comparison/pmnet_comparison.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "VTr3R3SqAkE7",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "collapsed": true,
        "id": "VTr3R3SqAkE7",
        "outputId": "0350da8e-0f3a-4c3e-bb5e-b5a5fa8516c6"
      },
      "outputs": [],
      "source": [
        "! pip install https://github.com/state-spaces/mamba/releases/download/v2.3.1/mamba_ssm-2.3.1+cu12torch2.10cxx11abiTRUE-cp312-cp312-linux_x86_64.whl\n",
        "! pip install https://github.com/Dao-AILab/causal-conv1d/releases/download/v1.6.1.post4/causal_conv1d-1.6.1+cu12torch2.10cxx11abiTRUE-cp312-cp312-linux_x86_64.whl\n",
        "! pip install --force transformers==4.57.6"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "e088c6c2",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e088c6c2",
        "outputId": "3fb57574-d805-4233-a9ef-65e4f5112789"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mon Mar 30 00:36:06 2026       \n",
            "+-----------------------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 580.82.07              Driver Version: 580.82.07      CUDA Version: 13.0     |\n",
            "+-----------------------------------------+------------------------+----------------------+\n",
            "| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |\n",
            "|                                         |                        |               MIG M. |\n",
            "|=========================================+========================+======================|\n",
            "|   0  NVIDIA A100-SXM4-40GB          Off |   00000000:00:04.0 Off |                    0 |\n",
            "| N/A   35C    P0             47W /  400W |       0MiB /  40960MiB |      0%      Default |\n",
            "|                                         |                        |             Disabled |\n",
            "+-----------------------------------------+------------------------+----------------------+\n",
            "\n",
            "+-----------------------------------------------------------------------------------------+\n",
            "| Processes:                                                                              |\n",
            "|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |\n",
            "|        ID   ID                                                               Usage      |\n",
            "|=========================================================================================|\n",
            "|  No running processes found                                                             |\n",
            "+-----------------------------------------------------------------------------------------+\n"
          ]
        }
      ],
      "source": [
        "! nvidia-smi"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "id": "d6c970d4",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 893,
          "referenced_widgets": [
            "718155cd08ad4645a6ab5ed9c40560a1",
            "cb9eccd3e1a845e68f6d19b08b584be5",
            "69e05d36c79f412e8e44b069b5b0d599",
            "2fb8a5922cd249c8a701897774d70013",
            "31bde5565e65472a8518bedd3e9902bf",
            "404f8191114f4ff2a6a43262916a61c4",
            "b735180998ee42c981708cef0ffb2e13",
            "846b7fee95ed4e92bd68d4d321b10ec0",
            "081ce52558254ddfb5fb6699e37627bf",
            "74a631a2775e440dabcb6308b37c44bb",
            "904d7de378aa4836b140b2541d1eb70e"
          ]
        },
        "id": "d6c970d4",
        "outputId": "ef1f0276-18fc-4b0c-9bf8-e0737a52b3a2"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Using device: cuda\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "718155cd08ad4645a6ab5ed9c40560a1",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Resolving data files:   0%|          | 0/23 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Number of tokenized test samples: 99\n",
            "Using 99 samples for evaluation\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 134.04it/s]\n",
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 135.00it/s]\n",
            "You are using a model of type pmnet to instantiate a model of type PMNet. This is not supported for all configurations of models and can yield errors.\n",
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 135.46it/s]\n",
            "You are using a model of type pmnet to instantiate a model of type PMNet. This is not supported for all configurations of models and can yield errors.\n",
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 134.35it/s]\n",
            "You are using a model of type pmnet to instantiate a model of type PMNet. This is not supported for all configurations of models and can yield errors.\n",
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 134.76it/s]\n",
            "You are using a model of type pmnet to instantiate a model of type PMNet. This is not supported for all configurations of models and can yield errors.\n",
            "Evaluating Samples: 100%|██████████| 99/99 [00:00<00:00, 133.44it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Graph saved to pmnet_long_context_bpb_comparison.pdf\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import json\n",
        "import os\n",
        "from pathlib import Path\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from datasets import load_dataset\n",
        "from safetensors.torch import load_file\n",
        "from tqdm import tqdm\n",
        "from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer\n",
        "\n",
        "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "torch.set_default_dtype(torch.float32)\n",
        "print(f\"Using device: {DEVICE}\")\n",
        "\n",
        "DATASET_ID = \"emozilla/pg19\"\n",
        "DATA_SPLIT = \"test\"\n",
        "TOKENIZER_ID = \"google/byt5-small\"\n",
        "\n",
        "PMNET_MODEL_ID = \"phasorkinetics/pmnet\"\n",
        "PMNET_CONFIG_PATH = \"config_pmnet.json\"\n",
        "PMNET_CKPT = \"model_pmnet.safetensors\"\n",
        "PMNET_LOSS_DIR = \"data/pmnet_losses\"\n",
        "\n",
        "PMNET_NO_MEM_CONFIG_PATH = \"config_pmnet_no_mem.json\"\n",
        "PMNET_NO_MEM_CKPT = \"model_pmnet_no_mem.safetensors\"\n",
        "PMNET_NO_MEM_LOSS_DIR = \"data/pmnet_no_mem_losses\"\n",
        "\n",
        "PMNET_NO_REC_LOSS_DIR = \"data/pmnet_no_rec_losses\"\n",
        "PMNET_NO_MEM_REC_LOSS_DIR = \"data/pmnet_no_mem_rec_losses\"\n",
        "\n",
        "MAMBA_MODEL_ID = \"state-spaces/mamba-130m-hf\"\n",
        "MAMBA_CONFIG_PATH = \"config_mamba.json\"\n",
        "MAMBA_CKPT = \"model_mamba.safetensors\"\n",
        "MAMBA_LOSS_DIR = \"data/mamba_losses\"\n",
        "\n",
        "SMOLLM_MODEL_ID = \"HuggingFaceTB/SmolLM-135M\"\n",
        "SMOLLM_CONFIG_PATH = \"config_smollm.json\"\n",
        "SMOLLM_CKPT = \"model_smollm.safetensors\"\n",
        "SMOLLM_LOSS_DIR = \"data/smollm_losses\"\n",
        "\n",
        "DATA_PROCESS_BATCH_SIZE = 1000\n",
        "NUM_PROC = 16\n",
        "NUM_EVAL_SAMPLES = 100\n",
        "SEQ_LENGTH_PMNET = 16 * 1024\n",
        "PLOT_SMOOTHING_WINDOW = 1000\n",
        "PLOT_SAVE_PATH = \"pmnet_long_context_bpb_comparison.pdf\"\n",
        "\n",
        "\n",
        "def loss_over_positions(model, input_ids):\n",
        "\tmodel.eval()\n",
        "\tdevice_local = next(model.parameters()).device\n",
        "\tcriterion = nn.CrossEntropyLoss(reduction=\"none\")\n",
        "\n",
        "\tinput_tensor = torch.tensor(\n",
        "\t\tinput_ids,\n",
        "\t\tdtype=torch.long,\n",
        "\t\tdevice=device_local,\n",
        "\t).unsqueeze(0)\n",
        "\n",
        "\twith torch.inference_mode():\n",
        "\t\toutputs = model(input_ids=input_tensor)\n",
        "\t\tshift_logits = outputs.logits[:, :-1, :].contiguous()\n",
        "\t\tshift_labels = input_tensor[:, 1:].contiguous()\n",
        "\t\tloss = criterion(\n",
        "\t\t\tshift_logits.view(-1, shift_logits.size(-1)),\n",
        "\t\t\tshift_labels.view(-1),\n",
        "\t\t)\n",
        "\n",
        "\tif torch.cuda.is_available():\n",
        "\t\ttorch.cuda.empty_cache()\n",
        "\n",
        "\treturn loss.float().cpu().numpy() / np.log(2)\n",
        "\n",
        "\n",
        "def prepare_dataset():\n",
        "\ttokenizer = AutoTokenizer.from_pretrained(TOKENIZER_ID, trust_remote_code=True)\n",
        "\ttest_dataset = load_dataset(DATASET_ID, split=DATA_SPLIT)\n",
        "\n",
        "\tdef process_batch(examples):\n",
        "\t\ttokenized = tokenizer(examples[\"text\"])\n",
        "\t\tinput_ids = tokenized[\"input_ids\"]\n",
        "\t\tinput_ids = [ids[:SEQ_LENGTH_PMNET] for ids in input_ids if len(ids) >= SEQ_LENGTH_PMNET]\n",
        "\t\treturn {\"input_ids\": input_ids}\n",
        "\n",
        "\ttest_dataset = test_dataset.map(\n",
        "\t\tprocess_batch,\n",
        "\t\tbatched=True,\n",
        "\t\tbatch_size=DATA_PROCESS_BATCH_SIZE,\n",
        "\t\tremove_columns=test_dataset.column_names,\n",
        "\t\tnum_proc=NUM_PROC,\n",
        "\t\tdesc=\"Tokenizing\",\n",
        "\t)\n",
        "\n",
        "\tprint(f\"Number of tokenized test samples: {len(test_dataset)}\")\n",
        "\tif NUM_EVAL_SAMPLES:\n",
        "\t\ttest_dataset = test_dataset.select(range(min(NUM_EVAL_SAMPLES, len(test_dataset))))\n",
        "\t\tprint(f\"Using {len(test_dataset)} samples for evaluation\")\n",
        "\n",
        "\treturn test_dataset\n",
        "\n",
        "\n",
        "def load_model_config(model_id: str, config_path=None):\n",
        "\tmodel_config = AutoConfig.from_pretrained(model_id, trust_remote_code=True)\n",
        "\n",
        "\tif config_path is not None:\n",
        "\t\tif not os.path.exists(config_path):\n",
        "\t\t\traise FileNotFoundError(f\"Config file not found: {config_path}\")\n",
        "\t\twith open(config_path, \"r\", encoding=\"utf-8\") as f:\n",
        "\t\t\tmodel_config.update(json.load(f))\n",
        "\n",
        "\t# Force float32 for numerically stable validation comparisons.\n",
        "\tmodel_config.torch_dtype = torch.float32\n",
        "\n",
        "\treturn model_config\n",
        "\n",
        "\n",
        "def load_model_from_safetensors(model_id: str, ckpt: str, config_path=None):\n",
        "\tmodel_config = load_model_config(model_id, config_path=config_path)\n",
        "\tmodel = AutoModelForCausalLM.from_config(model_config, trust_remote_code=True)\n",
        "\tstate_dict = load_file(ckpt)\n",
        "\tincompatible_keys = model.load_state_dict(state_dict, strict=False)\n",
        "\n",
        "\t# if incompatible_keys.missing_keys or incompatible_keys.unexpected_keys:\n",
        "\t# \traise RuntimeError(\n",
        "\t# \t\t\"Checkpoint load mismatch\\n\"\n",
        "\t# \t\tf\"missing_keys={incompatible_keys.missing_keys}\\n\"\n",
        "\t# \t\tf\"unexpected_keys={incompatible_keys.unexpected_keys}\"\n",
        "\t# \t)\n",
        "\n",
        "\tmodel.to(device=DEVICE, dtype=torch.float32)\n",
        "\tmodel.float()\n",
        "\tmodel.eval()\n",
        "\tif next(model.parameters()).dtype != torch.float32:\n",
        "\t\traise RuntimeError(f\"Model is not float32: {next(model.parameters()).dtype}\")\n",
        "\treturn model\n",
        "\n",
        "\n",
        "def evaluate(model_id: str, ckpt: str, dataset, losses_dir: str, config_path=None):\n",
        "\tPath(losses_dir).mkdir(parents=True, exist_ok=True)\n",
        "\tmodel = load_model_from_safetensors(model_id, ckpt, config_path=config_path)\n",
        "\tlosses_list = []\n",
        "\n",
        "\tfor i, sample in enumerate(tqdm(dataset, desc=\"Evaluating Samples\", total=len(dataset))):\n",
        "\t\tlosses_path = f\"{losses_dir}/sample_{i}_losses.npy\"\n",
        "\t\tif os.path.exists(losses_path):\n",
        "\t\t\tlosses = np.load(losses_path)\n",
        "\t\t\tlosses_list.append(losses)\n",
        "\t\t\tcontinue\n",
        "\n",
        "\t\tlosses = loss_over_positions(model, sample[\"input_ids\"])\n",
        "\t\tnp.save(losses_path, losses)\n",
        "\t\tlosses_list.append(losses)\n",
        "\n",
        "\treturn np.array(losses_list).mean(axis=0)\n",
        "\n",
        "\n",
        "def moving_average(a, n=PLOT_SMOOTHING_WINDOW):\n",
        "\tn = max(1, min(n, len(a)))\n",
        "\tret = np.cumsum(a, dtype=float)\n",
        "\tret[n:] = ret[n:] - ret[:-n]\n",
        "\treturn ret[n - 1 :] / n\n",
        "\n",
        "\n",
        "def smooth_curve(losses, window=PLOT_SMOOTHING_WINDOW):\n",
        "\twindow = max(1, min(window, len(losses)))\n",
        "\tsmoothed = moving_average(losses, window)\n",
        "\tx = np.arange(window - 1, window - 1 + len(smoothed))\n",
        "\treturn x, smoothed\n",
        "\n",
        "\n",
        "def main():\n",
        "\ttest_dataset = prepare_dataset()\n",
        "\n",
        "\tmamba_losses = evaluate(\n",
        "\t\tMAMBA_MODEL_ID,\n",
        "\t\tMAMBA_CKPT,\n",
        "\t\ttest_dataset,\n",
        "\t\tMAMBA_LOSS_DIR,\n",
        "\t\tconfig_path=MAMBA_CONFIG_PATH,\n",
        "\t)\n",
        "\tsmol_losses = evaluate(\n",
        "\t\tSMOLLM_MODEL_ID,\n",
        "\t\tSMOLLM_CKPT,\n",
        "\t\ttest_dataset,\n",
        "\t\tSMOLLM_LOSS_DIR,\n",
        "\t\tconfig_path=SMOLLM_CONFIG_PATH,\n",
        "\t)\n",
        "\tpmnet_losses = evaluate(\n",
        "\t\tPMNET_MODEL_ID,\n",
        "\t\tPMNET_CKPT,\n",
        "\t\ttest_dataset,\n",
        "\t\tPMNET_LOSS_DIR,\n",
        "\t\tconfig_path=PMNET_CONFIG_PATH,\n",
        "\t)\n",
        "\tpmnet_no_mem_losses = evaluate(\n",
        "\t\tPMNET_MODEL_ID,\n",
        "\t\tPMNET_NO_MEM_CKPT,\n",
        "\t\ttest_dataset,\n",
        "\t\tPMNET_NO_MEM_LOSS_DIR,\n",
        "\t\tconfig_path=PMNET_NO_MEM_CONFIG_PATH,\n",
        "\t)\n",
        "\tpmnet_no_rec_losses = evaluate(\n",
        "\t\t\tPMNET_MODEL_ID,\n",
        "\t\t\tPMNET_CKPT,\n",
        "\t\t\ttest_dataset,\n",
        "\t\t\tPMNET_NO_REC_LOSS_DIR,\n",
        "\t\t\tconfig_path=PMNET_NO_MEM_CONFIG_PATH,\n",
        "\t)\n",
        "\tpmnet_no_mem_rec_losses = evaluate(\n",
        "\t\t\tPMNET_MODEL_ID,\n",
        "\t\t\tPMNET_NO_MEM_CKPT,\n",
        "\t\t\ttest_dataset,\n",
        "\t\t\tPMNET_NO_MEM_REC_LOSS_DIR,\n",
        "\t\t\tconfig_path=PMNET_CONFIG_PATH,\n",
        "\t)\n",
        "\n",
        "\tmamba_x, mamba_smooth = smooth_curve(mamba_losses)\n",
        "\tsmol_x, smol_smooth = smooth_curve(smol_losses)\n",
        "\tx_pmnet, smooth_on = smooth_curve(pmnet_losses)\n",
        "\t_, smooth_off = smooth_curve(pmnet_no_mem_losses)\n",
        "\t_, smooth_no_rec = smooth_curve(pmnet_no_rec_losses)\n",
        "\t_, smooth_no_mem_rec = smooth_curve(pmnet_no_mem_rec_losses)\n",
        "\n",
        "\t# smol_config = load_model_config(SMOLLM_MODEL_ID, SMOLLM_CONFIG_PATH)\n",
        "\t# smol_ctx = smol_config.max_position_embeddings\n",
        "\n",
        "\tplt.figure(figsize=(12, 6))\n",
        "\tplt.plot(\n",
        "\t\tx_pmnet,\n",
        "\t\tsmooth_on,\n",
        "\t\tlabel=\"PMNet\",\n",
        "\t\tcolor=\"green\",\n",
        "\t\talpha=1,\n",
        "\t)\n",
        "\tplt.plot(mamba_x, mamba_smooth, label=\"Mamba\", color=\"blue\", alpha=0.6)\n",
        "\tplt.plot(smol_x, smol_smooth, label=\"SmolLM\", color=\"grey\", alpha=0.6)\n",
        "\tplt.plot(\n",
        "\t\tx_pmnet,\n",
        "\t\tsmooth_no_rec,\n",
        "\t\tlabel=\"PMNet No Recurrence\",\n",
        "\t\tcolor=\"orange\",\n",
        "\t\talpha=1,\n",
        "\t\tlinestyle=\"--\",\n",
        "\t)\n",
        "\tplt.plot(\n",
        "\t\tx_pmnet,\n",
        "\t\tsmooth_off,\n",
        "\t\tlabel=\"PMNet No Mememory / No Recurrence\",\n",
        "\t\tcolor=\"red\",\n",
        "\t\talpha=1,\n",
        "\t\tlinestyle=\"--\",\n",
        "\t)\n",
        "\tplt.plot(\n",
        "\t\tx_pmnet,\n",
        "\t\tsmooth_no_mem_rec,\n",
        "\t\tlabel=\"PMNet No Mememory / Recurrence\",\n",
        "\t\tcolor=\"purple\",\n",
        "\t\talpha=1,\n",
        "\t\tlinestyle=\"--\",\n",
        "\t)\n",
        "\n",
        "\t# approx_byte_limit = smol_ctx * 4\n",
        "\t# plt.axvline(\n",
        "\t# \tx=approx_byte_limit,\n",
        "\t# \tcolor=\"gray\",\n",
        "\t# \tlinestyle=\"--\",\n",
        "\t# \tlabel=f\"SmolLM\",\n",
        "\t# )\n",
        "\n",
        "\tplt.title(\"Long Context BPB Comparison\\nData: PG19 Test\", fontsize=20, fontweight=\"bold\")\n",
        "\tplt.xlabel(\"Position in Bytes\", fontsize=18)\n",
        "\tplt.ylabel(\"Bits Per Byte\", fontsize=18)\n",
        "\tplt.yscale(\"log\")\n",
        "\tplt.xticks(fontsize=16)\n",
        "\tplt.yticks(fontsize=16)\n",
        "\n",
        "\t# handles, labels = plt.gca().get_legend_handles_labels()\n",
        "\t# order = [\n",
        "\t# ]\n",
        "\t# index = [labels.index(label) for label in order]\n",
        "\t# plt.legend([handles[i] for i in index], [labels[i] for i in index], fontsize=16)\n",
        "\tplt.ylim(top=5.5)\n",
        "\tplt.legend()\n",
        "\tplt.grid(True, alpha=0.3)\n",
        "\t# plt.ylim(top=4.0)\n",
        "\tplt.tight_layout()\n",
        "\tplt.savefig(PLOT_SAVE_PATH)\n",
        "\tprint(f\"Graph saved to {PLOT_SAVE_PATH}\")\n",
        "\n",
        "\n",
        "main()\n"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "A100",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.11.13"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "081ce52558254ddfb5fb6699e37627bf": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "2fb8a5922cd249c8a701897774d70013": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_74a631a2775e440dabcb6308b37c44bb",
            "placeholder": "​",
            "style": "IPY_MODEL_904d7de378aa4836b140b2541d1eb70e",
            "value": " 23/23 [00:00&lt;00:00, 2988.51it/s]"
          }
        },
        "31bde5565e65472a8518bedd3e9902bf": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "404f8191114f4ff2a6a43262916a61c4": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "69e05d36c79f412e8e44b069b5b0d599": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_846b7fee95ed4e92bd68d4d321b10ec0",
            "max": 23,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_081ce52558254ddfb5fb6699e37627bf",
            "value": 23
          }
        },
        "718155cd08ad4645a6ab5ed9c40560a1": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_cb9eccd3e1a845e68f6d19b08b584be5",
              "IPY_MODEL_69e05d36c79f412e8e44b069b5b0d599",
              "IPY_MODEL_2fb8a5922cd249c8a701897774d70013"
            ],
            "layout": "IPY_MODEL_31bde5565e65472a8518bedd3e9902bf"
          }
        },
        "74a631a2775e440dabcb6308b37c44bb": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "846b7fee95ed4e92bd68d4d321b10ec0": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "904d7de378aa4836b140b2541d1eb70e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b735180998ee42c981708cef0ffb2e13": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cb9eccd3e1a845e68f6d19b08b584be5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_404f8191114f4ff2a6a43262916a61c4",
            "placeholder": "​",
            "style": "IPY_MODEL_b735180998ee42c981708cef0ffb2e13",
            "value": "Resolving data files: 100%"
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}