smolagents/examples/benchmark.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.2.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.3.1\u001b[0m\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "!pip install -e .. datasets sympy numpy matplotlib seaborn -q # Install dev version of smolagents + some packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/aymeric/venv/test/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>source</th>\n",
       "      <th>true_answer</th>\n",
       "      <th>true_reasoning</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>If Eliud Kipchoge could maintain his record-ma...</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>17</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>How many studio albums were published by Merce...</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>3</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Here's a fun riddle that I think you'll enjoy....</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>3</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>My family reunion is this week, and I was assi...</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>2</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>In Emily Midkiff's June 2014 article in a jour...</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>fluffy</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>What year was the municipality of San Carlos, ...</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>1786</td>\n",
       "      <td>['https://en.wikipedia.org/wiki/San_Carlos,_An...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>In which year was Maria Elena Walsh named Illu...</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>1985</td>\n",
       "      <td>['https://en.wikipedia.org/wiki/Mar%C3%ADa_Ele...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>What is the durability of the Istarelle spear ...</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>800</td>\n",
       "      <td>['http://demonssouls.wikidot.com/spear', 'http...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>What is the number of the executive order that...</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>7034</td>\n",
       "      <td>['https://www.loc.gov/collections/federal-thea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>Within plus or minus one minute, when was Marq...</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>77</td>\n",
       "      <td>['https://www.fifa.com/fifaplus/en/match-centr...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>132 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              question    source true_answer  \\\n",
       "0    If Eliud Kipchoge could maintain his record-ma...      GAIA          17   \n",
       "1    How many studio albums were published by Merce...      GAIA           3   \n",
       "2    Here's a fun riddle that I think you'll enjoy....      GAIA           3   \n",
       "3    My family reunion is this week, and I was assi...      GAIA           2   \n",
       "4    In Emily Midkiff's June 2014 article in a jour...      GAIA      fluffy   \n",
       "..                                                 ...       ...         ...   \n",
       "127  What year was the municipality of San Carlos, ...  SimpleQA        1786   \n",
       "128  In which year was Maria Elena Walsh named Illu...  SimpleQA        1985   \n",
       "129  What is the durability of the Istarelle spear ...  SimpleQA         800   \n",
       "130  What is the number of the executive order that...  SimpleQA        7034   \n",
       "131  Within plus or minus one minute, when was Marq...  SimpleQA          77   \n",
       "\n",
       "                                        true_reasoning  \n",
       "0                                                 None  \n",
       "1                                                 None  \n",
       "2                                                 None  \n",
       "3                                                 None  \n",
       "4                                                 None  \n",
       "..                                                 ...  \n",
       "127  ['https://en.wikipedia.org/wiki/San_Carlos,_An...  \n",
       "128  ['https://en.wikipedia.org/wiki/Mar%C3%ADa_Ele...  \n",
       "129  ['http://demonssouls.wikidot.com/spear', 'http...  \n",
       "130  ['https://www.loc.gov/collections/federal-thea...  \n",
       "131  ['https://www.fifa.com/fifaplus/en/match-centr...  \n",
       "\n",
       "[132 rows x 4 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import datasets\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "eval_ds = datasets.load_dataset(\"m-ric/smol_agents_benchmark\")[\"test\"]\n",
    "pd.DataFrame(eval_ds)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define utilities and tools\n",
    "To run the SERPAPI tool, you will need to have a [SerpAPI](https://serpapi.com/dashboard) API key: for this you need a paid account."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import os\n",
    "import re\n",
    "import string\n",
    "import time\n",
    "import warnings\n",
    "from typing import List\n",
    "\n",
    "from dotenv import load_dotenv\n",
    "from tqdm import tqdm\n",
    "\n",
    "from smolagents import (\n",
    "    AgentError,\n",
    "    CodeAgent,\n",
    "    GoogleSearchTool,\n",
    "    HfApiModel,\n",
    "    PythonInterpreterTool,\n",
    "    ToolCallingAgent,\n",
    "    VisitWebpageTool,\n",
    ")\n",
    "from smolagents.agents import ActionStep\n",
    "\n",
    "\n",
    "load_dotenv()\n",
    "os.makedirs(\"output\", exist_ok=True)\n",
    "\n",
    "\n",
    "def serialize_agent_error(obj):\n",
    "    if isinstance(obj, AgentError):\n",
    "        return {\"error_type\": obj.__class__.__name__, \"message\": obj.message}\n",
    "    else:\n",
    "        return str(obj)\n",
    "\n",
    "\n",
    "def answer_questions(eval_ds, file_name, agent, model_id, action_type, is_vanilla_llm=False):\n",
    "    answered_questions = []\n",
    "    if os.path.exists(file_name):\n",
    "        with open(file_name, \"r\") as f:\n",
    "            for line in f:\n",
    "                answered_questions.append(json.loads(line)[\"question\"])\n",
    "\n",
    "    for _, example in tqdm(enumerate(eval_ds), total=len(eval_ds)):\n",
    "        try:\n",
    "            question = example[\"question\"]\n",
    "            if example[\"source\"] == \"SimpleQA\":\n",
    "                question += \" Answer with only the final number.\"\n",
    "            if example[\"source\"] == \"MATH\":\n",
    "                question += \" Write code, not latex.\"\n",
    "            if question in answered_questions:\n",
    "                continue\n",
    "            start_time = time.time()\n",
    "\n",
    "            if is_vanilla_llm:\n",
    "                llm = agent\n",
    "                answer = str(llm([{\"role\": \"user\", \"content\": question}]).content)\n",
    "                token_count = {\n",
    "                    \"input\": llm.last_input_token_count,\n",
    "                    \"output\": llm.last_output_token_count,\n",
    "                }\n",
    "                intermediate_steps = str([])\n",
    "            else:\n",
    "                answer = str(agent.run(question))\n",
    "                token_count = agent.monitor.get_total_token_counts()\n",
    "                intermediate_steps = str(agent.logs)\n",
    "                # Remove memory from logs to make them more compact.\n",
    "                for step in agent.logs:\n",
    "                    if isinstance(step, ActionStep):\n",
    "                        step.agent_memory = None\n",
    "\n",
    "            end_time = time.time()\n",
    "            annotated_example = {\n",
    "                \"model_id\": model_id,\n",
    "                \"agent_action_type\": action_type,\n",
    "                \"question\": question,\n",
    "                \"answer\": answer,\n",
    "                \"true_answer\": example[\"true_answer\"],\n",
    "                \"source\": example[\"source\"],\n",
    "                \"intermediate_steps\": intermediate_steps,\n",
    "                \"start_time\": start_time,\n",
    "                \"end_time\": end_time,\n",
    "                \"token_counts\": token_count,\n",
    "            }\n",
    "\n",
    "            with open(file_name, \"a\") as f:\n",
    "                json.dump(annotated_example, f, default=serialize_agent_error)\n",
    "                f.write(\"\\n\")  # add a newline for JSONL format\n",
    "        except Exception as e:\n",
    "            print(\"Failed:\", e)\n",
    "\n",
    "\n",
    "def normalize_number_str(number_str: str) -> float:\n",
    "    # we replace these common units and commas to allow\n",
    "    # conversion to float\n",
    "    for char in [\"$\", \"%\", \",\"]:\n",
    "        number_str = number_str.replace(char, \"\")\n",
    "    try:\n",
    "        return float(number_str)\n",
    "    except ValueError:\n",
    "        return float(\"inf\")\n",
    "\n",
    "\n",
    "def split_string(\n",
    "    s: str,\n",
    "    char_list: list[str] = [\",\", \";\"],\n",
    ") -> list[str]:\n",
    "    pattern = f\"[{''.join(char_list)}]\"\n",
    "    return re.split(pattern, s)\n",
    "\n",
    "\n",
    "def is_float(element: any) -> bool:\n",
    "    try:\n",
    "        float(element)\n",
    "        return True\n",
    "    except ValueError:\n",
    "        return False\n",
    "\n",
    "\n",
    "def normalize_str(input_str, remove_punct=True) -> str:\n",
    "    \"\"\"\n",
    "    Normalize a string by:\n",
    "    - Removing all white spaces\n",
    "    - Optionally removing punctuation (if remove_punct is True)\n",
    "    - Converting to lowercase\n",
    "    Parameters:\n",
    "    - input_str: str, the string to normalize\n",
    "    - remove_punct: bool, whether to remove punctuation (default: True)\n",
    "    Returns:\n",
    "    - str, the normalized string\n",
    "    \"\"\"\n",
    "    # Remove all white spaces. Required e.g for seagull vs. sea gull\n",
    "    no_spaces = re.sub(r\"\\s\", \"\", input_str)\n",
    "\n",
    "    # Remove punctuation, if specified.\n",
    "    if remove_punct:\n",
    "        translator = str.maketrans(\"\", \"\", string.punctuation)\n",
    "        return no_spaces.lower().translate(translator)\n",
    "    else:\n",
    "        return no_spaces.lower()\n",
    "\n",
    "\n",
    "def extract_numbers(text: str) -> List[str]:\n",
    "    \"\"\"This pattern matches:\n",
    "    - Optional negative sign\n",
    "    - Numbers with optional comma thousand separators\n",
    "    - Optional decimal points with decimal numbers\n",
    "    \"\"\"\n",
    "    pattern = r\"-?(?:\\d{1,3}(?:,\\d{3})+|\\d+)(?:\\.\\d+)?\"\n",
    "\n",
    "    return [el.replace(\",\", \"\") for el in re.findall(pattern, text)]\n",
    "\n",
    "\n",
    "def get_question_score_gaia(\n",
    "    model_answer: str,\n",
    "    ground_truth: str,\n",
    ") -> bool:\n",
    "    \"\"\"Scoring function used to score functions from the GAIA benchmark\"\"\"\n",
    "    if is_float(ground_truth):\n",
    "        normalized_answer = normalize_number_str(str(model_answer))\n",
    "        return normalized_answer == float(ground_truth)\n",
    "\n",
    "    elif any(char in ground_truth for char in [\",\", \";\"]):  # if gt is a list\n",
    "        # question with the fish: normalization removes punct\n",
    "        gt_elems = split_string(ground_truth)\n",
    "        ma_elems = split_string(model_answer)\n",
    "\n",
    "        if len(gt_elems) != len(ma_elems):  # check length is the same\n",
    "            warnings.warn(\"Answer lists have different lengths, returning False.\", UserWarning)\n",
    "            return False\n",
    "\n",
    "        comparisons = []\n",
    "        for ma_elem, gt_elem in zip(ma_elems, gt_elems):  # compare each element as float or str\n",
    "            if is_float(gt_elem):\n",
    "                normalized_ma_elem = normalize_number_str(ma_elem)\n",
    "                comparisons.append(normalized_ma_elem == float(gt_elem))\n",
    "            else:\n",
    "                # we do not remove punct since comparisons can include punct\n",
    "                comparisons.append(\n",
    "                    normalize_str(ma_elem, remove_punct=False) == normalize_str(gt_elem, remove_punct=False)\n",
    "                )\n",
    "        return all(comparisons)\n",
    "\n",
    "    else:  # if gt is a str\n",
    "        return normalize_str(model_answer) == normalize_str(ground_truth)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Benchmark agents\n",
    "\n",
    "### Open models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_model_ids = [\n",
    "    \"meta-llama/Llama-3.3-70B-Instruct\",\n",
    "    # \"Qwen/QwQ-32B-Preview\",\n",
    "    \"Qwen/Qwen2.5-72B-Instruct\",\n",
    "    \"Qwen/Qwen2.5-Coder-32B-Instruct\",\n",
    "    \"meta-llama/Llama-3.2-3B-Instruct\",\n",
    "    \"meta-llama/Llama-3.1-8B-Instruct\",\n",
    "    \"mistralai/Mistral-Nemo-Instruct-2407\",\n",
    "    # \"HuggingFaceTB/SmolLM2-1.7B-Instruct\",\n",
    "    # \"meta-llama/Llama-3.1-70B-Instruct\",\n",
    "]\n",
    "\n",
    "for model_id in open_model_ids:\n",
    "    print(f\"Evaluating '{model_id}'...\")\n",
    "    # action_type = \"tool_calling\"\n",
    "    # agent = ToolCallingAgent(\n",
    "    #     tools=[GoogleSearchTool(), VisitWebpageTool(), PythonInterpreterTool()],\n",
    "    #     model=HfApiModel(model_id),\n",
    "    #     max_steps=10,\n",
    "    # )\n",
    "    # file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    # answer_questions(eval_ds, file_name, agent, model_id, action_type)\n",
    "\n",
    "    action_type = \"code\"\n",
    "    agent = CodeAgent(\n",
    "        tools=[GoogleSearchTool(), VisitWebpageTool()],\n",
    "        model=HfApiModel(model_id),\n",
    "        additional_authorized_imports=[\"numpy\", \"sympy\"],\n",
    "        max_steps=10,\n",
    "    )\n",
    "    file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    answer_questions(eval_ds, file_name, agent, model_id, action_type)\n",
    "\n",
    "    # Also evaluate vanilla model\n",
    "    action_type = \"vanilla\"\n",
    "    llm = HfApiModel(model_id)\n",
    "    file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    answer_questions(eval_ds, file_name, llm, model_id, action_type, is_vanilla_llm=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Closed models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from smolagents import LiteLLMModel\n",
    "\n",
    "\n",
    "litellm_model_ids = [\"gpt-4o\", \"anthropic/claude-3-5-sonnet-latest\"]\n",
    "\n",
    "for model_id in litellm_model_ids:\n",
    "    print(f\"Evaluating '{model_id}'...\")\n",
    "    action_type = \"tool_calling\"\n",
    "    agent = ToolCallingAgent(\n",
    "        tools=[\n",
    "            GoogleSearchTool(),\n",
    "            VisitWebpageTool(),\n",
    "            PythonInterpreterTool([\"numpy\", \"sympy\"]),\n",
    "        ],\n",
    "        model=LiteLLMModel(model_id),\n",
    "        max_steps=10,\n",
    "    )\n",
    "    file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    answer_questions(eval_ds, file_name, agent, model_id, action_type)\n",
    "\n",
    "    action_type = \"code\"\n",
    "    agent = CodeAgent(\n",
    "        tools=[GoogleSearchTool(), VisitWebpageTool()],\n",
    "        model=LiteLLMModel(model_id),\n",
    "        additional_authorized_imports=[\"numpy\", \"sympy\"],\n",
    "        max_steps=10,\n",
    "    )\n",
    "    file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    answer_questions(eval_ds, file_name, agent, model_id, action_type)\n",
    "\n",
    "    # Also evaluate vanilla model\n",
    "    action_type = \"vanilla\"\n",
    "    llm = LiteLLMModel(model_id)\n",
    "    file_name = f\"output/{model_id.replace('/', '_')}-{action_type}-26-dec-2024.jsonl\"\n",
    "    answer_questions(eval_ds, file_name, llm, model_id, action_type, is_vanilla_llm=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import glob\n",
    "# import json\n",
    "\n",
    "# jsonl_files = glob.glob(f\"output/*.jsonl\")\n",
    "\n",
    "# for file_path in jsonl_files:\n",
    "#     if \"-Nemo-\" in file_path and \"-vanilla-\" in file_path:\n",
    "#         print(file_path)\n",
    "#         # Read all lines and filter out SimpleQA sources\n",
    "#         filtered_lines = []\n",
    "#         removed = 0\n",
    "#         with open(file_path, \"r\", encoding=\"utf-8\") as f:\n",
    "#             for line in f:\n",
    "#                 try:\n",
    "#                     data = json.loads(line.strip())\n",
    "#                     data[\"answer\"] = data[\"answer\"][\"content\"]\n",
    "#                     # if not any([question in data[\"question\"] for question in eval_ds[\"question\"]]):\n",
    "#                     #     removed +=1\n",
    "#                     # else:\n",
    "#                     filtered_lines.append(json.dumps(data) + \"\\n\")\n",
    "#                 except json.JSONDecodeError:\n",
    "#                     print(\"Invalid line:\", line)\n",
    "#                     continue  # Skip invalid JSON lines\n",
    "#         print(f\"Removed {removed} lines.\")\n",
    "#         # Write filtered content back to the same file\n",
    "#         with open(\n",
    "#             str(file_path).replace(\"-vanilla-\", \"-vanilla2-\"), \"w\", encoding=\"utf-8\"\n",
    "#         ) as f:\n",
    "#             f.writelines(filtered_lines)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Score answers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/6m/9b1tts6d5w960j80wbw9tx3m0000gn/T/ipykernel_27085/1901135017.py:164: UserWarning: Answer lists have different lengths, returning False.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "import glob\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "res = []\n",
    "for file_path in glob.glob(\"output/*.jsonl\"):\n",
    "    data = []\n",
    "    with open(file_path) as f:\n",
    "        for line in f:\n",
    "            try:\n",
    "                # Use standard json module instead of pandas.json to handle large numbers better\n",
    "                record = json.loads(line)\n",
    "                data.append(record)\n",
    "            except json.JSONDecodeError as e:\n",
    "                print(f\"Error parsing line in {file_path}: {e}\")\n",
    "                continue\n",
    "\n",
    "    try:\n",
    "        smoldf = pd.DataFrame(data)\n",
    "        smoldf[\"action_type\"] = \"vanilla\" if \"-vanilla-\" in file_path else \"code\"\n",
    "        res.append(smoldf)\n",
    "    except Exception as e:\n",
    "        print(f\"Error creating DataFrame from {file_path}: {e}\")\n",
    "        continue\n",
    "\n",
    "result_df = pd.concat(res)\n",
    "\n",
    "\n",
    "def get_correct(row):\n",
    "    if row[\"source\"] == \"MATH\":  # Checks the last number in answer\n",
    "        numbers_answer = extract_numbers(str(row[\"answer\"]))\n",
    "        if len(numbers_answer) == 0:\n",
    "            return False\n",
    "        return float(numbers_answer[-1]) == float(row[\"true_answer\"])\n",
    "    else:\n",
    "        return get_question_score_gaia(str(row[\"answer\"]), str(row[\"true_answer\"]))\n",
    "\n",
    "\n",
    "result_df[\"correct\"] = result_df.apply(get_correct, axis=1)\n",
    "\n",
    "result_df = (result_df.groupby([\"model_id\", \"source\", \"action_type\"])[[\"correct\"]].mean() * 100).round(1).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "pivot_df = result_df.pivot_table(\n",
    "    index=[\"model_id\", \"source\"],\n",
    "    columns=[\"action_type\"],\n",
    "    values=\"correct\",\n",
    "    fill_value=float(\"nan\"),\n",
    ").reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Display results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>action_type</th>\n",
       "      <th>model_id</th>\n",
       "      <th>source</th>\n",
       "      <th>code</th>\n",
       "      <th>vanilla</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Qwen/Qwen2.5-72B-Instruct</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>28.1</td>\n",
       "      <td>6.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Qwen/Qwen2.5-72B-Instruct</td>\n",
       "      <td>MATH</td>\n",
       "      <td>76.0</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Qwen/Qwen2.5-72B-Instruct</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>88.0</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Qwen/Qwen2.5-Coder-32B-Instruct</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>25.0</td>\n",
       "      <td>3.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Qwen/Qwen2.5-Coder-32B-Instruct</td>\n",
       "      <td>MATH</td>\n",
       "      <td>86.0</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Qwen/Qwen2.5-Coder-32B-Instruct</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>86.0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>anthropic/claude-3-5-sonnet-latest</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>anthropic/claude-3-5-sonnet-latest</td>\n",
       "      <td>MATH</td>\n",
       "      <td>NaN</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>anthropic/claude-3-5-sonnet-latest</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>NaN</td>\n",
       "      <td>34.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>gpt-4o</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>25.6</td>\n",
       "      <td>3.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>gpt-4o</td>\n",
       "      <td>MATH</td>\n",
       "      <td>58.0</td>\n",
       "      <td>40.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>gpt-4o</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>86.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>meta-llama/Llama-3.1-8B-Instruct</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>3.1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>meta-llama/Llama-3.1-8B-Instruct</td>\n",
       "      <td>MATH</td>\n",
       "      <td>14.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>meta-llama/Llama-3.1-8B-Instruct</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>meta-llama/Llama-3.2-3B-Instruct</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>3.1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>meta-llama/Llama-3.2-3B-Instruct</td>\n",
       "      <td>MATH</td>\n",
       "      <td>40.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>meta-llama/Llama-3.2-3B-Instruct</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>meta-llama/Llama-3.3-70B-Instruct</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>31.2</td>\n",
       "      <td>3.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>meta-llama/Llama-3.3-70B-Instruct</td>\n",
       "      <td>MATH</td>\n",
       "      <td>72.0</td>\n",
       "      <td>40.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>meta-llama/Llama-3.3-70B-Instruct</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>78.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>mistralai/Mistral-Nemo-Instruct-2407</td>\n",
       "      <td>GAIA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>mistralai/Mistral-Nemo-Instruct-2407</td>\n",
       "      <td>MATH</td>\n",
       "      <td>30.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>mistralai/Mistral-Nemo-Instruct-2407</td>\n",
       "      <td>SimpleQA</td>\n",
       "      <td>30.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "action_type                              model_id    source  code  vanilla\n",
       "0                       Qwen/Qwen2.5-72B-Instruct      GAIA  28.1      6.2\n",
       "1                       Qwen/Qwen2.5-72B-Instruct      MATH  76.0     30.0\n",
       "2                       Qwen/Qwen2.5-72B-Instruct  SimpleQA  88.0     10.0\n",
       "3                 Qwen/Qwen2.5-Coder-32B-Instruct      GAIA  25.0      3.1\n",
       "4                 Qwen/Qwen2.5-Coder-32B-Instruct      MATH  86.0     60.0\n",
       "5                 Qwen/Qwen2.5-Coder-32B-Instruct  SimpleQA  86.0      8.0\n",
       "6              anthropic/claude-3-5-sonnet-latest      GAIA   NaN      3.1\n",
       "7              anthropic/claude-3-5-sonnet-latest      MATH   NaN     50.0\n",
       "8              anthropic/claude-3-5-sonnet-latest  SimpleQA   NaN     34.0\n",
       "9                                          gpt-4o      GAIA  25.6      3.1\n",
       "10                                         gpt-4o      MATH  58.0     40.0\n",
       "11                                         gpt-4o  SimpleQA  86.0      6.0\n",
       "12               meta-llama/Llama-3.1-8B-Instruct      GAIA   3.1      0.0\n",
       "13               meta-llama/Llama-3.1-8B-Instruct      MATH  14.0     18.0\n",
       "14               meta-llama/Llama-3.1-8B-Instruct  SimpleQA   2.0      6.0\n",
       "15               meta-llama/Llama-3.2-3B-Instruct      GAIA   3.1      0.0\n",
       "16               meta-llama/Llama-3.2-3B-Instruct      MATH  40.0     12.0\n",
       "17               meta-llama/Llama-3.2-3B-Instruct  SimpleQA  20.0      0.0\n",
       "18              meta-llama/Llama-3.3-70B-Instruct      GAIA  31.2      3.1\n",
       "19              meta-llama/Llama-3.3-70B-Instruct      MATH  72.0     40.0\n",
       "20              meta-llama/Llama-3.3-70B-Instruct  SimpleQA  78.0     12.0\n",
       "21           mistralai/Mistral-Nemo-Instruct-2407      GAIA   0.0      3.1\n",
       "22           mistralai/Mistral-Nemo-Instruct-2407      MATH  30.0     22.0\n",
       "23           mistralai/Mistral-Nemo-Instruct-2407  SimpleQA  30.0      6.0"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(pivot_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mnotebook controller is DISPOSED. \n",
      "\u001b[1;31mView Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib.legend_handler import HandlerTuple  # Added import\n",
    "\n",
    "\n",
    "# Assuming pivot_df is your original dataframe\n",
    "models = pivot_df[\"model_id\"].unique()\n",
    "sources = pivot_df[\"source\"].unique()\n",
    "\n",
    "# Create figure and axis\n",
    "plt.style.use(\"seaborn-v0_8-white\")\n",
    "fig, ax = plt.subplots(figsize=(15, 6))\n",
    "\n",
    "# Set the width of each bar group and positions of the bars\n",
    "width = 0.15  # width of each bar\n",
    "spacing = 0.02  # space between bars within a group\n",
    "group_spacing = 0.2  # space between model groups\n",
    "\n",
    "# Calculate positions for the bars\n",
    "num_sources = len(sources)\n",
    "total_width_per_group = (width + spacing) * num_sources * 2  # *2 for agent and vanilla\n",
    "x = np.arange(len(models)) * (total_width_per_group + group_spacing)\n",
    "\n",
    "# Plot bars for each source\n",
    "for i, source in enumerate(sources):\n",
    "    source_data = pivot_df[pivot_df[\"source\"] == source]\n",
    "    agent_scores = [\n",
    "        source_data[source_data[\"model_id\"] == model][\"code\"].values[0]\n",
    "        if len(source_data[source_data[\"model_id\"] == model]) > 0\n",
    "        else np.nan\n",
    "        for model in models\n",
    "    ]\n",
    "    vanilla_scores = [\n",
    "        source_data[source_data[\"model_id\"] == model][\"vanilla\"].values[0]\n",
    "        if len(source_data[source_data[\"model_id\"] == model]) > 0\n",
    "        else np.nan\n",
    "        for model in models\n",
    "    ]\n",
    "\n",
    "    # Position calculation for each pair of bars\n",
    "    pos = x + i * (width * 2 + spacing)\n",
    "\n",
    "    agent_bars = ax.bar(pos, agent_scores, width, label=f\"{source} (Agent)\", alpha=0.8)\n",
    "    vanilla_bars = ax.bar(\n",
    "        pos + width * 0.6,\n",
    "        vanilla_scores,\n",
    "        width,\n",
    "        hatch=\"////\",\n",
    "        alpha=0.5,\n",
    "        hatch_linewidth=2,\n",
    "        label=f\"{source} (Vanilla)\",\n",
    "        color=\"white\",\n",
    "        edgecolor=agent_bars[0].get_facecolor(),\n",
    "    )\n",
    "\n",
    "# Customize the plot\n",
    "ax.set_ylabel(\"Score\")\n",
    "ax.set_title(\"Model Performance Comparison\")\n",
    "\n",
    "# Set x-axis ticks in the middle of each group\n",
    "group_centers = x + (total_width_per_group - spacing) / 2\n",
    "ax.set_xticks(group_centers)\n",
    "\n",
    "# Wrap long model names to prevent overlap\n",
    "wrapped_labels = [\"\\n\".join(model.split(\"/\")) for model in models]\n",
    "ax.set_xticklabels(wrapped_labels, rotation=0, ha=\"center\")\n",
    "\n",
    "# Modify legend to combine agent and vanilla entries\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "unique_sources = sources\n",
    "legend_elements = [\n",
    "    (handles[i * 2], handles[i * 2 + 1], labels[i * 2].replace(\" (Agent)\", \"\")) for i in range(len(unique_sources))\n",
    "]\n",
    "custom_legend = ax.legend(\n",
    "    [(agent_handle, vanilla_handle) for agent_handle, vanilla_handle, _ in legend_elements],\n",
    "    [label for _, _, label in legend_elements],\n",
    "    handler_map={tuple: HandlerTuple(ndivide=None)},\n",
    "    bbox_to_anchor=(1.05, 1),\n",
    "    loc=\"upper left\",\n",
    ")\n",
    "\n",
    "ax.yaxis.grid(True, linestyle=\"--\", alpha=0.3)\n",
    "ax.set_ylim(bottom=0)\n",
    "plt.tight_layout()\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'formatted_df' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[12], line 45\u001b[0m\n\u001b[1;32m     41\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m mathjax_table\n\u001b[1;32m     44\u001b[0m \u001b[38;5;66;03m# Usage (after running your previous data processing code):\u001b[39;00m\n\u001b[0;32m---> 45\u001b[0m mathjax_table \u001b[38;5;241m=\u001b[39m create_mathjax_table(pivot_df, \u001b[43mformatted_df\u001b[49m)\n\u001b[1;32m     46\u001b[0m \u001b[38;5;28mprint\u001b[39m(mathjax_table)\n",
      "\u001b[0;31mNameError\u001b[0m: name 'formatted_df' is not defined"
     ]
    }
   ],
   "source": [
    "def create_mathjax_table(pivot_df, formatted_df):\n",
    "    # Start the matrix environment with 4 columns\n",
    "    # l for left-aligned model and task, c for centered numbers\n",
    "    mathjax_table = \"\\\\begin{array}{llcc}\\n\"\n",
    "    mathjax_table += \"\\\\text{Model} & \\\\text{Task} & \\\\text{Agent} & \\\\text{Vanilla} \\\\\\\\\\n\"\n",
    "    mathjax_table += \"\\\\hline\\n\"\n",
    "\n",
    "    # Sort the DataFrame by model_id and source\n",
    "    formatted_df = formatted_df.sort_values([\"model_id\", \"source\"])\n",
    "\n",
    "    current_model = None\n",
    "    for _, row in formatted_df.iterrows():\n",
    "        model = row[\"model_id\"]\n",
    "        source = row[\"source\"]\n",
    "\n",
    "        # Add a horizontal line between different models\n",
    "        if current_model is not None and current_model != model:\n",
    "            mathjax_table += \"\\\\hline\\n\"\n",
    "\n",
    "        # Format model name\n",
    "        model_display = model.replace(\"_\", \"\\\\_\")\n",
    "        if \"Qwen\" in model or \"anthropic\" in model:\n",
    "            model_display = f\"\\\\textit{{{model_display}}}\"\n",
    "\n",
    "        # If it's the same model as previous row, use empty space\n",
    "        if current_model == model:\n",
    "            model_display = \"\\\\;\"\n",
    "\n",
    "        # Add the data row\n",
    "        mathjax_table += f\"{model_display} & {source} & {row['agent']} & {row['vanilla']} \\\\\\\\\\n\"\n",
    "\n",
    "        current_model = model\n",
    "\n",
    "    mathjax_table += \"\\\\hline\\n\"\n",
    "    mathjax_table += \"\\\\end{array}\"\n",
    "\n",
    "    return mathjax_table\n",
    "\n",
    "\n",
    "# Usage (after running your previous data processing code):\n",
    "# mathjax_table = create_mathjax_table(pivot_df, formatted_df)\n",
    "# print(mathjax_table)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "test",
   "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": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}