Add demo notebook on observable patterns

notebooks/observable_patterns.ipynb

@@ -0,0 +1,178 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "99eee62b",
+   "metadata": {},
+   "source": [
+    "# Adding model observables by pattern\n",
+    "\n",
+    "This notebook shows how we can construct model observables using patterns of concepts to include in an observable."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "b660af76",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from mira.metamodel import *\n",
+    "from mira.modeling import Model\n",
+    "from mira.modeling.viz import GraphicalModel"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3badc26e",
+   "metadata": {},
+   "source": [
+    "We start with a standard SIR model using standard identifiers for the three compartments."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "9bc1a327",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "infected = Concept(name='infected population', identifiers={'ido': '0000511'})\n",
+    "susceptible = Concept(name='susceptible population', identifiers={'ido': '0000514'})\n",
+    "immune = Concept(name='immune population', identifiers={'ido': '0000592'})\n",
+    "\n",
+    "t1 = ControlledConversion(\n",
+    "    controller=infected,\n",
+    "    subject=susceptible,\n",
+    "    outcome=infected,\n",
+    ")\n",
+    "t2 = NaturalConversion(subject=infected, outcome=immune)\n",
+    "template_model = TemplateModel(templates=[t1, t2])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "948a2d99",
+   "metadata": {},
+   "source": [
+    "We now stratify the model into two cities."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "89ee6980",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model_2_city = stratify(\n",
+    "    template_model,\n",
+    "    key=\"city\",\n",
+    "    strata=[\n",
+    "        \"geonames:5128581\",  # NYC\n",
+    "        \"geonames:4930956\",  # Boston\n",
+    "    ],\n",
+    ")\n",
+    "\n",
+    "#GraphicalModel.for_jupyter(model_2_city)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "baa72e45",
+   "metadata": {},
+   "source": [
+    "First, let's create an observable for all susceptible people. Note that an alternative to achieving this would have been to add a susceptible observable before stratification. The stratification would then correctly rewrite the observable expression to reflect changes upon stratification."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "c8e84dc8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "add_observable_pattern(model_2_city,\n",
+    "                       'susceptible',\n",
+    "                       identifiers={'ido': '0000511'})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "2acb7d2b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "infected population_geonames_4930956 + infected population_geonames_5128581\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(model_2_city.observables['susceptible'].expression.args[0])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "b72dfc50",
+   "metadata": {},
+   "source": [
+    "We can also refer to a pattern just based on context to identify compartments to constructn an obserable. For instance, we can add an observable for all people in Boston irrespective of disease state, as follows."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "62fa02f0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "add_observable_pattern(model_2_city,\n",
+    "                       'Boston',\n",
+    "                       context={'city': 'geonames:4930956'})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "756e6fdf",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "immune population_geonames_4930956 + infected population_geonames_4930956 + susceptible population_geonames_4930956\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(model_2_city.observables['Boston'].expression.args[0])"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.10.14"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}