notebooks/LM_Eval_Demonstration.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "MCiLSwoWQK7z",
        "outputId": "5efbb6bc-0e2d-4df7-a5b2-36f4448960a8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting git+https://github.com/EleutherAI/lm-evaluation-harness.git\n",
            "  Cloning https://github.com/EleutherAI/lm-evaluation-harness.git to /tmp/pip-req-build-j2xmmhxh\n",
            "  Running command git clone --filter=blob:none --quiet https://github.com/EleutherAI/lm-evaluation-harness.git /tmp/pip-req-build-j2xmmhxh\n",
            "  Resolved https://github.com/EleutherAI/lm-evaluation-harness.git to commit b4cd85d406938f94ee5d451840a0d69bbda27006\n",
            "  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "  Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: accelerate>=0.21.0 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.30.1)\n",
            "Requirement already satisfied: evaluate in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.4.2)\n",
            "Requirement already satisfied: datasets>=2.16.0 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.19.1)\n",
            "Requirement already satisfied: jsonlines in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (4.0.0)\n",
            "Requirement already satisfied: numexpr in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.10.0)\n",
            "Requirement already satisfied: peft>=0.2.0 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.11.1)\n",
            "Requirement already satisfied: pybind11>=2.6.2 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.12.0)\n",
            "Requirement already satisfied: pytablewriter in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (1.2.0)\n",
            "Requirement already satisfied: rouge-score>=0.0.4 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.1.2)\n",
            "Requirement already satisfied: sacrebleu>=1.5.0 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.4.2)\n",
            "Requirement already satisfied: scikit-learn>=0.24.1 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (1.2.2)\n",
            "Requirement already satisfied: sqlitedict in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.1.0)\n",
            "Requirement already satisfied: torch>=1.8 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (2.3.0+cu121)\n",
            "Requirement already satisfied: tqdm-multiprocess in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.0.11)\n",
            "Requirement already satisfied: transformers>=4.1 in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (4.41.1)\n",
            "Requirement already satisfied: zstandard in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.22.0)\n",
            "Requirement already satisfied: dill in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (0.3.8)\n",
            "Requirement already satisfied: word2number in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (1.1)\n",
            "Requirement already satisfied: more-itertools in /usr/local/lib/python3.10/dist-packages (from lm_eval==0.4.2) (10.1.0)\n",
            "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (1.25.2)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (24.0)\n",
            "Requirement already satisfied: psutil in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (5.9.5)\n",
            "Requirement already satisfied: pyyaml in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (6.0.1)\n",
            "Requirement already satisfied: huggingface-hub in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (0.23.1)\n",
            "Requirement already satisfied: safetensors>=0.3.1 in /usr/local/lib/python3.10/dist-packages (from accelerate>=0.21.0->lm_eval==0.4.2) (0.4.3)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (3.14.0)\n",
            "Requirement already satisfied: pyarrow>=12.0.0 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (14.0.2)\n",
            "Requirement already satisfied: pyarrow-hotfix in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (0.6)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (2.0.3)\n",
            "Requirement already satisfied: requests>=2.19.0 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (2.31.0)\n",
            "Requirement already satisfied: tqdm>=4.62.1 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (4.66.4)\n",
            "Requirement already satisfied: xxhash in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (3.4.1)\n",
            "Requirement already satisfied: multiprocess in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (0.70.16)\n",
            "Requirement already satisfied: fsspec[http]<=2024.3.1,>=2023.1.0 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (2023.6.0)\n",
            "Requirement already satisfied: aiohttp in /usr/local/lib/python3.10/dist-packages (from datasets>=2.16.0->lm_eval==0.4.2) (3.9.5)\n",
            "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from rouge-score>=0.0.4->lm_eval==0.4.2) (1.4.0)\n",
            "Requirement already satisfied: nltk in /usr/local/lib/python3.10/dist-packages (from rouge-score>=0.0.4->lm_eval==0.4.2) (3.8.1)\n",
            "Requirement already satisfied: six>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from rouge-score>=0.0.4->lm_eval==0.4.2) (1.16.0)\n",
            "Requirement already satisfied: portalocker in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.5.0->lm_eval==0.4.2) (2.8.2)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.5.0->lm_eval==0.4.2) (2024.5.15)\n",
            "Requirement already satisfied: tabulate>=0.8.9 in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.5.0->lm_eval==0.4.2) (0.9.0)\n",
            "Requirement already satisfied: colorama in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.5.0->lm_eval==0.4.2) (0.4.6)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.5.0->lm_eval==0.4.2) (4.9.4)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.24.1->lm_eval==0.4.2) (1.11.4)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.24.1->lm_eval==0.4.2) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.24.1->lm_eval==0.4.2) (3.5.0)\n",
            "Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (4.11.0)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (1.12)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (3.3)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (3.1.4)\n",
            "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (8.9.2.26)\n",
            "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.3.1)\n",
            "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (11.0.2.54)\n",
            "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (10.3.2.106)\n",
            "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (11.4.5.107)\n",
            "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.0.106)\n",
            "Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (2.20.5)\n",
            "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (12.1.105)\n",
            "Requirement already satisfied: triton==2.3.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.8->lm_eval==0.4.2) (2.3.0)\n",
            "Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.10/dist-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>=1.8->lm_eval==0.4.2) (12.5.40)\n",
            "Requirement already satisfied: tokenizers<0.20,>=0.19 in /usr/local/lib/python3.10/dist-packages (from transformers>=4.1->lm_eval==0.4.2) (0.19.1)\n",
            "Requirement already satisfied: attrs>=19.2.0 in /usr/local/lib/python3.10/dist-packages (from jsonlines->lm_eval==0.4.2) (23.2.0)\n",
            "Requirement already satisfied: setuptools>=38.3.0 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (67.7.2)\n",
            "Requirement already satisfied: DataProperty<2,>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (1.0.1)\n",
            "Requirement already satisfied: mbstrdecoder<2,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (1.1.3)\n",
            "Requirement already satisfied: pathvalidate<4,>=2.3.0 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (3.2.0)\n",
            "Requirement already satisfied: tabledata<2,>=1.3.1 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (1.3.3)\n",
            "Requirement already satisfied: tcolorpy<1,>=0.0.5 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (0.1.6)\n",
            "Requirement already satisfied: typepy[datetime]<2,>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from pytablewriter->lm_eval==0.4.2) (1.3.2)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.16.0->lm_eval==0.4.2) (1.3.1)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.16.0->lm_eval==0.4.2) (1.4.1)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.16.0->lm_eval==0.4.2) (6.0.5)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.16.0->lm_eval==0.4.2) (1.9.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.16.0->lm_eval==0.4.2) (4.0.3)\n",
            "Requirement already satisfied: chardet<6,>=3.0.4 in /usr/local/lib/python3.10/dist-packages (from mbstrdecoder<2,>=1.0.0->pytablewriter->lm_eval==0.4.2) (5.2.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets>=2.16.0->lm_eval==0.4.2) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets>=2.16.0->lm_eval==0.4.2) (3.7)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets>=2.16.0->lm_eval==0.4.2) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets>=2.16.0->lm_eval==0.4.2) (2024.2.2)\n",
            "Requirement already satisfied: python-dateutil<3.0.0,>=2.8.0 in /usr/local/lib/python3.10/dist-packages (from typepy[datetime]<2,>=1.3.2->pytablewriter->lm_eval==0.4.2) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2018.9 in /usr/local/lib/python3.10/dist-packages (from typepy[datetime]<2,>=1.3.2->pytablewriter->lm_eval==0.4.2) (2023.4)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch>=1.8->lm_eval==0.4.2) (2.1.5)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score>=0.0.4->lm_eval==0.4.2) (8.1.7)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets>=2.16.0->lm_eval==0.4.2) (2024.1)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch>=1.8->lm_eval==0.4.2) (1.3.0)\n"
          ]
        }
      ],
      "source": [
        "# Install LM-Eval\n",
        "!pip install git+https://github.com/EleutherAI/lm-evaluation-harness.git"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "JbpEeufJQnTr"
      },
      "outputs": [],
      "source": [
        "from lm_eval import api"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "hgzFSI8hH59H"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "\n",
        "HF_TOKEN = \"\"  # generate a user access token from https://huggingface.co/settings/tokens and copy it here\n",
        "os.environ[\"HF_TOKEN\"] = HF_TOKEN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Knxt2sGYyBrY"
      },
      "source": [
        "# Configure Evaluation\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "9WS47SmXyQyC"
      },
      "outputs": [],
      "source": [
        "YAML_boolq_string = \"\"\"\n",
        "task: demo_boolq\n",
        "dataset_path: super_glue\n",
        "dataset_name: boolq\n",
        "output_type: multiple_choice\n",
        "training_split: train\n",
        "validation_split: validation\n",
        "doc_to_text: \"{{passage}}\\nQuestion: {{question}}?\\nAnswer:\"\n",
        "doc_to_target: label\n",
        "doc_to_choice: [\"no\", \"yes\"]\n",
        "should_decontaminate: true\n",
        "doc_to_decontamination_query: passage\n",
        "metric_list:\n",
        "  - metric: acc\n",
        "  - metric: bleu\n",
        "  - metric: f1\n",
        "\"\"\"\n",
        "with open(\"boolq.yaml\", \"w\") as f:\n",
        "    f.write(YAML_boolq_string)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HEqYUlYvGuhd",
        "outputId": "fd36c9ca-fdc3-4567-cce7-6818f9ff69a8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2024-05-30 06:24:29.336227: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
            "2024-05-30 06:24:29.336292: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
            "2024-05-30 06:24:29.338088: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
            "2024-05-30 06:24:30.997165: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
            "2024-05-30:06:24:35,343 INFO     [__main__.py:254] Verbosity set to INFO\n",
            "2024-05-30:06:24:35,343 INFO     [__main__.py:277] Including path: ./\n",
            "2024-05-30:06:24:43,787 WARNING  [__main__.py:293]  --limit SHOULD ONLY BE USED FOR TESTING.REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.\n",
            "2024-05-30:06:24:43,788 INFO     [__main__.py:344] Selected Tasks: ['demo_boolq']\n",
            "2024-05-30:06:24:43,790 INFO     [evaluator.py:141] Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234\n",
            "2024-05-30:06:24:43,790 INFO     [evaluator.py:178] Initializing hf model, with arguments: {'pretrained': 'EleutherAI/pythia-2.8b'}\n",
            "2024-05-30:06:24:43,812 INFO     [huggingface.py:165] Using device 'cuda'\n",
            "/usr/local/lib/python3.10/dist-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n",
            "  warnings.warn(\n",
            "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:774] [Task: demo_boolq] metric acc is defined, but aggregation is not. using default aggregation=mean\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:786] [Task: demo_boolq] metric acc is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:774] [Task: demo_boolq] metric bleu is defined, but aggregation is not. using default aggregation=bleu\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:786] [Task: demo_boolq] metric bleu is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:774] [Task: demo_boolq] metric f1 is defined, but aggregation is not. using default aggregation=f1\n",
            "2024-05-30:06:25:03,269 WARNING  [task.py:786] [Task: demo_boolq] metric f1 is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "/usr/local/lib/python3.10/dist-packages/datasets/load.py:1486: FutureWarning: The repository for super_glue contains custom code which must be executed to correctly load the dataset. You can inspect the repository content at https://hf.co/datasets/super_glue\n",
            "You can avoid this message in future by passing the argument `trust_remote_code=True`.\n",
            "Passing `trust_remote_code=True` will be mandatory to load this dataset from the next major release of `datasets`.\n",
            "  warnings.warn(\n",
            "2024-05-30:06:25:06,006 INFO     [task.py:398] Building contexts for demo_boolq on rank 0...\n",
            "100% 20/20 [00:00<00:00, 1266.87it/s]\n",
            "2024-05-30:06:25:06,024 INFO     [evaluator.py:395] Running loglikelihood requests\n",
            "Running loglikelihood requests: 100% 40/40 [00:02<00:00, 14.95it/s]\n",
            "/usr/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n",
            "  self.pid = os.fork()\n",
            "bootstrapping for stddev: f1_score\n",
            "100% 100/100 [01:59<00:00,  1.20s/it]\n",
            "fatal: not a git repository (or any of the parent directories): .git\n",
            "2024-05-30:06:27:09,982 INFO     [evaluation_tracker.py:132] Saving results aggregated\n",
            "2024-05-30:06:27:09,983 INFO     [evaluation_tracker.py:203] Saving samples results\n",
            "hf (pretrained=EleutherAI/pythia-2.8b), gen_kwargs: (None), limit: 20.0, num_fewshot: None, batch_size: 1\n",
            "|  Tasks   |Version|Filter|n-shot|Metric|Value |   |Stderr|\n",
            "|----------|-------|------|-----:|------|-----:|---|-----:|\n",
            "|demo_boolq|Yaml   |none  |     0|acc   |0.7500|±  |0.0993|\n",
            "|          |       |none  |     0|f1    |0.8485|±  |0.0690|\n",
            "\n"
          ]
        }
      ],
      "source": [
        "!lm_eval \\\n",
        "    --model hf \\\n",
        "    --model_args pretrained=EleutherAI/pythia-2.8b \\\n",
        "    --include_path ./ \\\n",
        "    --tasks demo_boolq \\\n",
        "    --output output/ \\\n",
        "    --limit 20 \\\n",
        "    --log_samples"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HDyMUJieyX-S",
        "outputId": "2307e7c9-fbcc-467e-8780-107924666e54"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2024-05-30 06:27:14.929536: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
            "2024-05-30 06:27:14.929584: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
            "2024-05-30 06:27:14.930843: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
            "2024-05-30 06:27:16.588649: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
            "2024-05-30:06:27:23,447 INFO     [__main__.py:254] Verbosity set to INFO\n",
            "2024-05-30:06:27:23,447 INFO     [__main__.py:277] Including path: ./\n",
            "2024-05-30:06:27:29,860 WARNING  [__main__.py:293]  --limit SHOULD ONLY BE USED FOR TESTING.REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.\n",
            "2024-05-30:06:27:29,861 INFO     [__main__.py:344] Selected Tasks: ['demo_boolq']\n",
            "2024-05-30:06:27:29,863 INFO     [evaluator.py:141] Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234\n",
            "2024-05-30:06:27:29,863 INFO     [evaluator.py:178] Initializing hf model, with arguments: {'pretrained': 'mistralai/Mistral-7B-v0.1'}\n",
            "2024-05-30:06:27:29,885 INFO     [huggingface.py:165] Using device 'cuda'\n",
            "/usr/local/lib/python3.10/dist-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n",
            "  warnings.warn(\n",
            "Loading checkpoint shards: 100% 2/2 [01:01<00:00, 30.54s/it]\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:774] [Task: demo_boolq] metric acc is defined, but aggregation is not. using default aggregation=mean\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:786] [Task: demo_boolq] metric acc is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:774] [Task: demo_boolq] metric bleu is defined, but aggregation is not. using default aggregation=bleu\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:786] [Task: demo_boolq] metric bleu is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:774] [Task: demo_boolq] metric f1 is defined, but aggregation is not. using default aggregation=f1\n",
            "2024-05-30:06:28:33,160 WARNING  [task.py:786] [Task: demo_boolq] metric f1 is defined, but higher_is_better is not. using default higher_is_better=True\n",
            "/usr/local/lib/python3.10/dist-packages/datasets/load.py:1486: FutureWarning: The repository for super_glue contains custom code which must be executed to correctly load the dataset. You can inspect the repository content at https://hf.co/datasets/super_glue\n",
            "You can avoid this message in future by passing the argument `trust_remote_code=True`.\n",
            "Passing `trust_remote_code=True` will be mandatory to load this dataset from the next major release of `datasets`.\n",
            "  warnings.warn(\n",
            "2024-05-30:06:28:35,330 INFO     [task.py:398] Building contexts for demo_boolq on rank 0...\n",
            "100% 20/20 [00:00<00:00, 1841.06it/s]\n",
            "2024-05-30:06:28:35,342 INFO     [evaluator.py:395] Running loglikelihood requests\n",
            "Running loglikelihood requests: 100% 40/40 [00:22<00:00,  1.80it/s]\n",
            "/usr/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n",
            "  self.pid = os.fork()\n",
            "bootstrapping for stddev: f1_score\n",
            "100% 100/100 [02:00<00:00,  1.20s/it]\n",
            "fatal: not a git repository (or any of the parent directories): .git\n",
            "2024-05-30:06:30:59,045 INFO     [evaluation_tracker.py:132] Saving results aggregated\n",
            "2024-05-30:06:30:59,046 INFO     [evaluation_tracker.py:203] Saving samples results\n",
            "hf (pretrained=mistralai/Mistral-7B-v0.1), gen_kwargs: (None), limit: 20.0, num_fewshot: None, batch_size: 1\n",
            "|  Tasks   |Version|Filter|n-shot|Metric|Value|   |Stderr|\n",
            "|----------|-------|------|-----:|------|----:|---|-----:|\n",
            "|demo_boolq|Yaml   |none  |     0|acc   |0.800|±  |0.0918|\n",
            "|          |       |none  |     0|f1    |0.875|±  |0.0642|\n",
            "\n"
          ]
        }
      ],
      "source": [
        "!lm_eval \\\n",
        "    --model hf \\\n",
        "    --model_args pretrained=mistralai/Mistral-7B-v0.1 \\\n",
        "    --include_path ./ \\\n",
        "    --tasks demo_boolq \\\n",
        "    --output output/ \\\n",
        "    --limit 20 \\\n",
        "    --log_samples"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ivXfua4qLggD"
      },
      "source": [
        "# Convert to Analytics Platform JSON\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qjyOXzRvMBQs"
      },
      "source": [
        "### Let's start with defining the `name`, `models`, and `metrics` we used in this demo\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "LNGj4ncsLqVq"
      },
      "outputs": [],
      "source": [
        "name = \"LM Evaluation Harness Demo\"\n",
        "\n",
        "# models -> List[dict]\n",
        "models = [\n",
        "    {\n",
        "        \"model_id\": \"EleutherAI/pythia-2.8b\",\n",
        "        \"name\": \"Pythia-2.9b\",\n",
        "        \"owner\": \"EleutherAI\",\n",
        "    },\n",
        "    {\n",
        "        \"model_id\": \"mistralai/Mistral-7B-v0.1\",\n",
        "        \"name\": \"Mistral-7B-v0.1\",\n",
        "        \"owner\": \"Mistral AI\",\n",
        "    },\n",
        "]\n",
        "\n",
        "# metrics -> List[dict]\n",
        "all_metrics = [\n",
        "    {\n",
        "        \"name\": \"F1\",\n",
        "        \"display_name\": \"F1\",\n",
        "        \"description\": \"F1 score \",\n",
        "        \"author\": \"algorithm\",\n",
        "        \"type\": \"numerical\",\n",
        "        \"aggregator\": \"average\",\n",
        "        \"range\": [0, 1.0, 0.1],\n",
        "    },\n",
        "    {\n",
        "        \"name\": \"Accuracy\",\n",
        "        \"display_name\": \"Accuracy\",\n",
        "        \"description\": \"Prediction accuracy\",\n",
        "        \"author\": \"algorithm\",\n",
        "        \"type\": \"numerical\",\n",
        "        \"aggregator\": \"average\",\n",
        "        \"range\": [0, 1.0, 0.1],\n",
        "    },\n",
        "]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HntEhvugQt2Y"
      },
      "source": [
        "## Now let's define `tasks`, `documents`, and `evaluations`\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "9yRse3PQQsxb"
      },
      "outputs": [],
      "source": [
        "import json\n",
        "\n",
        "outputs = []\n",
        "\n",
        "# modify output filepath for pythia-2.8b here\n",
        "with open(\n",
        "    \"output/EleutherAI__pythia-2.8b/samples_demo_boolq_2024-05-30T02-24-44.249027.json\",\n",
        "    \"r\",\n",
        ") as f:\n",
        "    model_1_samples = json.load(f)\n",
        "\n",
        "# modify output filepath for Mistral-7B-v0.1 here\n",
        "with open(\n",
        "    \"output/mistralai__Mistral-7B-v0.1/samples_demo_boolq_2024-05-30T02-28-34.024454.json\",\n",
        "    \"r\",\n",
        ") as f:\n",
        "    model_2_samples = json.load(f)\n",
        "\n",
        "all_tasks = []\n",
        "all_documents = []\n",
        "all_evaluations = []\n",
        "for model_1_sample, model_2_sample in zip(model_1_samples, model_2_samples):\n",
        "    assert model_1_sample[\"doc_id\"] == model_2_sample[\"doc_id\"]\n",
        "    doc_id = model_1_sample[\"doc_id\"]\n",
        "    content_1 = model_1_sample.get(\"doc\")\n",
        "    content_2 = model_2_sample.get(\"doc\")\n",
        "    passage_text = content_1.get(\"passage\")\n",
        "    document = {\"document_id\": f\"doc_{doc_id}\", \"text\": passage_text}\n",
        "\n",
        "    all_documents.extend([document])\n",
        "    instance = {\n",
        "        \"task_id\": f\"{doc_id}\",\n",
        "        \"task_type\": \"conversation\",\n",
        "        \"contexts\": [{\"document_id\": document[\"document_id\"]}],\n",
        "        \"input\": [{\"speaker\": \"user\", \"text\": f\"{model_1_sample['doc']['question']}\"}],\n",
        "        \"targets\": [{\"text\": \"yes\" if model_1_sample[\"target\"] else \"no\"}],\n",
        "    }\n",
        "    all_tasks.append(instance)\n",
        "\n",
        "    for i, pred in enumerate([model_1_sample, model_2_sample]):\n",
        "        model_id = models[i][\"model_id\"]\n",
        "        target = \"yes\" if pred[\"target\"] else \"no\"\n",
        "        prediction = (\n",
        "            \"no\"\n",
        "            if pred[\"filtered_resps\"][0][0] > pred[\"filtered_resps\"][1][0]\n",
        "            else \"yes\"\n",
        "        )\n",
        "        all_evaluations.append(\n",
        "            {\n",
        "                \"task_id\": f\"{doc_id}\",\n",
        "                \"model_id\": model_id,\n",
        "                \"model_response\": prediction,\n",
        "                \"annotations\": {\n",
        "                    \"Accuracy\": {\n",
        "                        \"system\": {\n",
        "                            \"value\": 1 if prediction == target else 0,\n",
        "                            \"duration\": 0,\n",
        "                        }\n",
        "                    },\n",
        "                    \"F1\": {\n",
        "                        \"system\": {\n",
        "                            \"value\": 1 if prediction == target else 0,\n",
        "                            \"duration\": 0,\n",
        "                        }\n",
        "                    },\n",
        "                },\n",
        "            }\n",
        "        )"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NM_VZxEU5UiX",
        "outputId": "7cb16261-b0ed-49dd-e2e2-0a1974c17f9f"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(20, 20, 40)"
            ]
          },
          "execution_count": 29,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(all_tasks), len(all_documents), len(all_evaluations)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bekGOYtEcABN"
      },
      "source": [
        "## Now we can write the output to file and import it into our dashboard for analysis :D\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "3tjuCibsYzG7"
      },
      "outputs": [],
      "source": [
        "import json\n",
        "\n",
        "output = {\n",
        "    \"name\": name,\n",
        "    \"models\": models,\n",
        "    \"metrics\": all_metrics,\n",
        "    \"documents\": all_documents,\n",
        "    \"tasks\": all_tasks,\n",
        "    \"evaluations\": all_evaluations,\n",
        "}\n",
        "\n",
        "with open(\n",
        "    file=\"lm-eval-harness-inspectorraget-demo.json\", mode=\"w\", encoding=\"utf-8\"\n",
        ") as fp:\n",
        "    json.dump(output, fp, indent=4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iIcWaE51cuAh"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8BEkotPhx-_w"
      },
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "machine_shape": "hm",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}