--- a
+++ b/src/eval/summary_eval.ipynb
@@ -0,0 +1,157 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "id": "afdeba94",
+   "metadata": {},
+   "source": [
+    "import os\n",
+    "import sys\n",
+    "\n",
+    "src_path = os.path.abspath(\"../..\")\n",
+    "print(src_path)\n",
+    "sys.path.append(src_path)"
+   ],
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "cell_type": "code",
+   "id": "0d5f2e19",
+   "metadata": {},
+   "source": "from src.utils import processed_data_path, set_seed, remote_project_path",
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "cell_type": "code",
+   "id": "e00815d2",
+   "metadata": {},
+   "source": [
+    "set_seed(seed=42)"
+   ],
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "cell_type": "code",
+   "id": "fd92d900",
+   "metadata": {},
+   "source": [
+    "import pandas as pd"
+   ],
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "cell_type": "code",
+   "id": "ef32981d",
+   "metadata": {},
+   "source": "model_path = os.path.join(remote_project_path, \"output\")",
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "cell_type": "code",
+   "id": "42a6306c",
+   "metadata": {},
+   "source": [
+    "import math\n",
+    "\n",
+    "\n",
+    "def ci95(s):\n",
+    "    return 1.96 * s.std() / math.sqrt(len(s))"
+   ],
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "source": [
+    "def print_score(answer_filename, ci=True):\n",
+    "    results = pd.read_json(os.path.join(model_path, f\"{answer_filename}/qa_output/answer_eval.jsonl\"), lines=True)\n",
+    "    print(\"Size:\", len(results))\n",
+    "    results[\"rel_score\"] = (results.score / results.base_score) * 100\n",
+    "\n",
+    "    if not ci:\n",
+    "        print(\"event:\")\n",
+    "        df = results[results.source == \"event\"]\n",
+    "        print(f\"GPT4 score: {df.base_score.mean():.4f} +- {df.base_score.std():.4f}\")\n",
+    "        print(f\"LLM score: {df.score.mean():.4f} +- {df.score.std():.4f}\")\n",
+    "        print(f\"Relative: {df.rel_score.mean():.4f} +- {df.rel_score.std():.4f}\")\n",
+    "\n",
+    "        print(\"note:\")\n",
+    "        df = results[results.source == \"note\"]\n",
+    "        print(f\"GPT4 score: {df.base_score.mean():.4f} +- {df.base_score.std():.4f}\")\n",
+    "        print(f\"LLM score: {df.score.mean():.4f} +- {df.score.std():.4f}\")\n",
+    "        print(f\"Relative: {df.rel_score.mean():.4f} +- {df.rel_score.std():.4f}\")\n",
+    "\n",
+    "        print(\"overall:\")\n",
+    "        print(f\"GPT4 score: {results.base_score.mean():.4f} +- {results.base_score.std():.4f}\")\n",
+    "        print(f\"LLM score: {results.score.mean():.4f} +- {results.score.std():.4f}\")\n",
+    "        print(f\"Relative: {results.rel_score.mean():.4f} +- {results.rel_score.std():.4f}\")\n",
+    "\n",
+    "    else:\n",
+    "        print(\"event:\")\n",
+    "        df = results[results.source == \"event\"]\n",
+    "        print(f\"GPT4 score: {df.base_score.mean():.4f} +- {ci95(df.base_score):.4f}\")\n",
+    "        print(f\"LLM score: {df.score.mean():.4f} +- {ci95(df.score):.4f}\")\n",
+    "        print(f\"Relative: {df.rel_score.mean():.4f} +- {ci95(df.rel_score):.4f}\")\n",
+    "\n",
+    "        print(\"note:\")\n",
+    "        df = results[results.source == \"note\"]\n",
+    "        print(f\"GPT4 score: {df.base_score.mean():.4f} +- {ci95(df.base_score):.4f}\")\n",
+    "        print(f\"LLM score: {df.score.mean():.4f} +- {ci95(df.score):.4f}\")\n",
+    "        print(f\"Relative: {df.rel_score.mean():.4f} +- {ci95(df.rel_score):.4f}\")\n",
+    "\n",
+    "        print(\"overall:\")\n",
+    "        print(f\"GPT4 score: {results.base_score.mean():.4f} +- {ci95(results.base_score):.4f}\")\n",
+    "        print(f\"LLM score: {results.score.mean():.4f} +- {ci95(results.score):.4f}\")\n",
+    "        print(f\"Relative: {results.rel_score.mean():.4f} +- {ci95(results.rel_score):.4f}\")\n",
+    "\n",
+    "    return"
+   ],
+   "id": "fbcddd14",
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "source": "print_score(\"llemr_vicuna\")",
+   "id": "9df914e9",
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "67bd046750285528"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "llm",
+   "language": "python",
+   "name": "llm"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.19"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}