example_usage.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import h5py\n",
    "\n",
    "h5filename = '/home/yxwang/Dataset/vqhlm/wikitext103_gpt2finetuned/test.h5'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 数据操作"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ChunkedDataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input IDs: torch.Size([128, 1024])\n",
      "Labels: torch.Size([128, 1024])\n",
      "Hidden States: torch.Size([128, 256, 768])\n"
     ]
    }
   ],
   "source": [
    "from dataloading import get_chunked_h5dataloader\n",
    "from constants import KEY_LM_INPUT_IDS, KEY_LM_HIDDEN_STATES, KEY_LM_LABELS\n",
    "\n",
    "dataloader = get_chunked_h5dataloader(\n",
    "    config_path='conf/data/example.yaml',\n",
    "    split='test',\n",
    ")\n",
    "\n",
    "for batch in dataloader:\n",
    "    input_ids = batch[KEY_LM_INPUT_IDS]\n",
    "    labels = batch[KEY_LM_LABELS]\n",
    "    hidden_states = batch[KEY_LM_HIDDEN_STATES]\n",
    "    \n",
    "    print(f\"Input IDs: {input_ids.shape}\")\n",
    "    print(f\"Labels: {labels.shape}\")\n",
    "    print(f\"Hidden States: {hidden_states.shape}\")\n",
    "    break  # 这里只打印一个批次的数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.60 :0.8160842776298523\n",
      "0.62 :0.7752801060676575\n",
      "0.64 :0.7344759583473206\n",
      "0.66 :0.6936716437339783\n",
      "0.68 :0.6528674244880677\n",
      "0.70 :0.6120632886886597\n",
      "0.72 :0.5712589859962464\n",
      "0.74 :0.5304548025131226\n",
      "0.76 :0.4896506190299988\n",
      "0.78 :0.448846435546875\n",
      "0.80 :0.4080421268939972\n",
      "0.82 :0.3672379791736603\n",
      "0.84 :0.32643377780914307\n",
      "0.86 :0.2856294929981232\n",
      "0.88 :0.2448253095149994\n",
      "0.90 :0.20402113497257232\n",
      "0.92 :0.16321683526039124\n",
      "0.94 :0.1224126547574997\n",
      "0.96 :0.0816084772348404\n",
      "0.98 :0.040804176777601245\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def calculate_dataset_stats(dataloader, ratio=0.7):\n",
    "    total_loss = 0\n",
    "    # 遍历整个数据集并收集所有的 hidden_states\n",
    "    for i, batch in enumerate(dataloader):\n",
    "        hidden_states = batch[KEY_LM_HIDDEN_STATES]\n",
    "        out = ratio * hidden_states\n",
    "        # out = out.clamp(-1., 1.)\n",
    "        rec_loss = (out - hidden_states).abs().mean()\n",
    "        total_loss += rec_loss.item()\n",
    "    total_loss /= len(dataloader)\n",
    "    return total_loss    \n",
    "\n",
    "# 示例用法：\n",
    "for ratio in range(60, 100, 2):\n",
    "    ratio = 0.01*ratio\n",
    "    total_loss = calculate_dataset_stats(dataloader, ratio)\n",
    "    print(\"%.2f\"%(ratio), f\":{total_loss}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## H5py直接操作"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 模型来自 https://huggingface.co/neulab/gpt2-finetuned-wikitext103\n",
    "- 模型tokenizer和定义如下\n",
    "```\n",
    "# Load model directly\n",
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"neulab/gpt2-finetuned-wikitext103\")\n",
    "model = AutoModelForCausalLM.from_pretrained(\"neulab/gpt2-finetuned-wikitext103\")\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total samples in the dataset: 548\n",
      "Hidden states shape: (548, 1024, 768)\n",
      "Input IDs shape: (548, 1024)\n",
      "Labels shape: (548, 1024)\n",
      "Sample 0 - Input IDs: [  796  5199   347 ... 50256 50256 50256]\n",
      "Sample 0 - Labels: [ 796 5199  347 ... -100 -100 -100]\n",
      "Sample 0 - Hidden States: [[-1.6174933  -1.1284641   0.0215988  ... -2.2583933   1.6741287\n",
      "  -0.35499385]\n",
      " [-5.498198    2.3329258  -2.2865224  ...  0.42031875  0.07713613\n",
      "   1.1435596 ]\n",
      " [ 1.8659987   1.5455958   1.0896978  ...  3.514292   -0.08421862\n",
      "   3.3854916 ]\n",
      " ...\n",
      " [ 0.798916    2.5979886  -0.07587534 ... -0.737859   -1.1983643\n",
      "   4.190173  ]\n",
      " [ 0.7338563   2.760131   -0.01490831 ... -0.6301186  -1.1114419\n",
      "   4.6592402 ]\n",
      " [ 0.7557825   1.4493941  -3.1341372  ... -2.043258    0.9624741\n",
      "   5.8210287 ]]\n"
     ]
    }
   ],
   "source": [
    "# 读取 HDF5 文件\n",
    "with h5py.File(h5filename, 'r') as f:\n",
    "    # 读取 total_samples 元数据\n",
    "    total_samples = f.attrs['total_samples']\n",
    "    print(f\"Total samples in the dataset: {total_samples}\")\n",
    "    \n",
    "    # 读取数据集\n",
    "    hidden_states = f['hidden_states'][:]\n",
    "    input_ids = f['input_ids'][:]\n",
    "    labels = f['labels'][:]\n",
    "\n",
    "    # 打印数据集的形状，确保读取正确\n",
    "    print(f\"Hidden states shape: {hidden_states.shape}\")\n",
    "    print(f\"Input IDs shape: {input_ids.shape}\")\n",
    "    print(f\"Labels shape: {labels.shape}\")\n",
    "    \n",
    "    # 获取第一个样本的数据并展示\n",
    "    sample_idx = 0  # 举例，读取第一个样本\n",
    "    print(f\"Sample {sample_idx} - Input IDs: {input_ids[sample_idx]}\")\n",
    "    print(f\"Sample {sample_idx} - Labels: {labels[sample_idx]}\")\n",
    "    \n",
    "    # 获取对应的 hidden states (flattened)\n",
    "    print(f\"Sample {sample_idx} - Hidden States: {hidden_states[sample_idx]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 预览index frequencies（需要已经跑完test）"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data loaded from CSV: {0: 0, 1: 0, 2: 0, 3: 0, 4: 0, 5: 1, 6: 3, 7: 0, 8: 1, 9: 1, 10: 0, 11: 2, 12: 0, 13: 0, 14: 0, 15: 0, 16: 1, 17: 0, 18: 1, 19: 2, 20: 0, 21: 2, 22: 0, 23: 0, 24: 0, 25: 0, 26: 0, 27: 0, 28: 0, 29: 1, 30: 0, 31: 0, 32: 0, 33: 0, 34: 4, 35: 1, 36: 1, 37: 0, 38: 0, 39: 0, 40: 1, 41: 0, 42: 0, 43: 0, 44: 0, 45: 0, 46: 0, 47: 0, 48: 1, 49: 1, 50: 0, 51: 3, 52: 0, 53: 1, 54: 0, 55: 0, 56: 2, 57: 1, 58: 0, 59: 0, 60: 0, 61: 0, 62: 0, 63: 0, 64: 0, 65: 1, 66: 0, 67: 1, 68: 0, 69: 0, 70: 0, 71: 3, 72: 1, 73: 1, 74: 0, 75: 0, 76: 0, 77: 1, 78: 0, 79: 0, 80: 0, 81: 1, 82: 1, 83: 1, 84: 0, 85: 0, 86: 2, 87: 0, 88: 0, 89: 0, 90: 0, 91: 2, 92: 0, 93: 1, 94: 1, 95: 0, 96: 1, 97: 0, 98: 3, 99: 0, 100: 2, 101: 4, 102: 0, 103: 0, 104: 2, 105: 0, 106: 0, 107: 0, 108: 0, 109: 0, 110: 0, 111: 0, 112: 0, 113: 0, 114: 0, 115: 1, 116: 0, 117: 0, 118: 0, 119: 1, 120: 0, 121: 0, 122: 0, 123: 0, 124: 2, 125: 0, 126: 2, 127: 1, 128: 0, 129: 0, 130: 1, 131: 1, 132: 0, 133: 0, 134: 0, 135: 1, 136: 0, 137: 0, 138: 0, 139: 0, 140: 0, 141: 1, 142: 1, 143: 0, 144: 1, 145: 0, 146: 0, 147: 0, 148: 0, 149: 1, 150: 0, 151: 0, 152: 0, 153: 0, 154: 0, 155: 0, 156: 0, 157: 1, 158: 0, 159: 0, 160: 4, 161: 0, 162: 0, 163: 0, 164: 0, 165: 0, 166: 0, 167: 0, 168: 0, 169: 0, 170: 0, 171: 1, 172: 0, 173: 0, 174: 0, 175: 0, 176: 0, 177: 1, 178: 0, 179: 2, 180: 0, 181: 1, 182: 2, 183: 0, 184: 0, 185: 0, 186: 0, 187: 0, 188: 3, 189: 1, 190: 0, 191: 0, 192: 2, 193: 0, 194: 0, 195: 0, 196: 2, 197: 0, 198: 1, 199: 0, 200: 1, 201: 0, 202: 0, 203: 0, 204: 1, 205: 2, 206: 0, 207: 0, 208: 0, 209: 1, 210: 0, 211: 0, 212: 2, 213: 0, 214: 0, 215: 0, 216: 1, 217: 1, 218: 2, 219: 3, 220: 0, 221: 0, 222: 0, 223: 1, 224: 1, 225: 0, 226: 0, 227: 0, 228: 1, 229: 0, 230: 0, 231: 0, 232: 0, 233: 0, 234: 0, 235: 0, 236: 0, 237: 0, 238: 0, 239: 0, 240: 0, 241: 0, 242: 1, 243: 0, 244: 0, 245: 0, 246: 0, 247: 0, 248: 0, 249: 0, 250: 0, 251: 0, 252: 6, 253: 1, 254: 1, 255: 0, 256: 0, 257: 0, 258: 0, 259: 1, 260: 1, 261: 0, 262: 1, 263: 0, 264: 0, 265: 0, 266: 1, 267: 0, 268: 0, 269: 0, 270: 1, 271: 0, 272: 0, 273: 0, 274: 0, 275: 0, 276: 0, 277: 2, 278: 0, 279: 3, 280: 0, 281: 0, 282: 0, 283: 0, 284: 0, 285: 0, 286: 0, 287: 0, 288: 0, 289: 0, 290: 3, 291: 0, 292: 1, 293: 0, 294: 0, 295: 0, 296: 1, 297: 2, 298: 0, 299: 2, 300: 1, 301: 1, 302: 0, 303: 1, 304: 0, 305: 0, 306: 0, 307: 0, 308: 0, 309: 0, 310: 0, 311: 1, 312: 0, 313: 0, 314: 0, 315: 0, 316: 0, 317: 0, 318: 1, 319: 2, 320: 1, 321: 0, 322: 0, 323: 0, 324: 2, 325: 1, 326: 0, 327: 1, 328: 0, 329: 0, 330: 0, 331: 0, 332: 0, 333: 0, 334: 0, 335: 1, 336: 0, 337: 0, 338: 4, 339: 0, 340: 0, 341: 0, 342: 1, 343: 1, 344: 0, 345: 0, 346: 0, 347: 0, 348: 0, 349: 0, 350: 3, 351: 0, 352: 0, 353: 2, 354: 0, 355: 0, 356: 0, 357: 1, 358: 0, 359: 0, 360: 0, 361: 1, 362: 0, 363: 0, 364: 4, 365: 0, 366: 0, 367: 0, 368: 0, 369: 0, 370: 1, 371: 0, 372: 4, 373: 0, 374: 1, 375: 2, 376: 0, 377: 1, 378: 0, 379: 1, 380: 0, 381: 0, 382: 0, 383: 0, 384: 0, 385: 0, 386: 0, 387: 1, 388: 0, 389: 0, 390: 1, 391: 4, 392: 0, 393: 2, 394: 0, 395: 0, 396: 0, 397: 0, 398: 0, 399: 0, 400: 0, 401: 2, 402: 0, 403: 0, 404: 1, 405: 0, 406: 0, 407: 0, 408: 0, 409: 1, 410: 1, 411: 0, 412: 0, 413: 0, 414: 0, 415: 0, 416: 0, 417: 0, 418: 1, 419: 0, 420: 5, 421: 0, 422: 2, 423: 1, 424: 0, 425: 0, 426: 0, 427: 0, 428: 0, 429: 0, 430: 1, 431: 0, 432: 1, 433: 0, 434: 0, 435: 1, 436: 1, 437: 0, 438: 0, 439: 4, 440: 1, 441: 0, 442: 0, 443: 3, 444: 0, 445: 1, 446: 0, 447: 0, 448: 1, 449: 1, 450: 2, 451: 0, 452: 1, 453: 0, 454: 0, 455: 0, 456: 1, 457: 1, 458: 0, 459: 0, 460: 0, 461: 2, 462: 2, 463: 0, 464: 0, 465: 1, 466: 0, 467: 1, 468: 0, 469: 0, 470: 0, 471: 1, 472: 3, 473: 1, 474: 0, 475: 0, 476: 0, 477: 0, 478: 0, 479: 1, 480: 0, 481: 1, 482: 0, 483: 0, 484: 0, 485: 0, 486: 4, 487: 1, 488: 0, 489: 0, 490: 0, 491: 0, 492: 0, 493: 0, 494: 1, 495: 0, 496: 3, 497: 0, 498: 0, 499: 1, 500: 0, 501: 0, 502: 0, 503: 1, 504: 1, 505: 0, 506: 1, 507: 0, 508: 0, 509: 1, 510: 0, 511: 1, 512: 0, 513: 0, 514: 0, 515: 0, 516: 0, 517: 0, 518: 0, 519: 1, 520: 0, 521: 0, 522: 1, 523: 1, 524: 0, 525: 0, 526: 0, 527: 0, 528: 0, 529: 0, 530: 0, 531: 0, 532: 0, 533: 0, 534: 0, 535: 0, 536: 0, 537: 0, 538: 0, 539: 0, 540: 0, 541: 0, 542: 0, 543: 2, 544: 0, 545: 1, 546: 0, 547: 1, 548: 1, 549: 0, 550: 2, 551: 0, 552: 0, 553: 1, 554: 0, 555: 0, 556: 2, 557: 2, 558: 0, 559: 1, 560: 0, 561: 1, 562: 1, 563: 2, 564: 1, 565: 1, 566: 0, 567: 1, 568: 1, 569: 0, 570: 3, 571: 2, 572: 0, 573: 0, 574: 0, 575: 4, 576: 1, 577: 0, 578: 0, 579: 1, 580: 0, 581: 2, 582: 2, 583: 0, 584: 0, 585: 4, 586: 0, 587: 0, 588: 1, 589: 1, 590: 0, 591: 0, 592: 2, 593: 0, 594: 0, 595: 1, 596: 0, 597: 0, 598: 1, 599: 1, 600: 1, 601: 0, 602: 1, 603: 2, 604: 0, 605: 4, 606: 1, 607: 1, 608: 0, 609: 0, 610: 2, 611: 0, 612: 1, 613: 0, 614: 0, 615: 0, 616: 0, 617: 1, 618: 0, 619: 0, 620: 1, 621: 0, 622: 0, 623: 0, 624: 0, 625: 0, 626: 1, 627: 2, 628: 0, 629: 0, 630: 0, 631: 0, 632: 0, 633: 1, 634: 0, 635: 0, 636: 1, 637: 0, 638: 1, 639: 0, 640: 1, 641: 0, 642: 2, 643: 1, 644: 0, 645: 1, 646: 2, 647: 2, 648: 1, 649: 0, 650: 1, 651: 0, 652: 0, 653: 0, 654: 1, 655: 1, 656: 0, 657: 1, 658: 1, 659: 0, 660: 0, 661: 2, 662: 2, 663: 0, 664: 2, 665: 0, 666: 1, 667: 0, 668: 0, 669: 1, 670: 0, 671: 0, 672: 0, 673: 0, 674: 2, 675: 0, 676: 1, 677: 0, 678: 0, 679: 3, 680: 2, 681: 1, 682: 0, 683: 0, 684: 2, 685: 0, 686: 0, 687: 0, 688: 0, 689: 0, 690: 1, 691: 0, 692: 1, 693: 0, 694: 1, 695: 0, 696: 3, 697: 0, 698: 1, 699: 0, 700: 0, 701: 0, 702: 1, 703: 0, 704: 0, 705: 0, 706: 0, 707: 1, 708: 1, 709: 0, 710: 2, 711: 0, 712: 0, 713: 1, 714: 2, 715: 0, 716: 0, 717: 0, 718: 0, 719: 2, 720: 0, 721: 0, 722: 2, 723: 1, 724: 1, 725: 2, 726: 1, 727: 0, 728: 1, 729: 0, 730: 0, 731: 0, 732: 0, 733: 0, 734: 2, 735: 0, 736: 0, 737: 0, 738: 0, 739: 0, 740: 0, 741: 1, 742: 1, 743: 0, 744: 0, 745: 0, 746: 3, 747: 2, 748: 1, 749: 0, 750: 0, 751: 0, 752: 0, 753: 0, 754: 0, 755: 1, 756: 1, 757: 0, 758: 0, 759: 0, 760: 0, 761: 0, 762: 0, 763: 1, 764: 1, 765: 0, 766: 2, 767: 1, 768: 0, 769: 0, 770: 0, 771: 0, 772: 0, 773: 0, 774: 2, 775: 0, 776: 0, 777: 0, 778: 1, 779: 0, 780: 0, 781: 2, 782: 1, 783: 0, 784: 0, 785: 1, 786: 0, 787: 0, 788: 0, 789: 1, 790: 1, 791: 0, 792: 1, 793: 0, 794: 1, 795: 0, 796: 1, 797: 4, 798: 0, 799: 1, 800: 0, 801: 0, 802: 0, 803: 0, 804: 2, 805: 0, 806: 0, 807: 1, 808: 0, 809: 1, 810: 0, 811: 0, 812: 1, 813: 0, 814: 0, 815: 1, 816: 0, 817: 0, 818: 0, 819: 0, 820: 0, 821: 0, 822: 3, 823: 0, 824: 0, 825: 0, 826: 0, 827: 0, 828: 3, 829: 1, 830: 1, 831: 2, 832: 0, 833: 0, 834: 0, 835: 0, 836: 3, 837: 1, 838: 0, 839: 0, 840: 0, 841: 0, 842: 0, 843: 1, 844: 1, 845: 0, 846: 0, 847: 0, 848: 1, 849: 0, 850: 3, 851: 3, 852: 0, 853: 2, 854: 0, 855: 0, 856: 1, 857: 0, 858: 0, 859: 0, 860: 1, 861: 0, 862: 0, 863: 0, 864: 0, 865: 0, 866: 0, 867: 0, 868: 0, 869: 1, 870: 0, 871: 4, 872: 1, 873: 0, 874: 0, 875: 1, 876: 0, 877: 2, 878: 0, 879: 0, 880: 1, 881: 0, 882: 1, 883: 0, 884: 2, 885: 0, 886: 0, 887: 0, 888: 0, 889: 0, 890: 1, 891: 0, 892: 0, 893: 1, 894: 0, 895: 0, 896: 0, 897: 1, 898: 0, 899: 0, 900: 0, 901: 0, 902: 0, 903: 0, 904: 2, 905: 1, 906: 0, 907: 2, 908: 1, 909: 1, 910: 2, 911: 2, 912: 1, 913: 2, 914: 1, 915: 0, 916: 0, 917: 1, 918: 0, 919: 0, 920: 0, 921: 1, 922: 0, 923: 0, 924: 0, 925: 0, 926: 0, 927: 1, 928: 0, 929: 1, 930: 1, 931: 0, 932: 0, 933: 0, 934: 3, 935: 0, 936: 0, 937: 0, 938: 0, 939: 0, 940: 0, 941: 0, 942: 1, 943: 1, 944: 0, 945: 1, 946: 1, 947: 0, 948: 1, 949: 1, 950: 1, 951: 0, 952: 0, 953: 0, 954: 0, 955: 1, 956: 0, 957: 0, 958: 0, 959: 4, 960: 0, 961: 0, 962: 0, 963: 0, 964: 5, 965: 1, 966: 0, 967: 0, 968: 0, 969: 0, 970: 2, 971: 7, 972: 1, 973: 2, 974: 3, 975: 0, 976: 0, 977: 0, 978: 1, 979: 0, 980: 0, 981: 0, 982: 0, 983: 1, 984: 0, 985: 1, 986: 0, 987: 1, 988: 0, 989: 1, 990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 3, 998: 0, 999: 1, 1000: 0, 1001: 2, 1002: 0, 1003: 0, 1004: 0, 1005: 3, 1006: 4, 1007: 0, 1008: 0, 1009: 0, 1010: 0, 1011: 0, 1012: 0, 1013: 0, 1014: 4, 1015: 0, 1016: 0, 1017: 1, 1018: 1, 1019: 1, 1020: 1, 1021: 1, 1022: 0, 1023: 2}\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "\n",
    "# 读取CSV文件并加载到字典中\n",
    "index_count_from_csv = {}\n",
    "\n",
    "with open('exp/1209_lowdim_CBDIM16/index_frequencies_0.csv', mode='r') as file:\n",
    "    reader = csv.reader(file)\n",
    "    next(reader)  # 跳过表头\n",
    "    for row in reader:\n",
    "        index = int(row[0])\n",
    "        frequency = int(row[1])\n",
    "        index_count_from_csv[index] = frequency\n",
    "\n",
    "print(\"Data loaded from CSV:\", index_count_from_csv)\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 提取所有索引的出现次数\n",
    "frequencies_from_csv = list(index_count_from_csv.values())\n",
    "\n",
    "# 绘制直方图\n",
    "plt.bar(range(len(frequencies_from_csv)), frequencies_from_csv, edgecolor='black', alpha=0.7)\n",
    "plt.title(\"Frequency of Codebook Indices (From CSV Data)\")\n",
    "plt.xlabel(\"Codebook Index\")\n",
    "plt.ylabel(\"Frequency\")\n",
    "plt.xticks(range(0, len(frequencies_from_csv), 50))  # 可调整间隔，防止标签过密\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "vqhlm",
   "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}