Experimenting with ArviZ

Experimenting with ArviZ.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pymc4 as pm4\n",
    "import numpy as np\n",
    "import arviz as az\n",
    "import seaborn as sns\n",
    "import tensorflow as tf\n",
    "import tensorflow_probability as tfp\n",
    "import matplotlib.pyplot as plt\n",
    "from pprint import pprint\n",
    "\n",
    "dtype = tf.float32\n",
    "plt.style.use(\"arviz-darkgrid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "mu = np.zeros(2, dtype=np.float32)\n",
    "cov = np.array([[1, 0.8], [0.8, 1]], dtype=np.float32)\n",
    "data = np.random.multivariate_normal(mu, cov, size=10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "@pm4.model\n",
    "def model():\n",
    "    density = yield pm4.MvNormal(\"density\", loc=mu, covariance_matrix=cov)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow/python/ops/linalg/linear_operator_lower_triangular.py:158: calling LinearOperator.__init__ (from tensorflow.python.ops.linalg.linear_operator) with graph_parents is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Do not pass `graph_parents`.  They will  no longer be used.\n",
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow_probability/python/math/minimize.py:77: calling <lambda> (from tensorflow_probability.python.vi.optimization) with loss is deprecated and will be removed after 2020-07-01.\n",
      "Instructions for updating:\n",
      "The signature for `trace_fn`s passed to `minimize` has changed. Trace functions now take a single `traceable_quantities` argument, which is a `tfp.math.MinimizeTraceableQuantities` namedtuple containing `traceable_quantities.loss`, `traceable_quantities.gradients`, etc. Please update your `trace_fn` definition.\n"
     ]
    }
   ],
   "source": [
    "approx_model = model()\n",
    "mean_field = pm4.fit(approx_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "posterior_samples = mean_field.approximation.sample(5000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/arviz/data/base.py:146: UserWarning: More chains (5000) than draws (2). Passed array should have shape (chains, draws, *shape)\n",
      "  UserWarning,\n"
     ]
    }
   ],
   "source": [
    "trace = az.from_dict(posterior=posterior_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inference data with groups:\n",
      "\t> posterior\n"
     ]
    }
   ],
   "source": [
    "print(trace)"
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}