Merge remote-tracking branch 'upstream/Development' into Development

Author: elseml

docs/source/tutorial_notebooks/Covid19_Initial_Posterior_Estimation.ipynb

@@ -1,5 +1,13 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "01fcb38f",
+   "metadata": {},
+   "source": [
+    "# Posterior Estimation for SIR-like Models"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "thirty-canada",
@@ -8,7 +16,7 @@
    },
    "source": [
     "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
-    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Defining-the-Generative-Model\" data-toc-modified-id=\"Defining-the-Generative-Model-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Defining the Generative Model</a></span><ul class=\"toc-item\"><li><span><a href=\"#Prior\" data-toc-modified-id=\"Prior-1.1\"><span class=\"toc-item-num\">1.1&nbsp;&nbsp;</span>Prior</a></span></li><li><span><a href=\"#Simulator-(Implicit-Likelihood-Function)\" data-toc-modified-id=\"Simulator-(Implicit-Likelihood-Function)-1.2\"><span class=\"toc-item-num\">1.2&nbsp;&nbsp;</span>Simulator (Implicit Likelihood Function)</a></span></li><li><span><a href=\"#Loading-Real-Data\" data-toc-modified-id=\"Loading-Real-Data-1.3\"><span class=\"toc-item-num\">1.3&nbsp;&nbsp;</span>Loading Real Data</a></span></li><li><span><a href=\"#Generative-Model\" data-toc-modified-id=\"Generative-Model-1.4\"><span class=\"toc-item-num\">1.4&nbsp;&nbsp;</span>Generative Model</a></span></li></ul></li><li><span><a href=\"#Prior-Checking\" data-toc-modified-id=\"Prior-Checking-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Prior Checking</a></span></li><li><span><a href=\"#Defining-the-Neural-Approximator\" data-toc-modified-id=\"Defining-the-Neural-Approximator-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Defining the Neural Approximator</a></span><ul class=\"toc-item\"><li><span><a href=\"#Summary-Network\" data-toc-modified-id=\"Summary-Network-3.1\"><span class=\"toc-item-num\">3.1&nbsp;&nbsp;</span>Summary Network</a></span></li><li><span><a href=\"#Inference-Network\" data-toc-modified-id=\"Inference-Network-3.2\"><span class=\"toc-item-num\">3.2&nbsp;&nbsp;</span>Inference Network</a></span></li><li><span><a href=\"#Amortized-Posterior\" data-toc-modified-id=\"Amortized-Posterior-3.3\"><span class=\"toc-item-num\">3.3&nbsp;&nbsp;</span>Amortized Posterior</a></span></li></ul></li><li><span><a href=\"#Defining-the-Configurator\" data-toc-modified-id=\"Defining-the-Configurator-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Defining the Configurator</a></span></li><li><span><a href=\"#Defining-the-Trainer\" data-toc-modified-id=\"Defining-the-Trainer-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>Defining the Trainer</a></span></li><li><span><a href=\"#Training-Phase\" data-toc-modified-id=\"Training-Phase-6\"><span class=\"toc-item-num\">6&nbsp;&nbsp;</span>Training Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Inspecting-the-Loss\" data-toc-modified-id=\"Inspecting-the-Loss-6.1\"><span class=\"toc-item-num\">6.1&nbsp;&nbsp;</span>Inspecting the Loss</a></span></li></ul></li><li><span><a href=\"#Validation-Phase\" data-toc-modified-id=\"Validation-Phase-7\"><span class=\"toc-item-num\">7&nbsp;&nbsp;</span>Validation Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Inspecting-the-Latent-Space\" data-toc-modified-id=\"Inspecting-the-Latent-Space-7.1\"><span class=\"toc-item-num\">7.1&nbsp;&nbsp;</span>Inspecting the Latent Space</a></span></li><li><span><a href=\"#Simulation-Based-Calibration---Rank-Histograms\" data-toc-modified-id=\"Simulation-Based-Calibration---Rank-Histograms-7.2\"><span class=\"toc-item-num\">7.2&nbsp;&nbsp;</span>Simulation-Based Calibration - Rank Histograms</a></span></li><li><span><a href=\"#Simulation-Based-Calibration---Rank-ECDF\" data-toc-modified-id=\"Simulation-Based-Calibration---Rank-ECDF-7.3\"><span class=\"toc-item-num\">7.3&nbsp;&nbsp;</span>Simulation-Based Calibration - Rank ECDF</a></span></li><li><span><a href=\"#Inferential-Adequacy-(Global)\" data-toc-modified-id=\"Inferential-Adequacy-(Global)-7.4\"><span class=\"toc-item-num\">7.4&nbsp;&nbsp;</span>Inferential Adequacy (Global)</a></span></li></ul></li><li><span><a href=\"#Inference-Phase\" data-toc-modified-id=\"Inference-Phase-8\"><span class=\"toc-item-num\">8&nbsp;&nbsp;</span>Inference Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Bivariate-Posteriors\" data-toc-modified-id=\"Bivariate-Posteriors-8.1\"><span class=\"toc-item-num\">8.1&nbsp;&nbsp;</span>Bivariate Posteriors</a></span><ul class=\"toc-item\"><li><span><a href=\"#Standalone\" data-toc-modified-id=\"Standalone-8.1.1\"><span class=\"toc-item-num\">8.1.1&nbsp;&nbsp;</span>Standalone</a></span></li><li><span><a href=\"#Compared-to-Prior\" data-toc-modified-id=\"Compared-to-Prior-8.1.2\"><span class=\"toc-item-num\">8.1.2&nbsp;&nbsp;</span>Compared to Prior</a></span></li></ul></li><li><span><a href=\"#Posterior-Retrodictive-Checks\" data-toc-modified-id=\"Posterior-Retrodictive-Checks-8.2\"><span class=\"toc-item-num\">8.2&nbsp;&nbsp;</span>Posterior Retrodictive Checks</a></span></li></ul></li></ul></div>"
+    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Introduction\" data-toc-modified-id=\"Introduction-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Introduction</a></span></li><li><span><a href=\"#Defining-the-Generative-Model\" data-toc-modified-id=\"Defining-the-Generative-Model-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Defining the Generative Model</a></span><ul class=\"toc-item\"><li><span><a href=\"#Prior\" data-toc-modified-id=\"Prior-2.1\"><span class=\"toc-item-num\">2.1&nbsp;&nbsp;</span>Prior</a></span></li><li><span><a href=\"#Simulator-(Implicit-Likelihood-Function)\" data-toc-modified-id=\"Simulator-(Implicit-Likelihood-Function)-2.2\"><span class=\"toc-item-num\">2.2&nbsp;&nbsp;</span>Simulator (Implicit Likelihood Function)</a></span></li><li><span><a href=\"#Loading-Real-Data\" data-toc-modified-id=\"Loading-Real-Data-2.3\"><span class=\"toc-item-num\">2.3&nbsp;&nbsp;</span>Loading Real Data</a></span></li><li><span><a href=\"#Generative-Model\" data-toc-modified-id=\"Generative-Model-2.4\"><span class=\"toc-item-num\">2.4&nbsp;&nbsp;</span>Generative Model</a></span></li></ul></li><li><span><a href=\"#Prior-Checking\" data-toc-modified-id=\"Prior-Checking-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Prior Checking</a></span></li><li><span><a href=\"#Defining-the-Neural-Approximator\" data-toc-modified-id=\"Defining-the-Neural-Approximator-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Defining the Neural Approximator</a></span><ul class=\"toc-item\"><li><span><a href=\"#Summary-Network\" data-toc-modified-id=\"Summary-Network-4.1\"><span class=\"toc-item-num\">4.1&nbsp;&nbsp;</span>Summary Network</a></span></li><li><span><a href=\"#Inference-Network\" data-toc-modified-id=\"Inference-Network-4.2\"><span class=\"toc-item-num\">4.2&nbsp;&nbsp;</span>Inference Network</a></span></li><li><span><a href=\"#Amortized-Posterior\" data-toc-modified-id=\"Amortized-Posterior-4.3\"><span class=\"toc-item-num\">4.3&nbsp;&nbsp;</span>Amortized Posterior</a></span></li></ul></li><li><span><a href=\"#Defining-the-Configurator\" data-toc-modified-id=\"Defining-the-Configurator-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>Defining the Configurator</a></span></li><li><span><a href=\"#Defining-the-Trainer\" data-toc-modified-id=\"Defining-the-Trainer-6\"><span class=\"toc-item-num\">6&nbsp;&nbsp;</span>Defining the Trainer</a></span></li><li><span><a href=\"#Training-Phase\" data-toc-modified-id=\"Training-Phase-7\"><span class=\"toc-item-num\">7&nbsp;&nbsp;</span>Training Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Inspecting-the-Loss\" data-toc-modified-id=\"Inspecting-the-Loss-7.1\"><span class=\"toc-item-num\">7.1&nbsp;&nbsp;</span>Inspecting the Loss</a></span></li></ul></li><li><span><a href=\"#Validation-Phase\" data-toc-modified-id=\"Validation-Phase-8\"><span class=\"toc-item-num\">8&nbsp;&nbsp;</span>Validation Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Inspecting-the-Latent-Space\" data-toc-modified-id=\"Inspecting-the-Latent-Space-8.1\"><span class=\"toc-item-num\">8.1&nbsp;&nbsp;</span>Inspecting the Latent Space</a></span></li><li><span><a href=\"#Simulation-Based-Calibration---Rank-Histograms\" data-toc-modified-id=\"Simulation-Based-Calibration---Rank-Histograms-8.2\"><span class=\"toc-item-num\">8.2&nbsp;&nbsp;</span>Simulation-Based Calibration - Rank Histograms</a></span></li><li><span><a href=\"#Simulation-Based-Calibration---Rank-ECDF\" data-toc-modified-id=\"Simulation-Based-Calibration---Rank-ECDF-8.3\"><span class=\"toc-item-num\">8.3&nbsp;&nbsp;</span>Simulation-Based Calibration - Rank ECDF</a></span></li><li><span><a href=\"#Inferential-Adequacy-(Global)\" data-toc-modified-id=\"Inferential-Adequacy-(Global)-8.4\"><span class=\"toc-item-num\">8.4&nbsp;&nbsp;</span>Inferential Adequacy (Global)</a></span></li></ul></li><li><span><a href=\"#Inference-Phase\" data-toc-modified-id=\"Inference-Phase-9\"><span class=\"toc-item-num\">9&nbsp;&nbsp;</span>Inference Phase</a></span><ul class=\"toc-item\"><li><span><a href=\"#Bivariate-Posteriors\" data-toc-modified-id=\"Bivariate-Posteriors-9.1\"><span class=\"toc-item-num\">9.1&nbsp;&nbsp;</span>Bivariate Posteriors</a></span><ul class=\"toc-item\"><li><span><a href=\"#Standalone\" data-toc-modified-id=\"Standalone-9.1.1\"><span class=\"toc-item-num\">9.1.1&nbsp;&nbsp;</span>Standalone</a></span></li><li><span><a href=\"#Compared-to-Prior\" data-toc-modified-id=\"Compared-to-Prior-9.1.2\"><span class=\"toc-item-num\">9.1.2&nbsp;&nbsp;</span>Compared to Prior</a></span></li></ul></li><li><span><a href=\"#Posterior-Retrodictive-Checks\" data-toc-modified-id=\"Posterior-Retrodictive-Checks-9.2\"><span class=\"toc-item-num\">9.2&nbsp;&nbsp;</span>Posterior Retrodictive Checks</a></span></li></ul></li></ul></div>"
    ]
   },
   {
@@ -18,10 +26,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import os\n",
-    "import sys\n",
-    "\n",
-    "sys.path.append(os.path.abspath(os.path.join(\"../../..\")))\n",
     "import datetime\n",
     "from functools import partial\n",
     "\n",
@@ -40,7 +44,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "C:\\Users\\Stefan Radev\\Desktop\\Projects\\BayesFlow\\bayesflow\\trainers.py:26: TqdmExperimentalWarning: Using `tqdm.autonotebook.tqdm` in notebook mode. Use `tqdm.tqdm` instead to force console mode (e.g. in jupyter console)\n",
+      "D:\\Anaconda3\\envs\\TensorFlowDev\\lib\\site-packages\\bayesflow\\trainers.py:26: TqdmExperimentalWarning: Using `tqdm.autonotebook.tqdm` in notebook mode. Use `tqdm.tqdm` instead to force console mode (e.g. in jupyter console)\n",
       "  from tqdm.autonotebook import tqdm\n"
      ]
     }
@@ -58,8 +62,8 @@
    "id": "polished-warning",
    "metadata": {},
    "source": [
-    "<h1>Introduction</h1>\n",
-    "<br>\n",
+    "## Introduction\n",
+    "\n",
     "In this tutorial, we will illustrate how to perform posterior inference on simple, stationary SIR-like models (complex models will be tackled in a further notebook). SIR-like models comprise suitable illustrative examples, since they generate time-series and their outputs represent the results of solving a system of ordinary differential equations (ODEs).\n",
     "\n",
     "The details for tackling stochastic epidemiological models are described in our corresponding paper, which you can consult for a more formal exposition and a more comprehensive treatment of neural architectures:\n",
@@ -1614,7 +1618,7 @@
    "source": [
     "That's it for this tutorial! You now know how to use the basic building blocks of `BayesFlow` to create amortized neural approximators. :)\n",
     "\n",
-    "In the [next tutorial](./PriorSensitivity_Covid19_Initial.ipynb), we will go through a <strong>prior sensitivity analysis</strong> with `BayesFlow`, which is as easy to perform as it is important for ascertaining the robustness of our inferences."
+    "<!-- In the [next tutorial](./PriorSensitivity_Covid19_Initial.ipynb), we will go through a <strong>prior sensitivity analysis</strong> with `BayesFlow`, which is as easy to perform as it is important for ascertaining the robustness of our inferences. -->"
    ]
   }
  ],