Merge pull request #1 from quanted/gemini-models

Gemini models

notebooks/.ipynb_checkpoints/gcolab-epa_nmf_py_workflow_01-checkpoint.ipynb

@@ -487,7 +487,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "5afced49-c9ff-4a44-9558-e327d07d46bc",
+   "id": "d8514eed-0747-49b0-b387-ce3379213a2e",
    "metadata": {
     "id": "5afced49-c9ff-4a44-9558-e327d07d46bc",
     "outputId": "499ec5ca-8d55-4537-a069-a7b004821e1c"
@@ -498,6 +498,35 @@
     "model_analysis.plot_factor_contributions(feature_idx=1)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3e479211-fc1d-46d8-a84a-13e339143c1c",
+   "metadata": {
+    "id": "5afced49-c9ff-4a44-9558-e327d07d46bc",
+    "outputId": "499ec5ca-8d55-4537-a069-a7b004821e1c"
+   },
+   "outputs": [],
+   "source": [
+    "# New Graphic: Factor Profile Composition Radar Graph\n",
+    "model_analysis.plot_factor_composition()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "be971709-def7-4528-85f2-b5a1dd7e8506",
+   "metadata": {
+    "id": "5afced49-c9ff-4a44-9558-e327d07d46bc",
+    "outputId": "499ec5ca-8d55-4537-a069-a7b004821e1c"
+   },
+   "outputs": [],
+   "source": [
+    "# New Graphic: Factor Contribution Surface Plot\n",
+    "factor_idx = 1\n",
+    "model_analysis.plot_factor_surface(factor_idx=factor_idx)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "3ccba89f-ec29-4791-b221-ecfbd4f15009",

notebooks/.ipynb_checkpoints/gemini-analysis-checkpoint.ipynb

@@ -0,0 +1,95 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "2ce7bb70-e4ee-4e27-a8c9-9ce1ab0b5fc0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Notebook imports\n",
+    "import os\n",
+    "import sys\n",
+    "import json\n",
+    "\n",
+    "module_path = os.path.abspath(os.path.join('..'))\n",
+    "if module_path not in sys.path:\n",
+    "    sys.path.append(module_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "bccedc41-222f-46b6-9461-e64a35cbeb7e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "LS_NMF = \"D:\\\\\\\\projects\\\\nmf_py\\\\funsearch\\\\algorithms\\\\ls-nmf\\\\summary.json\"\n",
+    "WS_NMF = \"D:\\\\\\\\projects\\\\nmf_py\\\\funsearch\\\\algorithms\\\\ws-nmf\\\\summary.json\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "44131ca7-2cb7-40d9-ba1c-0097708e202f",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Best Model: 267, Q(robust): 54418.75, Base Q(robust): 54424.27\n"
+     ]
+    }
+   ],
+   "source": [
+    "best_Q = None\n",
+    "best_model = None\n",
+    "base_Q = None\n",
+    "with open(LS_NMF, 'r') as gemini_file:\n",
+    "    model_summary = json.load(gemini_file)\n",
+    "    for k, v in model_summary.items():\n",
+    "        k_Q = v[\"Q(robust)\"][0]\n",
+    "        if best_Q is None:\n",
+    "            best_Q = k_Q\n",
+    "            best_model = k\n",
+    "        elif k_Q < best_Q:\n",
+    "            best_Q = k_Q\n",
+    "            best_model = k\n",
+    "        if k == '0':\n",
+    "            base_Q = k_Q\n",
+    "\n",
+    "print(f\"Best Model: {best_model}, Q(robust): {best_Q}, Base Q(robust): {base_Q}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fe5ac9bc-3767-4529-90ce-e2e5e6f4eb1a",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "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": "3.11.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}