cr_speak_equal.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from diart import SpeakerDiarization as DiartSpeakerDiarization, SpeakerDiarizationConfig\n",
    "from diart.sources import FileAudioSource\n",
    "from diart.inference import StreamingInference\n",
    "from diart.models import SegmentationModel, EmbeddingModel\n",
    "from diart.optim import Optimizer\n",
    "\n",
    "from pyannote.core import Segment, Annotation, notebook\n",
    "from pyannote.database.util import load_rttm\n",
    "from pyannote.audio import Pipeline\n",
    "from pyannote.audio.pipelines import SpeakerDiarization\n",
    "from pyannote.metrics.diarization import DiarizationErrorRate\n",
    "\n",
    "from huggingface_hub import login\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import json\n",
    "import warnings\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "HUGGING_FACE_TOKEN = \"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\"\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "#login(HUGGING_FACE_TOKEN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "segmentation = SegmentationModel.from_pretrained(\"pyannote/segmentation\")\n",
    "embedding = EmbeddingModel.from_pretrained(\"pyannote/wespeaker-voxceleb-resnet34-LM\")\n",
    "\n",
    "config = SpeakerDiarizationConfig(\n",
    "    segmentation=segmentation,\n",
    "    embedding=embedding,\n",
    "    tau_active=0.6,\n",
    "    rho_update=0.3,\n",
    "    delta_new=0.9,\n",
    "    max_speakers=4\n",
    ")\n",
    "\n",
    "pipeline = DiartSpeakerDiarization(config)\n",
    "source = FileAudioSource(\"data/example3.wav\", 16000)\n",
    "\n",
    "inference = StreamingInference(\n",
    "    pipeline, \n",
    "    source, \n",
    "    do_profile=False, \n",
    "    do_plot=False, \n",
    "    show_progress=False\n",
    ")\n",
    "\n",
    "prediction = inference()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipeline_params = Pipeline \\\n",
    "    .from_pretrained(\"pyannote/speaker-diarization-3.1\") \\\n",
    "    .parameters(instantiated=True)\n",
    "\n",
    "pipeline = SpeakerDiarization(\n",
    "    segmentation=\"pyannote/segmentation\",\n",
    "    embedding=\"pyannote/wespeaker-voxceleb-resnet34-LM\"\n",
    ")\n",
    "\n",
    "pipeline = pipeline.instantiate(pipeline_params)\n",
    "diarization = pipeline(\"data/example3.wav\")\n",
    "\n",
    "reference = Annotation()\n",
    "for turn, _, speaker in diarization.itertracks(yield_label=True):\n",
    "    reference[Segment(turn.start, turn.end)] = speaker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "der = DiarizationErrorRate()\n",
    "print(f\"Diarization Error Rate: {der(prediction, reference)}\")\n",
    "print(json.dumps(der.compute_components(prediction, reference), indent=3))\n",
    "\n",
    "plt.figure(figsize=(15, 4))\n",
    "notebook.reset()\n",
    "notebook.plot_annotation(prediction)\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(15, 4))\n",
    "notebook.reset()\n",
    "notebook.plot_annotation(reference)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Compare of tester.py, diart_mic.ipynb and webapp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "FOLDER = \"data/derived/webapp\"\n",
    "\n",
    "for f in [file for file in os.listdir(FOLDER) if file.endswith(\".rttm\")]:\n",
    "    plt.figure(figsize=(15, 4))\n",
    "    plt.title(f)\n",
    "    notebook.reset()\n",
    "    notebook.plot_annotation(load_rttm(FOLDER + \"/\" +  f)[\"mic://localhost\"])\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKMAAAGMCAYAAAAGOGFVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA2zUlEQVR4nO3deXRV5dk34DtAgAAhDDJPxgoOiIigFrR1QFGrVkCrKCqo1a8Kjq+11olqq1U6vFqggrbaOo9VrNaBIg5YFbCOxWKlKigzCCTMkP39wcp5jUxJCDsD17UWi2TvfZ7n3tPJOb+z93OykiRJAgAAAABSUKuyCwAAAABg5yGMAgAAACA1wigAAAAAUiOMAgAAACA1wigAAAAAUiOMAgAAACA1wigAAAAAUiOMAgAAACA1wigAAAAAUiOMAgAAACA1wigAAAAAUiOMAgAAACA1wigAAAAAUiOMAoBqYurUqTF8+PDo2rVrNGzYMDp27BinnHJKfPzxx5ss+9FHH8UxxxwTjRo1imbNmsWZZ54ZCxcuLLHMv//977jyyitjv/32i9zc3GjTpk0cd9xxMW3atG3WctRRR0VWVlYMHz681PUXFRXFyJEjIz8/P+rXrx/77rtvPPTQQ5tddvTo0bHXXntFvXr1ol27dnH55ZfHihUrStVPYWFhjBgxIo455pho1qxZZGVlxZ/+9Ket1nXHHXfEfvvtFzk5OdG8efM44ogj4r333iv1ukVEvPjii3HuuefGPvvsE7Vr145dd911s8v97Gc/i6ysrC3+e/3110ssX5p9GRFx0003xfe///1o1apVZGVlxc9+9rMy9V+/fv1Nlt1SjbfcckuZts03/e1vf9tsfStXroyf/exn8fLLL29X+wBA1VansgsAAErn1ltvjddffz1+8IMfxL777hvz5s2L0aNHx/777x9vvvlm7LPPPhER8cUXX8R3v/vdyMvLi5tvvjkKCwvj17/+dXzwwQcxZcqUqFu3bkRE/OEPf4g//vGPcdJJJ8WFF14Yy5Yti3HjxsW3v/3teP755+PII4/cbB1/+ctf4o033ihz/ddcc03ccsstcd5558UBBxwQ48ePj9NPPz2ysrJi0KBBmeV+8pOfxMiRI+Pkk0+OSy65JKZPnx6jRo2Kf/3rX/HCCy9ss59FixbFjTfeGB07dozu3btvM9g455xz4oEHHoizzjorhg8fHitWrIh33nknFixYUKb1e/DBB+ORRx6J/fffP9q2bbvF5QYOHBi77777JtOvvvrqKCwsjAMOOCAzrbT7MiLi2muvjdatW0ePHj1KtZ3uuOOOaNSoUeb32rVrb3a5o446Ks4666wS03r06LHN9rfmb3/7W4wZM2aTQGrlypVxww03RETEYYcdtl19AABVWELqhgwZkpx44omVXQaVzHFAMccCSVK64+D1119P1qxZU2Laxx9/nNSrVy8ZPHhwZtoFF1yQ5OTkJJ9//nlm2oQJE5KISMaNG5eZNm3atKSgoKBEe4sWLUpatGiRHHzwwZutYdWqVcmuu+6a3HjjjUlEJMOGDSvV+n3xxRdJdnZ2ieWLioqS73znO0n79u2T9evXJ0mSJHPmzEnq1KmTnHnmmSUeP2rUqCQikqeffnqbfa1evTqZO3dukiRJMnXq1CQiknvuuWezyz7yyCNJRCR/+ctfSrUeW/Pll18ma9euTZIkSY477rikU6dOpX7srFmzkqysrOS8884rMb20+zJJkuTTTz9NkiRJFi5cmEREMmLEiM32NWLEiCQikoULF26zrrLs49IoLCxMkiRJhg0blmzuZei2agcAaga36dUwY8aMiV133TXq168fBx10UEyZMqWySyJlr776apxwwgnRtm3byMrKiqeeeqqyS6KS/PKXv4wDDjggcnNzo2XLltG/f/+YMWNGZZfFdujTp0+JK2EiIjp37hxdu3aNjz76KDPtiSeeiOOPPz46duwYd9xxR+y7774xcODAqFWrVvz4xz+O5557LiIievbsWeLKmIiI5s2bx3e+850S7X3dyJEjo6ioKK644ooy1T5+/PhYt25dXHjhhZlpWVlZccEFF8QXX3yRudLqjTfeiPXr15e4UioiMr8//PDD2+yrXr160bp161LV9dvf/jYOPPDAGDBgQBQVFZX6VsDNadu2bWRnZ5frsQ899FAkSRKDBw8uMf3r+7LYkUceGV26dIlHH320xLJbui1wS5IkieXLl0eSJNtcdtWqVbF69eoytV98O+D06dPj9NNPj6ZNm8YhhxwSQ4cOjTFjxkREydsAP/vss2jRokVERNxwww2Z6cVXTw0dOjQaNWoUs2bNiuOPPz4aNWoU7dq1y7T1wQcfxBFHHBENGzaMTp06xYMPPlimegGA9FT4bXobFi+u6Ca3qnbz5qn2VxWtXbs26tatG4888khcfvnlMXbs2DjooIPitttui6OPPjpmzJgRLVu2TL2ur1asTbW/pg3rbnuhGqz4OFixYkV07949zjnnnBg4cGBllxXL1ixLtb+8enmp9lcVFR8Lr7zySgwbNiwOOOCAWL9+fVx99dXRr1+/mD59ejRs2DD1ulYtK9sb2e2Vk7fp+Dc1UZIkMX/+/OjatWtERHz55ZexYMGC2G+//SIion379nHLLbdE586d44orrogXX3wxTjzxxHjnnXcyj/mmefPmxS677LLJ9FmzZsUtt9wSd999d+Tk5JSpznfeeScaNmwYe+21V4npBx54YGb+IYccEmvWrImI2KT9Bg0aRETE22+/XaZ+t2b58uUxZcqUuPDCC+Pqq6+OUaNGRWFhYeTn58ctt9wSp5xySoX1tS0PPPBAdOjQIb773e9mphXvy169em2y/IEHHhh/+9vftqvP3XbbLQoLC6Nhw4bRv3//+M1vfhOtWrXaZLk//elP8fvf/z6SJIm99torrr322jj99NNL3c8PfvCD6Ny5c9x8882RJEn06NEj5syZExMmTIj77rsvs1yLFi3ijjvuiAsuuCAGDBiQ+Ru27777ZpbZsGFDHHvssfHd7343Ro4cGQ888EAMHz48GjZsGNdcc00MHjw4Bg4cGGPHjo2zzjorevfuHfn5+duxlQCAHaHCw6h5++5X0U1uVbsvZ5f5MY8//njccMMN8cknn0SDBg2iR48eMX78+Bg2bFgsXbo0evToEaNHj441a9bE6aefHr/73e8yn0QXFRXFrbfeGnfeeWfMmzcvunTpEtddd12cfPLJEbHxRdL5558fL730UsybNy86duwYF154YVxyySVbrGfq1Knxve99L6644or4yU9+EkuXLo0rrrgixo8fH2vWrIlevXrF//7v/0b37t0jYuMnjU899VQMHz48brrppvj888+jqKgofvvb38Z5550XZ599dkREjB07Np599tm4++6746qrrirzdtpex46clGp/b95wdJmWr6nHwbHHHhvHHntsObdixTvzudK/YakIT/d/tsyPqanHwvPPP1+i3T/96U/RsmXLePvtt0u84U3LvWc9nmp//2/8GWVavroeBx988EF8+eWX0bNnz9hvv/3i+OOPj4iN4wddffXVccIJJ2Ta3GOPPeLpp5+Opk2bxptvvrnZMOq1116LN954I6699tpN5v3P//xP9OjRY5Orlkpj7ty5mYG1v65NmzYRETFnzpxMjRERr7/+ehx++OEl6orYGNBUlJkzZ0aSJPHwww9HnTp1YuTIkZGXlxe33357DBo0KBo3bhzHHHNMhfW3Jf/617/i/fffjyuvvLLE9pk7d25E/N82+ro2bdrEkiVLYs2aNVGvXr0y9de0adMYPnx49O7dO+rVqxevvfZajBkzJqZMmRLTpk2Lxo0bZ5bt06dPnHLKKZGfnx9z5syJMWPGxODBg2PZsmVxwQUXlKq/7t27b3KVUpcuXWLChAlxxhklz9OTTz45Lrjggth33303mRcRsXr16jjjjDPipz/9aUREnH766dG2bds455xz4qGHHopTTz01IjaOc7XnnnvGn//85y0O5A4AVJ6dbgDzuXPnxmmnnRYjR46MAQMGREFBQbz22muZS9QnTpwY9evXj5dffjk+++yzOPvss6N58+Zx0003RcTG217uv//+GDt2bHTu3DleffXVOOOMM6JFixZx6KGHRlFRUbRv3z4ee+yxaN68efzjH/+I888/P9q0abPZT1hfeumlGDhwYIwcOTLOP//8iNj4CWJOTk4899xzkZeXF+PGjYu+ffvGxx9/HM2aNYuIiE8++SSeeOKJ+Mtf/hK1a9eOtWvXxttvv515cRYRUatWrTjyyCPLNchsTVdTjwPKbmc6FpYt23iVWvFj+D/V9Tg47LDDYsOGDdG7d+/o3r17TJw4Mf7+979HxMbBzr+pODxbsWJF9O7de5P5CxYsiNNPPz3y8/PjyiuvLDFv0qRJ8cQTT8Rbb71Vrm28atWqzYYmxd/gtmrVqoiI2H///eOggw6KW2+9Ndq1axeHH354fPTRR3HBBRdEdnZ2ZrmKUFhYGBERixcvjjfffDMOOuigiIj4/ve/H/n5+fGLX/wilTDqgQceiIjY5Ba94nXd1nYraxj1zRD0pJNOigMPPDAGDx4cv//970t8gPXNb/Y755xzomfPnnH11VfH0KFDS3WF3I9+9KMy1bctP/zhDzM/N2nSJPbYY4/45JNPSpxLe+yxRzRp0iT++9//VmjfAEDF2CnDqPXr18fAgQOjU6dOERHRrVu3zPy6devG3XffHQ0aNIiuXbvGjTfeGD/+8Y/j5z//eaxbty5uvvnm+Pvf/555Eb/bbrvF5MmTY9y4cXHooYdGdnZ25ltgIiLy8/PjjTfeiEcffXSTNxxPPvlknHXWWfGHP/wh80ne5MmTY8qUKbFgwYLMi8tf//rX8dRTT8Xjjz+eeVOydu3auPfeezNjK8yZMyc2bNiwyeX1rVq1in//+98VuQlrhJp6HFB2O8uxUFRUFJdeemkcfPDBmW9c4/9Ux+PgiiuuiNtvvz1yc3Pj8ccfjzvvvDPWrl0bP//5z6Nfv34lxkz64IMPonfv3rFy5cqIiHj00Udj7733LtHvihUr4vjjj4+CgoKYPHlyibGk1q9fHxdffHGceeaZJb7pbXPmzZtX4ve8vLzIycmJnJyczC14X1c8DtHXQ40nnngiTj311DjnnHMiYuO3vF1++eXxyiuvZMY927BhQyxcuLBEW82aNdtkTK2tKe4zPz8/E0RFRDRq1ChOOOGEuP/++2P9+vVRp86Oe7mUJEk8+OCDsc8++5S4He3r9ZV2u22P008/Pf7nf/4n/v73v2/1auq6devG8OHD40c/+lG8/fbbccghh2yz7Yq8Ta5+/fqbPM/l5eVF+/btN7nqLi8vL7766qsK6xsAqDg7XRjVvXv36Nu3b3Tr1i2OPvro6NevX5x88snRtGnTzPzicSkiInr37h2FhYUxe/bsKCwsjJUrV8ZRRx1Vos21a9eW+IrjMWPGxN133x2zZs2KVatWxdq1azNjdxR766234plnnonHH388+vfvn5n+3nvvRWFhYTT/xlhYq1atipkzZ2Z+79SpkwBiOzgOKLazHAvDhg2LDz/8MCZPnlzqbbMzqY7HwapVq6KoqCgGDBgQbdu2jYiNx0FxyFR8i1fExqtE3n333bj44otj0qRJ8cMf/jA6d+6cWXbt2rUxcODAeP/99+OFF17YJLC89957Y8aMGTFu3Lj47LPPSswrKCiIzz77LFq2bBkNGjTY5Jaye+65J4YOHRpt2rSJSZMmRZIkm70VrXgdIiLatWsXkydPjv/85z8xb9686Ny5c7Ru3Tratm0bXbp0iYiI2bNnbxJyTJo0KQ477LAoreI+NzdOUsuWLWPdunWxYsWKyMvbcWPRvf766/H555/HL3/5y03mFW/Lr+/LYnPnzo1mzZqV+aqorenQoUMsWbKkVMtFRKmWjai4wCwitnjl55aml2ZwdgAgfRUeRrV+/92KbrJC1a5dOyZMmBD/+Mc/4sUXX4xRo0bFNddcU6rbDoov53/22WejXbt2JeYVvxh8+OGH44orrojf/OY30bt378jNzY1f/epXm7T/rW99K5o3bx533313HHfccZlv3yksLIw2bdrEyy+/vEn/TZo0yfz8zcGHd9lll6hdu3bMnz+/xPT58+eX+huFKtpzVx6+7YUqSU09Dqqi+46t2t9mtDMcC8OHD49nnnkmXn311Wjfvv0212tHOevekyut722pTsfBmjVrYujQofHhhx/GvffeW2JcpYYNG0a7du2iRYsWMW3atMz0unXrxu677x4zZ86MPn36RFZWVtx+++0xbty4KCoqirPOOismTpwYjz76aBx66KGbrOOsWbNi3bp1cfDBB28y795774177703nnzyyejfv39MmDChxPzican222+/+MMf/hAfffRRiauyirfBN4O5iI3fFNi5c+eIiJg+fXrMnTs3hg4dGhERrVu33qSv4nHUSqtt27bRunXrzY5DNWfOnKhfv37k5uaWqc2yeuCBByIrK2uzA4Jvbl8WmzJlyma3WXklSRKfffZZiQB1S4pvfdueD0O+eRXTtqYDADVLhYdR1eHb7bKysuLggw+Ogw8+OK6//vro1KlTPPnkkxGx8dPnVatWZT7Fe/PNN6NRo0bRoUOHzCeQs2bN2uyL9YiNn3D26dOnxFdXf/3qhWK77LJL/OUvf4nDDjssTjnllHj00UcjOzs79t9//5g3b17UqVOnTF/RXLdu3ejZs2dMnDgx82l6UVFRTJw4MYYPH17qdipSVf92u5p4HFRF1eHb7WrqsZAkSVx00UXx5JNPxssvv1zp3yhV1b/drjocB1lZWXH11VfHu+++G+PHj4/vfe97m+3vpJNOij//+c8xe/bszFUsEydOjI8//jguu+yyePTRRzO3fl100UXxyCOPxLhx47b4DZyDBg3abPAxYMCA+N73vhfnnXde5ja3I488crNtnHjiiXHZZZfF73//+xg9enREbDxGx44dG+3atYs+ffps9nERG/+eXXnlldGgQYPM+EP169ffYl9lceqpp8btt98eEyZMyFzdtmjRohg/fnwcccQRUatWre3uY0vWrVsXjz32WBxyyCHRsWPHzS6zrX1ZHgsXLtwkSLrjjjti4cKFJcbI2txyBQUFcdttt8Uuu+wSPXv2zExftGhRLFq0KDp27FjiKsItKQ7Qly5dWiJYL37s0qVLy7paAEA1stPdpvfWW2/FxIkTo1+/ftGyZct46623YuHChbHXXnvF+++/H2vXro1zzz03rr322vjss89ixIgRMXz48KhVq1bk5ubGFVdcEZdddlkUFRXFIYccEsuWLYvXX389GjduHEOGDInOnTvHvffeGy+88ELk5+fHfffdF1OnTt3sm8CWLVvGSy+9FIcffnicdtpp8fDDD8eRRx4ZvXv3jv79+8fIkSOjS5cuMWfOnHj22WdjwIABm/1652KXX355DBkyJHr16hUHHnhg3HbbbbFixYrMt+vxf2rycVBYWBiffPJJ5vdPP/003n333WjWrNkW3+zszGrysTBs2LB48MEHY/z48ZGbm5sZy6d4DB/+T3U5Dr797W/HokWLom/fvvHmm2/GuHHjomfPnrHbbrvF+++/n2nj6quvjsceeywOP/zwyM/Pj/z8/HjkkUeic+fOMXPmzHj55ZfjhRdeiNtuuy1+//vfR+/evaNBgwZx//33l6hlwIAB0bBhw9hzzz1jzz333Oy2y8/PL3FL4Za0b98+Lr300vjVr34V69atiwMOOCCeeuqpeO211+KBBx4ocZvVJZdcEqtXr4799tsv1q1bFw8++GBMmTIl/vznP5f6eWz06NGxdOnSzLf0/fWvf40vvvgiIjYGcMW33v30pz+NRx99NE466aS4/PLLIy8vL8aOHZsZC6ws3n///Xj66acjYuOXCixbtix+8YtfRMTGq7a+/s2GEREvvPBCLF68eJOBy7/u6/vykksuicLCwvjVr34V3bp12+Tv+3333Reff/55ZmywV199NdP/mWeemRkPrVOnTnHqqadGt27don79+jF58uR4+OGHY7/99ov/9//+X6a9MWPGxFNPPRUnnHBCdOzYMebOnZu51fS+++4rMT7X6NGj44Ybbij1rZLFQdbFF18cRx99dNSuXTsGDRoUOTk5sffee8cjjzwSXbp0iWbNmsU+++xjrDsAqGmSncz06dOTo48+OmnRokVSr169pEuXLsmoUaOSJEmSIUOGJCeeeGJy/fXXJ82bN08aNWqUnHfeecnq1aszjy8qKkpuu+22ZI899kiys7OTFi1aJEcffXTyyiuvJEmSJKtXr06GDh2a5OXlJU2aNEkuuOCC5Kqrrkq6d++eaaO4n2Jz5sxJunTpkpxyyinJ+vXrk+XLlycXXXRR0rZt2yQ7Ozvp0KFDMnjw4GTWrFlJkiTJiBEjSrT3daNGjUo6duyY1K1bNznwwAOTN998s2I3YA1Rk4+DSZMmJRGxyb8hQ4ZU+HasCWrysbC54yAiknvuuafCt2N1V12Og3bt2m1xv0ZEifY+/PDDpF+/fknt2rWTrKyspFatWknz5s2Tvn37Ji+++GKmz6219+mnn251u0VEMmzYsFJv5w0bNiQ333xz0qlTp6Ru3bpJ165dk/vvv3+T5e65556ke/fuScOGDZPc3Nykb9++yUsvvVTqfpIkSTp16lTq9Zo5c2YyYMCApHHjxklOTk5yxBFHJFOmTClTf8V1b6nPzT0HDxo0KMnOzk4WL1681XaL92WDBg2SJk2aJIMHD07mzZu3yXKHHnroFvufNGlSZrkf/vCHyd57753k5uYm2dnZye6775785Cc/SZYvX16ivRdffDE56qijktatWyfZ2dlJkyZNkn79+iUTJ07cpO8RI0Zs0k/xtIULF26y/Pr165OLLrooadGiRZKVlZV8/SXpP/7xj6Rnz55J3bp1k4hIRowYkSTJxuO1YcOGm13vrl27bjK9U6dOyXHHHbfJdACg8mUliZEdiw0dOjSWLl0aTz31VGWXQiVyHFDMsUCE4wAAACrajhsIAQAAAAC+YacbMwoAoCyKx1vbkpycnMwYVAAAbJvb9AAAtiIrK2ur84cMGRJ/+tOf0ikGAKAGcGUUAMBWTJgwYavz27Ztm1IlAAA1gyujAAAAAEiNAcwBAAAASE25b9MrKiqKOXPmRG5u7jbHUgAAAACg5kqSJAoKCqJt27ZRq9bWr30qdxg1Z86c6NChQ3kfDgAAAEANM3v27Gjfvv1Wlyl3GJWbm5vppHHjxuVtBgAAAIBqbvny5dGhQ4dMXrQ15Q6jim/Na9y4sTAKAAAAgFIN5WQAcwAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXbHUZtWLCgIuqgGtswf34s/81vY8P8+an2ufSGG2PpDTdu0u+igjVx16RPYlHBmtTqoWJtbh/urPu1eL0/nrt8p1z/qubfr0yLX5x9Y/z7lWkV1uYni+bGFc+Nik8Wza2wNktryeol8ccP/hBj3hkdf/zgD7Fk9ZJYsnpJPPjRA7Fk9ZId3vfW+kmrjrKq7Loq6rlwS+2Ut/3KPI6pGr6cOyfu+d398eXcORXW5oolK2PaQ+/FiiUrK3TZWbOXxg2/mxyzZi+tgCqrpkUFa+L25/8dtz//7wp57VDe54Yt7Zfi6bNmL61Rr2/Ksp121te221Jd3gdU9N/RmmJxGdZ7+8OohQu3twmquQ0LFkTBb/831WByw4IFseLOu2LFnXdt0u+igjXxx5dn7rRPADXB5vbhzrpfi9f7vwsKd8r1r2pmT58Zz3Q8KGZPn1lhbX721YL4eM3z8dlX6X+489XqJTF+5pPxwufPxfiZT8ZXq5fEV6uXxMMzHoyvdnDYsq1+0qqjrCq7rop6LtxSO+VtvzKPY6qGBfMXxtqJG/+vKCu/WhVvP/xBrPxqVYUu++Xc5fHc4hXx5dzlFVFmlbSoYE089Mbn8dAbn1dYGFWe54Yt7Zfi6V/OXV6jXt+UZTvtrK9tt6W6vA+o6L+jNcXiwhTDKAAAAAAoLWEUAAAAAKkRRgEAAACQGmEUAAAAAKkRRgEAAACQGmEUAAAAAKkRRgEAAACQGmEUAAAAAKkRRgEAAACQGmEUAAAAAKkRRgEAAACQmjrb20DRsuWxYfHiiqiFaqpo6bLKLmGzClati69WrK3sMiiHglXrtjpvZ9qv39wWO9v6VzVr1xVFRMSKoqwK2w8r16yPiIhV61fEsjXpPp8Wri3c6rwdWc/W+k6zjrIqbd072vY+F2ztebY87VfmcUzVsGrdyoiIWLdifaxatrpC2lxTWPZjfE3h2m32v27lxuO/cO36Gvs39evneEW8dtjWc8a2fHO/fHPf1pTXN+XZTjVl3StKdXkfUNF/R2uKglXrS73sdodRS84+J9bVcoEVVc9F906r7BLYAXb2/bqzr39l22/2FxEdmsUtC/LilpGTKqTNOjkLotmeEXfNuCnumlEhTVaI6/5xTWWXEBFVp46qZkc/F5S1/ap6HJOexouaxnfimHhn5EfxTnxUaXU8e/3EbS6zKKdOxB5N47oJ/4mY8J8UqqpcVeG1w7b2S1WosbLszOteVtVpW1WnWivS+jUrSr2sFAkAAACA1AijAAAAAEiNMAoAAACA1Gz3mFHN7rk7mh/QqyJqoZpaN/2jWDzotMouYxOjzuoVu7fOrewyKIdP5hVs8T7rnW2/fnNb7GzrX9X8449L4t3FEVe1XBaHDu1fIW2+9ukH8YdPIs7b45r47m5dK6TN0vps2adbHJPp531uil3z8iul7zTrKKvS1r2jbe9zwdaeZ8vTfmUex1QN/3p/enzw15nR48q9ots+FXMMLP7sq1KNAfV1x93YN5rv2nSry0x958v466RP4udHdY5ePdpvT4lV1tfP8Yp47bCt54xt+eZ++ea+rSmvb8qznWrKuleU6vI+oKL/jtYU7/5nThxxa+mW3e4wqlZe46jdvPn2NkM1tqFJXmWXsFm5OdnRtGHdyi6DcsjNyd7qvJ1pv35zW+xs61/V1M3eeEFxw1pJhe2HBvU2/inOqdMw8uql+3zaqG6jrc7bkfVsre806yir0ta9o23vc8HWnmfL035lHsdUDTnZDSIiIrthncjJq18hbdZrVPZjvF6jutvsP7vBxuO/Ud06NfZv6tfP8Yp47bCt54xt+eZ++ea+rSmvb8qznWrKuleU6vI+oKL/jtYUuTmlj5jcpgcAAABAaoRRAAAAAKRGGAUAAABAaoRRAAAAAKRGGAUAAABAaoRRAAAAAKRGGAUAAABAaoRRAAAAAKRGGAUAAABAaoRRAAAAAKRGGAUAAABAaupsbwO1W7SoiDqoxmq3bBm5l18WtVu2TLXPhuefl/n563bJrRfnHvat2CW3Xmr1ULE2tw931v1avN67tWy0U65/VdNh72/F8VP+Fh2O+F6Ftblr05bRpd4xsWvT9J5DizWt3yxO/NaAWL1+ddSvUz+a1m8WERGD9jg98/OO7Htr/WxrfmWp7Loq6rlwS+2Ut/3KPI6pGlq2ahF1+278v6I0aJoTPQd1iwZNcyp02XZtGsexzRtGuzaNK6LMKmmX3HpxWu9OmZ8ror3yPDdsab8UT2/ZpnGNen1Tlu20s7623Zbq8j6gov+O1hTNG5V+vbOSJEnK08ny5csjLy8vli1bFo0b19wncgAAAAC2riw5kdv0AAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAGAns6hgTdw16ZNYVLCmxM8VacWSlTHtofdixZKVlfL4HWVHbK8dtQ8gLRvmz4+lN9wYS2+4MTbMn1/Z5QBQDQijAGAns6hgTfzx5ZmZMKr454q08qtV8fbDH8TKr1ZVyuN3lB2xvXbUPoC0bFiwIFbceVesuPOu2LBgQWWXA0A1IIwCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDV1KrsAAKByFKxat8P7WFO4NlYtW12ux1VlBavWxVcrKqbGNPYDAEBVIowCgJ3URfdO2+F9PHv9xB3eR2VIY9sBANRUbtMDAAAAIDXCKAAAAABSI4wCAAAAIDXGjAKAndSos3pFxI4d/+i4G/tG812blvlxiz/7qkqPNzXqrF6xe+vcCmnrk3kFxqACAHYqwigA2Enl5mTv8D7qNaobOXn1y/W4qiw3JzuaNqyYGtPYDwAAVYnb9AAAAABIjTAKAAAAgNQIowAAAABIjTAKAAAAgNQIowAAAABIjTAKAAAAgNQIowAAAABIjTAKAAAAgNQIowAAAABIjTAKAAAAgNQIowAAAABITZ3KLgAASNcuufXi3MO+Fbvk1ouIKPFzRWnQNCd6DuoWDZrmVMrjd5Rvbruq2iakqXbLltHw/PMyPwPAtmQlSZKU54HLly+PvLy8WLZsWTRu3Lii6wIAAACgmihLTuQ2PQAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKADYia1YsjKmPfRerFiyskq1BQBAzSWMAoCd2MqvVsXbD38QK79aVaXaAgCg5hJGAQAAAJAaYRQAAAAAqRFGAQAAAJAaYRQAAAAAqRFGAQAAAJAaYRQAAAAAqRFGAQAAAJAaYRQAAAAAqRFGAQAAAJAaYRQAAAAAqRFGAQAAAJCaOpVdAABQ+dYUro1Vy1ZvdxsAALAtwigAIJ69fmJllwAAwE7CbXoAAAAApEYYBQAAAEBqhFEAAAAApMaYUQBAHHdj32i+a9PtamPxZ18ZewoAgG0SRgEAUa9R3cjJq7/dbQAAwLa4TQ8AAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAGAn1qBpTvQc1C0aNM2pUm0BAFBzZSVJkpTngcuXL4+8vLxYtmxZNG7cuKLrAgAAAKCaKEtO5MooAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFIjjAIAAAAgNcIoAAAAAFJTp7wPTJIkIiKWL19eYcUAAAAAUP0U50PFedHWlDuMKigoiIiIDh06lLcJAAAAAGqQgoKCyMvL2+oyWUlpIqvNKCoqijlz5kRubm5kZWWVq0Aq3vLly6NDhw4xe/bsaNy4cWWXAzWGcwt2HOcX7BjOLdgxnFuweUmSREFBQbRt2zZq1dr6qFDlvjKqVq1a0b59+/I+nB2scePGnhhhB3BuwY7j/IIdw7kFO4ZzCza1rSuiihnAHAAAAIDUCKMAAAAASI0wqoapV69ejBgxIurVq1fZpUCN4tyCHcf5BTuGcwt2DOcWbL9yD2AOAAAAAGXlyigAAAAAUiOMAgAAACA1wigAAAAAUiOMAgAAACA1wqhq6Je//GUccMABkZubGy1btoz+/fvHjBkzSiyzevXqGDZsWDRv3jwaNWoUJ510UsyfP7+SKobq6ZZbbomsrKy49NJLM9OcW1B+X375ZZxxxhnRvHnzyMnJiW7dusW0adMy85Mkieuvvz7atGkTOTk5ceSRR8Z//vOfSqwYqr4NGzbEddddF/n5+ZGTkxPf+ta34uc//3l8/TuKnFtQOq+++mqccMIJ0bZt28jKyoqnnnqqxPzSnEtLliyJwYMHR+PGjaNJkyZx7rnnRmFhYYprAdWDMKoaeuWVV2LYsGHx5ptvxoQJE2LdunXRr1+/WLFiRWaZyy67LP7617/GY489Fq+88krMmTMnBg4cWIlVQ/UyderUGDduXOy7774lpju3oHy++uqrOPjggyM7Ozuee+65mD59evzmN7+Jpk2bZpYZOXJk/O53v4uxY8fGW2+9FQ0bNoyjjz46Vq9eXYmVQ9V26623xh133BGjR4+Ojz76KG699dYYOXJkjBo1KrOMcwtKZ8WKFdG9e/cYM2bMZueX5lwaPHhw/Otf/4oJEybEM888E6+++mqcf/75aa0CVB8J1d6CBQuSiEheeeWVJEmSZOnSpUl2dnby2GOPZZb56KOPkohI3njjjcoqE6qNgoKCpHPnzsmECROSQw89NLnkkkuSJHFuwfb4yU9+khxyyCFbnF9UVJS0bt06+dWvfpWZtnTp0qRevXrJQw89lEaJUC0dd9xxyTnnnFNi2sCBA5PBgwcnSeLcgvKKiOTJJ5/M/F6ac2n69OlJRCRTp07NLPPcc88lWVlZyZdffpla7VAduDKqBli2bFlERDRr1iwiIt5+++1Yt25dHHnkkZll9txzz+jYsWO88cYblVIjVCfDhg2L4447rsQ5FOHcgu3x9NNPR69eveIHP/hBtGzZMnr06BF33XVXZv6nn34a8+bNK3F+5eXlxUEHHeT8gq3o06dPTJw4MT7++OOIiHjvvfdi8uTJceyxx0aEcwsqSmnOpTfeeCOaNGkSvXr1yixz5JFHRq1ateKtt95KvWaoyupUdgFsn6Kiorj00kvj4IMPjn322SciIubNmxd169aNJk2alFi2VatWMW/evEqoEqqPhx9+OP75z3/G1KlTN5nn3ILy++9//xt33HFHXH755XH11VfH1KlT4+KLL466devGkCFDMudQq1atSjzO+QVbd9VVV8Xy5ctjzz33jNq1a8eGDRvipptuisGDB0dEOLeggpTmXJo3b160bNmyxPw6depEs2bNnG/wDcKoam7YsGHx4YcfxuTJkyu7FKj2Zs+eHZdccklMmDAh6tevX9nlQI1SVFQUvXr1iptvvjkiInr06BEffvhhjB07NoYMGVLJ1UH19eijj8YDDzwQDz74YHTt2jXefffduPTSS6Nt27bOLQCqLLfpVWPDhw+PZ555JiZNmhTt27fPTG/dunWsXbs2li5dWmL5+fPnR+vWrVOuEqqPt99+OxYsWBD7779/1KlTJ+rUqROvvPJK/O53v4s6depEq1atnFtQTm3atIm99967xLS99torZs2aFRGROYe++e2Uzi/Yuh//+Mdx1VVXxaBBg6Jbt25x5plnxmWXXRa//OUvI8K5BRWlNOdS69atY8GCBSXmr1+/PpYsWeJ8g28QRlVDSZLE8OHD48knn4yXXnop8vPzS8zv2bNnZGdnx8SJEzPTZsyYEbNmzYrevXunXS5UG3379o0PPvgg3n333cy/Xr16xeDBgzM/O7egfA4++OCYMWNGiWkff/xxdOrUKSIi8vPzo3Xr1iXOr+XLl8dbb73l/IKtWLlyZdSqVfIlfe3ataOoqCginFtQUUpzLvXu3TuWLl0ab7/9dmaZl156KYqKiuKggw5KvWaoytymVw0NGzYsHnzwwRg/fnzk5uZm7j/Oy8uLnJycyMvLi3PPPTcuv/zyaNasWTRu3Dguuuii6N27d3z729+u5Oqh6srNzc2MvVasYcOG0bx588x05xaUz2WXXRZ9+vSJm2++OU455ZSYMmVK3HnnnXHnnXdGRERWVlZceuml8Ytf/CI6d+4c+fn5cd1110Xbtm2jf//+lVs8VGEnnHBC3HTTTdGxY8fo2rVrvPPOO/Hb3/42zjnnnIhwbkFZFBYWxieffJL5/dNPP4133303mjVrFh07dtzmubTXXnvFMcccE+edd16MHTs21q1bF8OHD49BgwZF27ZtK2mtoIqq7K/zo+wiYrP/7rnnnswyq1atSi688MKkadOmSYMGDZIBAwYkc+fOrbyioZo69NBDk0suuSTzu3MLyu+vf/1rss8++yT16tVL9txzz+TOO+8sMb+oqCi57rrrklatWiX16tVL+vbtm8yYMaOSqoXqYfny5ckll1ySdOzYMalfv36y2267Jddcc02yZs2azDLOLSidSZMmbfZ91pAhQ5IkKd25tHjx4uS0005LGjVqlDRu3Dg5++yzk4KCgkpYG6jaspIkSSopBwMAAABgJ2PMKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCAAAAIDXCKAAAAABSI4wCANiGoUOHRv/+/Su7DACAGqFOZRcAAFCZsrKytjp/xIgRcfvtt0eSJClVBABQswmjAICd2ty5czM/P/LII3H99dfHjBkzMtMaNWoUjRo1qozSAABqJLfpAQA7tdatW2f+5eXlRVZWVolpjRo12uQ2vcMOOywuuuiiuPTSS6Np06bRqlWruOuuu2LFihVx9tlnR25ubuy+++7x3HPPlejrww8/jGOPPTYaNWoUrVq1ijPPPDMWLVqU8hoDAFQuYRQAQDn8+c9/jl122SWmTJkSF110UVxwwQXxgx/8IPr06RP//Oc/o1+/fnHmmWfGypUrIyJi6dKlccQRR0SPHj1i2rRp8fzzz8f8+fPjlFNOqeQ1AQBIlzAKAKAcunfvHtdee2107tw5fvrTn0b9+vVjl112ifPOOy86d+4c119/fSxevDjef//9iIgYPXp09OjRI26++ebYc889o0ePHnH33XfHpEmT4uOPP67ktQEASI8xowAAymHffffN/Fy7du1o3rx5dOvWLTOtVatWERGxYMGCiIh47733YtKkSZsdf2rmzJnRpUuXHVwxAEDVIIwCACiH7OzsEr9nZWWVmFb8LX1FRUUREVFYWBgnnHBC3HrrrZu01aZNmx1YKQBA1SKMAgBIwf777x9PPPFE7LrrrlGnjpdgAMDOy5hRAAApGDZsWCxZsiROO+20mDp1asycOTNeeOGFOPvss2PDhg2VXR4AQGqEUQAAKWjbtm28/vrrsWHDhujXr19069YtLr300mjSpEnUquUlGQCw88hKkiSp7CIAAAAA2Dn4GA4AAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEiNMAoAAACA1AijAAAAAEjN/wf3SRajlsh/rQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "FOLDER = \"data/derived/tester\"\n",
    "\n",
    "for f in [file for file in os.listdir(FOLDER) if file.endswith(\".rttm\")]:\n",
    "    plt.figure(figsize=(15, 4))\n",
    "    plt.title(f)\n",
    "    notebook.reset()\n",
    "    notebook.plot_annotation(load_rttm(FOLDER + \"/\" +  f)[\"mic://localhost\"])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 4))\n",
    "notebook.reset()\n",
    "notebook.plot_annotation(load_rttm(\"data/derived/eval-streamlit.rttm\")[\"mic://localhost\"])\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": 2
}