- Repeat for many Gibbs steps.\n",
    "\n",
    "\n",
    "\n",
    "### Benefits of Gibbs Sampling <a class=\"anchor\" id=\"onea\"></a>\n",
    "\n",
    "- By drawing directly from the posterior conditionals, the **auto-correlation length can be exceptionally small**, with **minimal burn-in**. \n",
    "- **It's also fast**; sequential draws directly from the parameter posterior conditionals means that we are not rejecting any points. \n",
    "\n",
    "### Drawbacks of Gibbs Sampling <a class=\"anchor\" id=\"oneb\"></a>\n",
    "\n",
    "- You need to know the form of the conditional probability distributions for each parameter (or parameter blocks), and how to draw samples from it. \n",
    "- This can be a non-trivial problem, so typically a lot of effort is placed in manipulating the form of the posterior to find a conditional that is a standard probability distribution. This is where **conjugate priors** become really handy. \n",
    "\n",
    "### Conjugate Priors (Again) <a class=\"anchor\" id=\"onec\"></a>\n",
    "\n",
    "> *These are parameter priors for which the the posterior lies in the same family of distributions as the prior.* \n",
    "\n",
    "***EXAMPLE***\n",
    "\n",
    "- Imagine we have a Gaussian likelihood function with mean $\\mu$ and variance $\\sigma^2$ parameters. \n",
    "- The conjugate prior on $\\mu$ when $\\sigma^2$ is known (as is assumed in Gibbs when sweeping through each parameter) is a Gaussian distribution.\n",
    "- Thus the conditional posterior distribution on $\\mu$ is simply a Gaussian. \n",
    "- Similarly, the conjugate prior on $\\sigma^2$ when $\\mu$ is an inverse gamma distribution.\n",
    "- Thus the conditional posterior distribution on $\\sigma^2$ is an inverse gamma distribution.\n",
    "\n",
    "\n",
    "Let's look at the speciifc example we had in `Lecture_11` where we used the `PTMCMCSampler` to generate an MCMC chain of posterior samples of $\\mu$ and $\\sigma$ for homoescedastic Gaussian data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metropolis-Hastings Approach"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# Execute this cell\n",
    "import numpy as np\n",
    "from scipy import stats\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format='retina'\n",
    "\n",
    "from scipy import integrate\n",
    "from scipy.stats import norm\n",
    "from astroML.plotting import hist\n",
    "from astroML.plotting.mcmc import convert_to_stdev\n",
    "from astroML.plotting import setup_text_plots\n",
    "setup_text_plots(fontsize=8, usetex=True)\n",
    "from astroML.plotting.mcmc import convert_to_stdev\n",
    "import warnings; warnings.simplefilter('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# don't worry about any warning messages regarding mpi4py\n",
    "from PTMCMCSampler.PTMCMCSampler import PTSampler as ptmcmc\n",
    "\n",
    "# generating data\n",
    "np.random.seed(21)\n",
    "Ndata = 100\n",
    "mu = 1.0\n",
    "sigma = 0.5 \n",
    "data = stats.norm(mu, sigma).rvs(Ndata)\n",
    "\n",
    "# mcmc functions\n",
    "#################\n",
    "def Likelihood(x, data):\n",
    "    # Gaussian likelihood \n",
    "    return np.prod(stats.norm(loc=x[0], scale=x[1]).pdf(data))\n",
    "\n",
    "def Prior(x):\n",
    "    # create a uniform prior in mu and sigma\n",
    "    prior_mu = stats.uniform.pdf(x[0], loc=0.5, scale=1.0)\n",
    "    prior_sigma = stats.uniform.pdf(x[1], loc=0.1, scale=0.8)\n",
    "    return  prior_mu * prior_sigma\n",
    "\n",
    "def logLikelihood(x):\n",
    "    return np.log(Likelihood(x, data))\n",
    "\n",
    "def logPrior(x):\n",
    "    return np.log(Prior(x))\n",
    "###############\n",
    "\n",
    "\n",
    "# dimension of parameter space\n",
    "ndim = 2\n",
    "\n",
    "# initial jump covariance matrix\n",
    "cov = np.diag(np.ones(ndim) * 0.01**2)\n",
    "\n",
    "# intialize sampler\n",
    "sampler = ptmcmc(ndim, logLikelihood, logPrior, cov, \n",
    "                 outDir='./my_ptmcmc_chain', resume=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sampler for N steps\n",
    "N = int(3e4)\n",
    "x0 = np.array([0.9, 0.3])\n",
    "sampler.sample(x0, N, SCAMweight=30, AMweight=15, DEweight=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load in the chain\n",
    "chain = np.loadtxt('./my_ptmcmc_chain/chain_1.txt')\n",
    "\n",
    "import acor\n",
    "thin = int(acor.acor(chain[:,0])[0])\n",
    "print('Taking every n = ' + str(thin) + ' samples based on ac length.')\n",
    "print('Burning first n = ' + str(500) + ' samples.')\n",
    "\n",
    "plt.plot(chain[500::thin,0]);\n",
    "plt.plot(chain[500::thin,1]);\n",
    "\n",
    "# injected values\n",
    "plt.hlines(1.0, 0.0, 1500, color='k', lw=2, ls='dashed');\n",
    "plt.hlines(0.5, 0.0, 1500, color='k', lw=2, ls='dashed');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**On my machine it took about 30 seconds to get the $3000$ chain samples, of which I burnt the first 500, and took every second sample for autocorrelation thinning. This left me with $1250$ samples generated in 30 seconds with an acceptance rate of $\\sim 50\\%$.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Gibbs Approach\n",
    "\n",
    "Code for this example is adapted from https://wiseodd.github.io/techblog/2015/10/09/gibbs-sampling/, https://bookdown.org/rdpeng/advstatcomp/gibbs-sampler.html, and https://www.wikiwand.com/en/Conjugate_prior. \n",
    "\n",
    "---\n",
    "\n",
    "To make the Gibbs approach work, we need the conjugate priors for the mean and variance of a Gaussian likelihood. These are a Gaussian for the mean, and an inverse Gamma for the variance. We'll make these broad so that our posterior constraints are data-dominated.\n",
    "\n",
    "$$ p(\\mu) = \\mathcal{N}(\\mu_0,\\sigma_0) = \\frac{1}{\\sqrt{2\\pi\\sigma_0^2}}\\exp\\left(\\frac{-\\mu^2}{2\\sigma_0^2} \\right)$$\n",
    "\n",
    "where $\\mu_0,\\sigma_0$ are the location and scale parameters of the Gaussian prior on $\\mu$, and \n",
    "\n",
    "$$ p(\\sigma^2) = \\mathrm{IG}(\\alpha,\\beta) = \\frac{\\beta^\\alpha}{\\Gamma(\\alpha)} (1/\\sigma^2)^{\\alpha+1}\\exp(-\\beta/\\sigma^2)$$\n",
    "\n",
    "where $\\alpha,\\beta$ are parameters defining the shape of the inverse Gamma prior on $\\sigma^2$.\n",
    "\n",
    "It is algebraically tedious and relatively straightforward to multiply the Gaussian likelihood of the data with each of these priors in turn to work out the posterior conditional distributions on $\\mu$ and $\\sigma^2$. These turn out to be\n",
    "\n",
    "$$ p(\\mu|\\{x_i\\}) = \\mathcal{N}(\\mu_*,\\sigma_*)$$\n",
    "\n",
    "where \n",
    "\n",
    "$$\\mu_* = \\left( \\frac{1}{\\sigma_0^2} + \\frac{n}{\\sigma^2}\\right)^{-1} \\left(\\frac{\\mu_0}{\\sigma_0^2} + \\frac{\\sum^n_{i=1}x_i}{\\sigma^2} \\right)$$\n",
    "\n",
    "$$\\sigma^2_* = \\left( \\frac{1}{\\sigma_0^2} + \\frac{n}{\\sigma^2}\\right)^{-1} $$\n",
    "\n",
    "and \n",
    "\n",
    "$$ p(\\sigma^2|\\{x_i\\}) = \\mathrm{IG}(\\alpha_*,\\beta_*)$$\n",
    "\n",
    "where \n",
    "\n",
    "$$ \\alpha_* = \\alpha + n/2$$\n",
    "\n",
    "$$ \\beta_* = \\beta + \\frac{\\sum^n_{i=1}(x_i-\\mu)^2}{2}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# conditional of \\mu\n",
    "def p_mu_given_sigma2(sigma2, x, mu0, sigma0):\n",
    "    sigma_star2 = 1.0 / ( (1.0 / sigma0**2.) + (x.shape[0] / sigma2) )\n",
    "    mu_star = sigma_star2 * ((mu0/sigma0**2.) + (np.sum(x)/sigma2))\n",
    "    return np.random.normal(loc=mu_star, scale=np.sqrt(sigma_star2))\n",
    "\n",
    "# conditional of \\sigma^2\n",
    "def p_sigma2_given_mu(mu, x, alpha, beta):\n",
    "    alpha_star = alpha + x.shape[0]/2.\n",
    "    beta_star = beta + np.sum((x-mu)**2.)/2.\n",
    "    return stats.invgamma.rvs(a=alpha_star, scale=beta_star)\n",
    "\n",
    "\n",
    "def gibbs_sampling(data, mu_initial, sigma2_initial,\n",
    "                   mu0, sigma0, alpha, beta, iter=2000):\n",
    "    samples = np.zeros((iter, 2))\n",
    "    samples[0,:] = [mu_initial, sigma2_initial]\n",
    "\n",
    "    for ii in range(1,iter):\n",
    "        # sequential sampling of posterior conditionals\n",
    "        ### mu first\n",
    "        mu_new = p_mu_given_sigma2(samples[ii-1,1], data, mu0, sigma0)\n",
    "        samples[ii,0] = mu_new\n",
    "        ### then sigma2\n",
    "        sigma2_new = p_sigma2_given_mu(samples[ii,0], data, alpha, beta)\n",
    "        samples[ii,1] = sigma2_new\n",
    "\n",
    "    return samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# starting point\n",
    "mu_initial = 0.2\n",
    "sigma2_initial = 0.6\n",
    "\n",
    "# priors\n",
    "mu0 = 1.0\n",
    "sigma0 = 0.9\n",
    "alpha = 1.0\n",
    "beta = 1.0\n",
    "\n",
    "## SAMPLE!\n",
    "def run_gibbs():\n",
    "    samples = gibbs_sampling(data, mu_initial, sigma2_initial, \n",
    "                             mu0, sigma0, alpha, beta)\n",
    "    return samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = run_gibbs()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "thin = int(acor.acor(samples[:,0])[0])\n",
    "print('Taking every n = ' + str(thin) + ' samples based on ac length.')\n",
    "print('Burning first n = ' + str(500) + ' samples.')\n",
    "\n",
    "plt.plot(___); # burned and thinned mean\n",
    "plt.plot(___**0.5); # burned and thinned sigma\n",
    "\n",
    "# injected values\n",
    "plt.hlines(1.0, 0.0, 1500, color='k', lw=2, ls='dashed');\n",
    "plt.hlines(0.5, 0.0, 1500, color='k', lw=2, ls='dashed');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Run the following code to time the Gibbs sampling of $2000$ posterior draws. No points are rejected. You'll see that this is about **two orders of magnitude faster than Metropolis-Hastings**, while also giving a greater number of sampled values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%timeit -n1 run_gibbs()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Gibbs sampling is awesome...provided that you can work out the analytic conditional posteriors as standard distribution forms.** \n",
    "\n",
    "But even if you can't, all is not lost. You can get inventive by embedding a short Metropolis-Hastings block within the Gibbs algorithm. For example, if there are parameters for which you can not directly draw from the conditional, then your Gibbs step for that parameter could be a short Metropolis-Hastings MCMC run. \n",
    "\n",
    "The goal is to run this sufficiently long enough that you have drawn a single quasi-independent random sample from the parameter posterior conditional distribution, with all other parameters fixed. You can then proceed through the remainder of your Gibbs steps as normal. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Practical Evidence Evaluation & Model Selection <a class=\"anchor\" id=\"two\"></a>\n",
    "\n",
    "We've discussed Bayesian parameter estimation algorithms in detail. But what about evidence evaluation and model comparison? We already have the AIC and BIC as proxies. Grid-based methods for integrating the likelihood $\\times$ prior are far too crude. \n",
    "\n",
    "There are some further numerical techniques that we'll briefly discuss, before looking at examples of **Nested Sampling** algorithms to compute the Bayesian evidence. \n",
    "\n",
    "### Savage-Dickey Density Ratio\n",
    "\n",
    "This is great for when we want to compare models that have nested parameters, e.g. (1) a model with just noise, and (2) a model with a signal and noise. We need to be able to identify a parameter in the signal model that acts as an amplitude, $A$, that can effectively switch the signal off when it is zero. \n",
    "\n",
    "Thus $p_1(d|n) = p_2(d|A=0,n)$.\n",
    "\n",
    "With MCMC sampling we can map out the full parameter posterior distribution of model (2). We need to make sure we get good mixing and exploration of the parameter $A$ so that we can reconstruct its posterior density at $A=0$. \n",
    "\n",
    "The Bayes factor of Model $2$ (signal and noise) versus Model $1$ (noise alone) can be computed using the **[Savage-Dickey Density Ratio](https://www.jstor.org/stable/2958475?casa_token=Dc8wjaKRj5UAAAAA%3A4IhKgqNXc4jpan6hRxW7tBpXTI4i2236BulPTRxj4FcivD988vqJw-N75Dtw7z4Lh3YtPsDIceS8NDtIXAIRIO1T227h9VYZ2prQpwb46WLBB0GGz8w&seq=1#metadata_info_tab_contents)** approximation,\n",
    "\n",
    "$$ \\mathcal{B} = \\frac{\\mathcal{Z}_2}{\\mathcal{Z}_1} = \\frac{p(A=0)}{p(A=0|d,n)}$$\n",
    "\n",
    "which is the prior density at $A=0$ divided by the conditional posterior density at $A=0$. We can easily compute the latter by making a histogram of posterior samples in $A$ and reading off the normalized pdf at $A=0$.\n",
    "\n",
    "\n",
    "### Product-space Sampling\n",
    "\n",
    "Product-space sampling sounds fancy, but is super easy. In this type of model comparison, we never explicitly try to compute the Bayesian evidence. \n",
    "\n",
    "- We run a big MCMC chain that includes all parameters of all models, in addition to a model indexing parameter, $n$. \n",
    "- We define distinct ranges for this parameter, such that in certain sampling ranges it will trigger one of the model likelihoods to be used, e.g. if $-0.5<n<0.5$ use the likelihood for model $1$. \n",
    "- After sampling sufficiently long, the MCMC chain will spend different amounts of time in differnt regions of $n$ in proportion to the Bayesian odds ratio. \n",
    "- Hence, the ratio of samples in model sub-ranges of $n$ gives the Bayesian odds ratio. Simple!\n",
    "- We use this alot in PTA GW searches, see [https://arxiv.org/abs/1801.02617](https://arxiv.org/abs/1801.02617), [https://arxiv.org/abs/2009.04496](https://arxiv.org/abs/2009.04496), and [https://arxiv.org/abs/2006.04810](https://arxiv.org/abs/2006.04810). \n",
    "- Refs [here](https://rss.onlinelibrary.wiley.com/doi/abs/10.1111/j.2517-6161.1995.tb02042.x) and [here](https://amstat.tandfonline.com/doi/abs/10.1198/10618600152627924?casa_token=Kgpt2ZvdMrcAAAAA%3Al5JB7EitQ7MjPmxBObu2XJXocZCE1tI2-ZAceCwBjJOi2TRajF7SaxFcQjo1Nn5AF9q6vKhkxpBd&#.YEZH4F1KiLo).\n",
    "\n",
    "### Thermodynamic Integration\n",
    "\n",
    "This technique is pretty fancy, and is derived from analogies with thermodynamics. It is used for exact calculation of the Bayesian evidence. \n",
    "\n",
    "- Many MCMC chains are launched in parallel. However, their likelihoods differ slightly in their \"temperature\", $T$. We actually use the inverse temperature $\\beta=1/T$ when writing the math. The temperature of a chain is used to modify the likelihood as follows, where \n",
    "\n",
    "$$ p(d|\\theta)_\\beta = p(d|\\theta)^\\beta = p(d|\\theta)^{1/T}.$$\n",
    "\n",
    "- Thus only the $T=0$ chain corresponds to the actual model likelhiood. \n",
    "- But the other temperature chains have the effect of taking higher and higher roots of the likelihood, smoothing it out and making it more similar to the original prior. \n",
    "- These high temperature chains explore the parameter volume well, communicate back to the $T=0$ chain, and improve MCMC mixing. \n",
    "- Crucially, we can also use them in evidence computation:\n",
    "\n",
    "$$ \\ln\\mathcal{Z} = \\int_0^1 \\langle \\ln p(d|\\theta)\\rangle_\\beta\\, d\\beta, $$\n",
    "\n",
    "where we take the mean log-likelihood of each $\\beta$ chain, then sum these over all temperatures to get the log-evidence. Refs [here](https://www.jstor.org/stable/2676756?casa_token=nGU-oa0w6lQAAAAA%3A8UNqj0_rBP7aZ4vocZ_NzbxKhre4WvzhbNOs_TTI_FfexQmppcxQSlxaO5PCHPZNAw9fZUmy4rPeXLe_CVJgAD2AYR6qS8D6xQDicpc1R45ZhjEtOV4&seq=1#metadata_info_tab_contents), and [here](https://link.springer.com/content/pdf/10.1007/BF01406511.pdf).\n",
    "\n",
    "\n",
    "### Nested Sampling\n",
    "\n",
    "Nested sampling is a very different MCMC method from Metropolis-Hastings or even Gibbs approaches. \n",
    "\n",
    "- One starts with many points spread throughout the prior volume.\n",
    "- The points are ranked in terms of their likelihood.\n",
    "- The lowest ranked point is removed, and a higher likelihood point is randomly added.\n",
    "- This is repeated, until the cluster of points climbs up the likelihood surface, accumulating Bayesian evidence information along the way. \n",
    "\n",
    "The prior volume is defined as $dX = p(\\theta)d^n\\theta$ such that $X(\\lambda) = \\int_{p(d|\\theta)>\\lambda}p(\\theta)d^n\\theta$, where the integral is over the model parameter region that exceeds the iso-likelihood contour $p(d|\\theta)=L(\\theta)=\\lambda$.\n",
    "\n",
    "This transforms the evidence integral into \n",
    "\n",
    "$$\\mathcal{Z} = \\int p(d|\\theta)p(\\theta)d^n\\theta = \\int_0^1 L(X)dX.$$\n",
    "\n",
    "There are several popular packages that perform nested sampling, and not only do they all compute the Bayesian evidence but they also produce posterior samples as a by-product. \n",
    "\n",
    "By far the most historically well known is [`MultiNest`](https://github.com/rjw57/MultiNest) and its Python wrapper [`PyMultiNest`](https://johannesbuchner.github.io/PyMultiNest/). It used to be (and may still be) a bit of a pain to install though (there is C and Fortran involved). Refs [here](https://ui.adsabs.harvard.edu/abs/2004AIPC..735..395S/abstract), and [here](https://academic.oup.com/mnras/article/398/4/1601/981502?login=true).\n",
    "\n",
    "\n",
    "So while this is the big one, there seem to be other more pythonic implementations now on the market. Let's look at those.\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dynesty <a class=\"anchor\" id=\"twoa\"></a>\n",
    "\n",
    "https://arxiv.org/abs/1904.02180\n",
    "\n",
    "![](https://github.com/joshspeagle/dynesty/blob/master/docs/images/title.gif?raw=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This example comes from the [dynesty](https://arxiv.org/abs/1904.02180) [\"crash course\"](https://dynesty.readthedocs.io/en/latest/crashcourse.html). We define our likelihood to be a $3$D correlated multivariate Gaussian distribution, and our prior to be uniform in each dimension from $-10$ to $10$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dynesty\n",
    "\n",
    "# Define the dimensionality of our problem.\n",
    "ndim = 3\n",
    "\n",
    "# Define our 3-D correlated multivariate normal likelihood.\n",
    "C = np.identity(ndim)  # set covariance to identity matrix\n",
    "C[C==0] = 0.95  # set off-diagonal terms\n",
    "Cinv = np.linalg.inv(C)  # define the inverse (i.e. the precision matrix)\n",
    "lnorm = -0.5 * (np.log(2 * np.pi) * ndim +\n",
    "                np.log(np.linalg.det(C)))  # ln(normalization)\n",
    "\n",
    "def loglike(x):\n",
    "    \"\"\"The log-likelihood function.\"\"\"\n",
    "\n",
    "    return -0.5 * np.dot(x, np.dot(Cinv, x)) + lnorm\n",
    "\n",
    "# Define our uniform prior.\n",
    "def ptform(u):\n",
    "    \"\"\"Transforms samples `u` drawn from the unit cube to samples to those\n",
    "    from our uniform prior within [-10., 10.) for each variable.\"\"\"\n",
    "\n",
    "    return 10. * (2. * u - 1.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Run the sampler. *(You may get some warnings about messages not being able to write out anymore because the buffer was exceeded. Don't worry. Just check the circle in the top right of your notebook to see when it goes from filled to hollow to show that the cell is finished.)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# \"Static\" nested sampling.\n",
    "sampler = dynesty.NestedSampler(loglike, ptform, ndim)\n",
    "sampler.run_nested()\n",
    "sresults = sampler.results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## \"Dynamic\" nested sampling.\n",
    "# COME BACK and RUN THIS IF YOU HAVE TIME AT THE END. \n",
    "\n",
    "#dsampler = dynesty.DynamicNestedSampler(loglike, ptform, ndim)\n",
    "#dsampler.run_nested()\n",
    "#dresults = dsampler.results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You've just estimated the Bayesian evidence and posterior. Plot the results. These figures will have an $x$-axis that is $-\\ln X$ which is a bit confusing to think about to begin with, but remember that we are moving up the likelihood contours when going from left to right on the graphs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from dynesty import plotting as dyplot\n",
    "\n",
    "# Plot a summary of the run.\n",
    "rfig, raxes = dyplot.runplot(sresults)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot parameter traces and 1-D marginalized posteriors. On the left panels, you'll see that the live points started out scattereed across the prior volume, then move inwards as the sampler climbs the likelihood."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfig, taxes = dyplot.traceplot(sresults)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the 2-D marginalized posteriors using `dynesty` routines. (But you could also use `corner` if you wish)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cfig, caxes = dyplot.cornerplot(sresults)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`dynesty` also lets you access many other utilities and analysis outputs. See below. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from dynesty import utils as dyfunc\n",
    "\n",
    "# Extract sampling results.\n",
    "samples = sresults.samples  # samples\n",
    "weights = np.exp(sresults.logwt - sresults.logz[-1])  # normalized weights\n",
    "\n",
    "# Compute 10%-90% quantiles.\n",
    "quantiles = [dyfunc.quantile(samps, [0.16, 0.84], weights=weights)\n",
    "             for samps in samples.T]\n",
    "print('68% parameter credible regions are:\\n ' + str(quantiles) + '\\n')\n",
    "\n",
    "# Compute weighted mean and covariance.\n",
    "mean, cov = dyfunc.mean_and_cov(samples, weights)\n",
    "print('Mean and covariance of parameters are: ' + str(mean) + '\\n' + str(cov))\n",
    "\n",
    "# Resample weighted samples.\n",
    "samples_equal = dyfunc.resample_equal(samples, weights)\n",
    "\n",
    "# Generate a new set of results with statistical+sampling uncertainties.\n",
    "results_sim = dyfunc.simulate_run(sresults)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you want to plot your posterior samples using `corner` or another package, then access the *equally weighted samples*. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import corner\n",
    "corner.corner(samples_equal);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The code even helpfully tells you what work you should be citing based on the run configurations that you used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(sampler.citations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a lot to explore in the `dynesty` package. It is not a panacea, and there are some things you may have to tune, e.g. the **number of initialized live points**, and **the evidence tolerance**. Dig into these caveats in your own time for your own research problems.\n",
    "\n",
    "Additionally, it implements **dynamic nested sampling** that dynamically allocates live points to explore higher posterior regions better at the expense of increasing the relative evidence uncertainty. See https://dynesty.readthedocs.io/en/latest/dynamic.html. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  UltraNest <a class=\"anchor\" id=\"twob\"></a>\n",
    "\n",
    "*Refs: MLFriends (Buchner, 2014; 2019) using UltraNest https://johannesbuchner.github.io/UltraNest/*\n",
    "\n",
    "To see how `UltraNest` works compared to standard MCMC algorithms, go [here](https://johannesbuchner.github.io/UltraNest/method.html#visualisation). \n",
    "\n",
    "You should have seen this kind of visualization before for Metropolis-Hastings, but this helps make it clear how the new live points are being chosen from within the prior volume, and how the whole set of points are moving up the likelihood surface.\n",
    "\n",
    "We'll look at an example from [UltraNest's documentation](https://johannesbuchner.github.io/UltraNest/using-ultranest.html). This involves fitting a **spectral line feature**. There are many other tutorials on the documentation to dig into in your own time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate data over a wavelength range with a noisy signal\n",
    "x = np.linspace(400, 800, 100)\n",
    "yerr = 1.0\n",
    "y = np.random.normal(20 * np.exp(-0.5 * ((x-500)/4.2)**2), yerr)\n",
    "\n",
    "plt.errorbar(x, y, yerr=yerr, ls=' ', marker='x')\n",
    "plt.plot(x, 20 * np.exp(-0.5 * ((x-500)/4.2)**2), ls=':', alpha=0.5, color='k')\n",
    "plt.xlabel('x')\n",
    "plt.ylabel('y');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This problem has 3 parameters: The location, amplitude and width of the gaussian line. We assume that the noise level is known."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_names = ['location', 'amplitude', 'width']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As in `dynesty`, the parameter space is defined through a transform from unit values ($0,\\ldots,1$) to the physical range.\n",
    "\n",
    "The next function demonstrates:\n",
    "- a uniform prior for a location parameter\n",
    "- a log-uniform prior for a scale parameter\n",
    "- a log-normal prior"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_prior_transform(cube):\n",
    "    params = cube.copy()\n",
    "\n",
    "    # transform location parameter: uniform prior\n",
    "    lo = 400\n",
    "    hi = 800\n",
    "    params[0] = cube[0] * (hi - lo) + lo\n",
    "\n",
    "    # transform amplitude parameter: log-uniform prior\n",
    "    lo = 0.1\n",
    "    hi = 100\n",
    "    params[1] = 10**(cube[1] * (np.log10(hi) - np.log10(lo)) + np.log10(lo))\n",
    "\n",
    "    # More complex prior, you can use the ppf functions\n",
    "    # from scipy.stats, such as scipy.stats.norm(mean, std).ppf\n",
    "\n",
    "    # transform for width:\n",
    "    # a log-normal centered at 1 +- 1dex\n",
    "    params[2] = 10**stats.norm.ppf(cube[2], 0, 1)\n",
    "\n",
    "    return params"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to specify the data likelihood given certain parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_likelihood(params):\n",
    "    location, amplitude, width = params\n",
    "    # compute intensity at every x position according to the model\n",
    "    y_model = amplitude * np.exp(-0.5 * ((x - location)/width)**2)\n",
    "    # compare model and data with gaussian likelihood:\n",
    "    like = -0.5 * (((y_model - y)/yerr)**2).sum()\n",
    "    return like"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ready to sample!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import ultranest\n",
    "\n",
    "sampler = ultranest.ReactiveNestedSampler(param_names, my_likelihood, \n",
    "                                          my_prior_transform)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Take a second to install ipython widgets for some interactive run updates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip3 install ipywidgets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = sampler.run()\n",
    "sampler.print_results()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ultranest.plot import cornerplot\n",
    "cornerplot(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**There is a ton more to explore in nested sampling, evidence evaluation, moddel comparison, and parameter estimation.** \n",
    "\n",
    "**Alas, we must move on. However, these hands-on tutorials using real packages gives you an idea of the rich array of resources you can use for your own research.** \n",
    "\n",
    "**Make sure you know what is going on under the hood though-- remember we're physicists after all!**"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:astr8070] *",
   "language": "python",
   "name": "conda-env-astr8070-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}