blog figures added

stories-inference.ipynb

@@ -0,0 +1,52 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "import numpy as np\n",
+    "\n",
+    "# Define the range for the heatmap\n",
+    "min_score = min(stories[\"log_score\"].min(), stories[\"predictions\"].min())\n",
+    "max_score = max(stories[\"log_score\"].max(), stories[\"predictions\"].max())\n",
+    "\n",
+    "# Create a more continuous 2D histogram\n",
+    "plt.figure(figsize=(12, 10))\n",
+    "sns.kdeplot(\n",
+    "    data=stories.sample(10000, seed=42),\n",
+    "    x=\"log_score\",\n",
+    "    y=\"predictions\",\n",
+    "    cmap=\"YlOrRd\",\n",
+    "    fill=True,\n",
+    "    cbar=True,\n",
+    "    cbar_kws={\"label\": \"Density\"},\n",
+    ")\n",
+    "\n",
+    "plt.title(\"Density Heatmap of Predicted Score vs Actual Score\")\n",
+    "plt.xlabel(\"Actual Score (log_score)\")\n",
+    "plt.ylabel(\"Predicted Score\")\n",
+    "\n",
+    "# Set axis limits to match the data range\n",
+    "plt.xlim(min_score, max_score)\n",
+    "plt.ylim(min_score, max_score)\n",
+    "\n",
+    "# Add a diagonal line for reference\n",
+    "plt.plot([min_score, max_score], [min_score, max_score], \"k--\", alpha=0.5)\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}