kaggle.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0b36dc7e",
   "metadata": {
    "papermill": {
     "duration": 0.004511,
     "end_time": "2024-01-31T10:29:04.799333",
     "exception": false,
     "start_time": "2024-01-31T10:29:04.794822",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## 1. 处理数据集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ca9f21ce",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:04.809454Z",
     "iopub.status.busy": "2024-01-31T10:29:04.809091Z",
     "iopub.status.idle": "2024-01-31T10:29:11.932263Z",
     "shell.execute_reply": "2024-01-31T10:29:11.931237Z"
    },
    "papermill": {
     "duration": 7.131076,
     "end_time": "2024-01-31T10:29:11.934874",
     "exception": false,
     "start_time": "2024-01-31T10:29:04.803798",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.24.3\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import torch\n",
    "from PIL import Image\n",
    "from sklearn.model_selection import train_test_split\n",
    "from torch.utils.data import DataLoader, IterableDataset\n",
    "from torchvision import transforms\n",
    "from transformers import BertTokenizer\n",
    "\n",
    "EMOTION2ID = {\"positive\": 0, \"neutral\": 1, \"negative\": 2, \"null\": 3}\n",
    "ID2EMOTION = {0: \"positive\", 1: \"neutral\", 2: \"negative\", 3: \"null\"}\n",
    "SEED = 42\n",
    "\n",
    "dataset_loc = \"/kaggle/input/lab5-dataset/dataset\"\n",
    "\n",
    "\n",
    "class MultimodalDataset(IterableDataset):\n",
    "    def __init__(self, samples, tokenizer):\n",
    "        self.data = samples\n",
    "        self.tokenizer = tokenizer\n",
    "        self.samples = []\n",
    "\n",
    "        for guid, tag in zip(self.data[\"guid\"], self.data[\"tag\"]):\n",
    "            # load img\n",
    "            img = Image.open(f\"{dataset_loc}/data/{guid}.jpg\")\n",
    "            transform = transforms.Compose(\n",
    "                [\n",
    "                    transforms.Resize(224),\n",
    "                    transforms.CenterCrop(224),\n",
    "                    transforms.ToTensor(),\n",
    "                ]\n",
    "            )\n",
    "            img = transform(img)\n",
    "\n",
    "            # load txt\n",
    "            with open(f\"{dataset_loc}/data/{guid}.txt\", \"r\", encoding=\"gb18030\") as f:\n",
    "                txt = f.read()\n",
    "                txt = txt.replace(\"#\", \"\")\n",
    "\n",
    "            txt = self.tokenizer.encode(txt, add_special_tokens=True)\n",
    "            self.samples.append((guid, txt, img, EMOTION2ID[tag]))\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.data)\n",
    "\n",
    "    def __iter__(self):\n",
    "        for bundles in self.samples:\n",
    "            yield bundles\n",
    "\n",
    "\n",
    "class DataHandler:\n",
    "    def __init__(self, batch_size):\n",
    "        self.batch_size = batch_size\n",
    "        self.tokenizer = BertTokenizer.from_pretrained(\n",
    "            \"bert-base-uncased\", mirror=\"bfsu\"\n",
    "        )\n",
    "        self.get_dataloader()\n",
    "\n",
    "    @staticmethod\n",
    "    def _collate_fn(batch):\n",
    "        guid = []\n",
    "        txt = []\n",
    "        img = []\n",
    "        tag = []\n",
    "        max_len = 0\n",
    "        for sample in batch:\n",
    "            guid.append(sample[0])\n",
    "            txt.append(sample[1])\n",
    "            max_len = max(max_len, len(sample[1]))\n",
    "            img.append(np.array(sample[2]).tolist())\n",
    "            tag.append(sample[3])\n",
    "\n",
    "        padded_txt = []\n",
    "        txt_mask = []\n",
    "        for t in txt:\n",
    "            padded = t + [0] * (max_len - len(t))\n",
    "            mask = [1] * len(t) + [0] * (max_len - len(t))\n",
    "            padded_txt.append(padded)\n",
    "            txt_mask.append(mask)\n",
    "\n",
    "        padded_txt = torch.LongTensor(padded_txt)\n",
    "        txt_mask = torch.BoolTensor(txt_mask)\n",
    "        img = torch.FloatTensor(img)\n",
    "        tag = torch.LongTensor(tag)\n",
    "        return guid, padded_txt, txt_mask, img, tag\n",
    "\n",
    "\n",
    "    def get_dataloader(self):\n",
    "        print(\"===== Data Info =====\")\n",
    "        samples = pd.read_csv(f\"{dataset_loc}/train.txt\")\n",
    "        print(samples[\"tag\"].value_counts())\n",
    "\n",
    "        train_samples, val_samples = train_test_split(\n",
    "            samples, test_size=0.2, random_state=SEED\n",
    "        )\n",
    "        test_samples = pd.read_csv(f\"{dataset_loc}/test_without_label.txt\").fillna(\n",
    "            \"null\"\n",
    "        )\n",
    "\n",
    "        self.train_size = len(train_samples)\n",
    "        self.val_size = len(val_samples)\n",
    "        self.test_size = len(test_samples)\n",
    "\n",
    "        print(\"\\nTrain size:\", self.train_size)\n",
    "        print(\"Val size:\", self.val_size)\n",
    "        print(\"Test size:\", self.test_size, \"\\n\")\n",
    "        print(\"Loading dataset...\\n\")\n",
    "\n",
    "        train_data = MultimodalDataset(train_samples, self.tokenizer)\n",
    "        val_data = MultimodalDataset(val_samples, self.tokenizer)\n",
    "        test_data = MultimodalDataset(test_samples, self.tokenizer)\n",
    "\n",
    "        self.train_loader = DataLoader(\n",
    "            train_data, batch_size=self.batch_size, collate_fn=self._collate_fn\n",
    "        )\n",
    "        self.val_loader = DataLoader(\n",
    "            val_data, batch_size=self.batch_size, collate_fn=self._collate_fn\n",
    "        )\n",
    "        self.test_loader = DataLoader(\n",
    "            test_data, batch_size=self.batch_size, collate_fn=self._collate_fn\n",
    "        )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "187640ac",
   "metadata": {
    "papermill": {
     "duration": 0.004479,
     "end_time": "2024-01-31T10:29:11.944057",
     "exception": false,
     "start_time": "2024-01-31T10:29:11.939578",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## 2. 定义模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "92d8fcae",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:11.955514Z",
     "iopub.status.busy": "2024-01-31T10:29:11.954987Z",
     "iopub.status.idle": "2024-01-31T10:29:12.277727Z",
     "shell.execute_reply": "2024-01-31T10:29:12.276898Z"
    },
    "papermill": {
     "duration": 0.331171,
     "end_time": "2024-01-31T10:29:12.280138",
     "exception": false,
     "start_time": "2024-01-31T10:29:11.948967",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "from torchvision.models import ResNet18_Weights\n",
    "from transformers import BertModel\n",
    "\n",
    "\n",
    "class BertResnetWithConcat(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.txt_model = BertModel.from_pretrained(\"bert-base-uncased\")\n",
    "        self.img_model = torchvision.models.resnet18(\n",
    "            weights=ResNet18_Weights.IMAGENET1K_V1\n",
    "        )\n",
    "\n",
    "        self.fc = nn.Linear(768 + 1000, 3)\n",
    "        self.only_img_fc = nn.Linear(1000, 3)\n",
    "        self.only_txt_fc = nn.Linear(768, 3)\n",
    "\n",
    "    def forward(self, txt, txt_mask, img, ablate):\n",
    "        if ablate == 0:  # both\n",
    "            img = self.img_model(img)\n",
    "            txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "            txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "            txt.view(txt.shape[0], -1)\n",
    "\n",
    "            out = torch.cat((txt, img), dim=-1)\n",
    "            out = self.fc(out)\n",
    "        elif ablate == 1:  # img only\n",
    "            img = self.img_model(img)\n",
    "            out = self.only_img_fc(img)\n",
    "        else:  # txt only\n",
    "            txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "            txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "            txt.view(txt.shape[0], -1)\n",
    "            out = self.only_txt_fc(txt)\n",
    "\n",
    "        return out\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "41d4ae25",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:12.290882Z",
     "iopub.status.busy": "2024-01-31T10:29:12.290348Z",
     "iopub.status.idle": "2024-01-31T10:29:12.301117Z",
     "shell.execute_reply": "2024-01-31T10:29:12.300260Z"
    },
    "papermill": {
     "duration": 0.01818,
     "end_time": "2024-01-31T10:29:12.303088",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.284908",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "from torchvision.models import ResNet18_Weights\n",
    "from transformers import BertModel\n",
    "\n",
    "\n",
    "class BertResnetWithAttention(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.txt_model = BertModel.from_pretrained(\"bert-base-uncased\")\n",
    "        self.img_model = torchvision.models.resnet18(\n",
    "            weights=ResNet18_Weights.IMAGENET1K_V1\n",
    "        )\n",
    "\n",
    "        self.txt_fc = nn.Linear(768, 256)\n",
    "        self.img_fc = nn.Linear(1000, 256)\n",
    "\n",
    "        self.fc = nn.Linear(512, 3)\n",
    "\n",
    "        self.transformer_encoder = nn.TransformerEncoder(\n",
    "            encoder_layer=nn.TransformerEncoderLayer(\n",
    "                d_model=512,\n",
    "                nhead=4,\n",
    "                dim_feedforward=512,\n",
    "                dropout=0.4,\n",
    "            ),\n",
    "            num_layers=2,\n",
    "        )\n",
    "\n",
    "        self.activate = nn.LeakyReLU()\n",
    "\n",
    "    def forward(self, txt, txt_mask, img, _):\n",
    "        img = self.img_model(img)\n",
    "        img = self.img_fc(img)\n",
    "        img = self.activate(img)\n",
    "\n",
    "        txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "        txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "        txt.view(txt.shape[0], -1)\n",
    "        txt = self.txt_fc(txt)\n",
    "        txt = self.activate(txt)\n",
    "\n",
    "        out = torch.cat((txt, img), dim=-1)\n",
    "        out = self.transformer_encoder(out)\n",
    "        out = self.fc(out)\n",
    "\n",
    "        return out\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ee682e79",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:12.313369Z",
     "iopub.status.busy": "2024-01-31T10:29:12.313055Z",
     "iopub.status.idle": "2024-01-31T10:29:12.323984Z",
     "shell.execute_reply": "2024-01-31T10:29:12.323027Z"
    },
    "papermill": {
     "duration": 0.018494,
     "end_time": "2024-01-31T10:29:12.326213",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.307719",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "from torchvision.models import DenseNet121_Weights\n",
    "from transformers import BertModel\n",
    "\n",
    "\n",
    "class BertDensenetWithConcat(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.txt_model = BertModel.from_pretrained(\"bert-base-uncased\")\n",
    "        self.img_model = torchvision.models.densenet121(\n",
    "            weights=DenseNet121_Weights.IMAGENET1K_V1\n",
    "        )\n",
    "\n",
    "        self.fc = nn.Linear(768 + 1000, 3)\n",
    "        self.only_img_fc = nn.Linear(1000, 3)\n",
    "        self.only_txt_fc = nn.Linear(768, 3)\n",
    "\n",
    "    def forward(self, txt, txt_mask, img, ablate):\n",
    "        if ablate == 0:  # both\n",
    "            img = self.img_model(img)\n",
    "            txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "            txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "            txt.view(txt.shape[0], -1)\n",
    "\n",
    "            out = torch.cat((txt, img), dim=-1)\n",
    "            out = self.fc(out)\n",
    "        elif ablate == 1:  # img only\n",
    "            img = self.img_model(img)\n",
    "            out = self.only_img_fc(img)\n",
    "        else:  # txt only\n",
    "            txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "            txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "            txt.view(txt.shape[0], -1)\n",
    "            out = self.only_txt_fc(txt)\n",
    "\n",
    "        return out\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e397a0df",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:12.336768Z",
     "iopub.status.busy": "2024-01-31T10:29:12.336026Z",
     "iopub.status.idle": "2024-01-31T10:29:12.346286Z",
     "shell.execute_reply": "2024-01-31T10:29:12.345605Z"
    },
    "papermill": {
     "duration": 0.017544,
     "end_time": "2024-01-31T10:29:12.348315",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.330771",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "from torchvision.models import DenseNet121_Weights\n",
    "from transformers import BertModel\n",
    "\n",
    "\n",
    "class BertDensenetWithAttention(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.txt_model = BertModel.from_pretrained(\"bert-base-uncased\")\n",
    "        self.img_model = torchvision.models.densenet121(\n",
    "            weights=DenseNet121_Weights.IMAGENET1K_V1\n",
    "        )\n",
    "\n",
    "        self.txt_fc = nn.Linear(768, 256)\n",
    "        self.img_fc = nn.Linear(1000, 256)\n",
    "\n",
    "        self.fc = nn.Linear(512, 3)\n",
    "\n",
    "        self.transformer_encoder = nn.TransformerEncoder(\n",
    "            encoder_layer=nn.TransformerEncoderLayer(\n",
    "                d_model=512,\n",
    "                nhead=4,\n",
    "                dim_feedforward=512,\n",
    "                dropout=0.4,\n",
    "            ),\n",
    "            num_layers=2,\n",
    "        )\n",
    "\n",
    "        self.activate = nn.LeakyReLU()\n",
    "\n",
    "    def forward(self, txt, txt_mask, img, _):\n",
    "        img = self.img_model(img)\n",
    "        img = self.img_fc(img)\n",
    "        img = self.activate(img)\n",
    "\n",
    "        txt = self.txt_model(input_ids=txt, attention_mask=txt_mask)  # type: ignore\n",
    "        txt = txt.last_hidden_state[:, 0, :]  # CLS vector\n",
    "        txt.view(txt.shape[0], -1)\n",
    "        txt = self.txt_fc(txt)\n",
    "        txt = self.activate(txt)\n",
    "\n",
    "        out = torch.cat((txt, img), dim=-1)\n",
    "        out = self.transformer_encoder(out)\n",
    "        out = self.fc(out)\n",
    "\n",
    "        return out\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df3379fe",
   "metadata": {
    "papermill": {
     "duration": 0.004217,
     "end_time": "2024-01-31T10:29:12.356944",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.352727",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## 3. 定义训练过程"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c383023d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:12.366672Z",
     "iopub.status.busy": "2024-01-31T10:29:12.366391Z",
     "iopub.status.idle": "2024-01-31T10:29:12.458650Z",
     "shell.execute_reply": "2024-01-31T10:29:12.457794Z"
    },
    "papermill": {
     "duration": 0.09993,
     "end_time": "2024-01-31T10:29:12.461169",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.361239",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import random\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from sklearn.metrics import f1_score, precision_score, recall_score, accuracy_score\n",
    "from torch import optim\n",
    "from tqdm import tqdm\n",
    "\n",
    "output_loc = \".\"\n",
    "\n",
    "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "\n",
    "def set_seed(seed):\n",
    "    random.seed(seed)\n",
    "    np.random.seed(seed)\n",
    "    torch.manual_seed(seed)\n",
    "    torch.cuda.manual_seed_all(seed)\n",
    "    torch.backends.cudnn.deterministic = True\n",
    "\n",
    "\n",
    "class Model:\n",
    "    def __init__(\n",
    "        self,\n",
    "        max_epochs=10,\n",
    "        lr=1e-5,\n",
    "        batch_size=16,\n",
    "        model=0,\n",
    "        ablate=0,\n",
    "    ):\n",
    "        set_seed(SEED)\n",
    "\n",
    "        self.epochs = max_epochs\n",
    "        self.batch_size = batch_size\n",
    "        self.ablate = ablate\n",
    "        self.model_type = model\n",
    "        if model == 0:\n",
    "            self.model = BertResnetWithConcat().to(DEVICE)\n",
    "        elif model == 1:\n",
    "            self.model = BertResnetWithAttention().to(DEVICE)\n",
    "        elif model == 2:\n",
    "            self.model = BertDensenetWithConcat().to(DEVICE)\n",
    "        elif model == 3:\n",
    "            self.model = BertDensenetWithAttention().to(DEVICE)\n",
    "\n",
    "        self.optimizer = optim.AdamW(self.model.parameters(), lr=lr)\n",
    "        self.criterion = nn.CrossEntropyLoss()\n",
    "        self.data = DataHandler(self.batch_size)\n",
    "\n",
    "    def train(self):\n",
    "        self.train_loss = []\n",
    "        self.val_accuracy = []\n",
    "        print(\"===== Traning Info =====\")\n",
    "        print(\"Device:\", DEVICE)\n",
    "        print(\"Batch size:\", self.batch_size)\n",
    "\n",
    "        if self.ablate == 0:\n",
    "            if self.model_type == 0:\n",
    "                print(\"Model: BertResnetWithConcat\")\n",
    "            elif self.model_type == 1:\n",
    "                print(\"Model: BertResnetWithCrossAttention\")\n",
    "            elif self.model_type == 2:\n",
    "                print(\"Model: BertDensenetWithConcat\")\n",
    "            elif self.model_type == 3:\n",
    "                print(\"Model: BertDensenetWithCrossAttention\")\n",
    "        elif self.ablate == 1:\n",
    "            if self.model_type == 0:\n",
    "                print(\"Model: Resnet Only\")\n",
    "            elif self.model_type == 2:\n",
    "                print(\"Model: Densenet Only\")\n",
    "        elif self.ablate == 2:\n",
    "            print(\"Model: Bert only\")\n",
    "\n",
    "        print(\"\\n==== Starting Train ====\")\n",
    "\n",
    "        best_metrics = [float(\"inf\"), 0, 0, 0, 0]\n",
    "        early_stop_patience = 2\n",
    "        early_stop_count = 0\n",
    "        epoch = 0\n",
    "\n",
    "        for epoch in range(1, self.epochs + 1):\n",
    "            self._epoch_train(epoch)\n",
    "            metrics = self._evaluate()\n",
    "            if metrics[1] > best_metrics[1]:\n",
    "                best_metrics = metrics\n",
    "                torch.save(self.model.state_dict(), f\"{output_loc}/model.pt\")\n",
    "                early_stop_count = 0\n",
    "            else:\n",
    "                early_stop_count += 1\n",
    "                if early_stop_count == early_stop_patience:\n",
    "                    print(\"Early stop...\")\n",
    "                    break\n",
    "\n",
    "        print(f\"The training epoch is {epoch}\")\n",
    "        print(f\"Choose model with best accuracy: {best_metrics[1]}, with:\")\n",
    "        print(f\"Precision: {best_metrics[2]}\")\n",
    "        print(f\"Recall: {best_metrics[3]}\")\n",
    "        print(f\"F1: {best_metrics[4]}\")\n",
    "        self.model.load_state_dict(torch.load(f\"{output_loc}/model.pt\"))\n",
    "\n",
    "        def moving_average(data, window_size):\n",
    "            weights = np.repeat(1.0, window_size) / window_size\n",
    "            smoothed_data = np.convolve(data, weights, \"valid\")\n",
    "            return smoothed_data\n",
    "\n",
    "        window_size = 15\n",
    "        self.train_loss = moving_average(self.train_loss, window_size)\n",
    "\n",
    "    def _epoch_train(self, epoch):\n",
    "        self.model.train()\n",
    "        epoch_loss = 0\n",
    "        correct = 0\n",
    "        for _, txt, txt_mask, image, label in tqdm(self.data.train_loader):\n",
    "            txt = txt.to(DEVICE)\n",
    "            txt_mask = txt_mask.to(DEVICE)\n",
    "            image = image.to(DEVICE)\n",
    "            label = label.to(DEVICE)\n",
    "\n",
    "            self.optimizer.zero_grad()\n",
    "            output = self.model(txt, txt_mask, image, self.ablate)\n",
    "\n",
    "            loss = self.criterion(output, label)\n",
    "            loss.backward()\n",
    "\n",
    "            # used to prevent gradient explosion\n",
    "            # clip_grad_norm_(self.model.parameters(), 1)\n",
    "\n",
    "            self.optimizer.step()\n",
    "\n",
    "            pred = output.argmax(dim=1)\n",
    "            correct += (pred == label).sum().item()\n",
    "\n",
    "            self.train_loss.append(loss.item())\n",
    "            epoch_loss += loss.item()\n",
    "\n",
    "        epoch_loss /= len(self.data.train_loader)\n",
    "        epoch_acc = correct / self.data.train_size\n",
    "        print(f\"Train Epoch {epoch}\")\n",
    "        print(\"Train set: \\nLoss: {}, Accuracy: {}\".format(epoch_loss, epoch_acc))\n",
    "\n",
    "    def _evaluate(self):\n",
    "        self.model.eval()\n",
    "        epoch_loss = 0\n",
    "        preds = []\n",
    "        labels = []\n",
    "\n",
    "        with torch.no_grad():\n",
    "            for _, txt, txt_mask, image, label in self.data.val_loader:\n",
    "                txt = txt.to(DEVICE)\n",
    "                txt_mask = txt_mask.to(DEVICE)\n",
    "                image = image.to(DEVICE)\n",
    "                label = label.to(DEVICE)\n",
    "\n",
    "                self.optimizer.zero_grad()\n",
    "                output = self.model(txt, txt_mask, image, self.ablate)\n",
    "\n",
    "                loss = self.criterion(output, label)\n",
    "\n",
    "                pred = output.argmax(dim=1)\n",
    "                epoch_loss += loss.item()\n",
    "\n",
    "                preds.append(pred)\n",
    "                labels.append(label)\n",
    "\n",
    "        epoch_loss /= len(self.data.val_loader)\n",
    "\n",
    "        preds = torch.cat(preds).cpu().numpy()\n",
    "        labels = torch.cat(labels).cpu().numpy()\n",
    "\n",
    "        epoch_acc = accuracy_score(labels, preds)\n",
    "        self.val_accuracy.append(epoch_acc)\n",
    "\n",
    "        f1 = f1_score(labels, preds, average=\"macro\")\n",
    "        precision = precision_score(labels, preds, average=\"macro\")\n",
    "        recall = recall_score(labels, preds, average=\"macro\")\n",
    "        print(\n",
    "            \"Valid set: \\nLoss: {}, Accuracy: {}, Precision: {}, Recall: {}, F1: {}\".format(\n",
    "                epoch_loss, epoch_acc, precision, recall, f1\n",
    "            )\n",
    "        )\n",
    "        return epoch_loss, epoch_acc, precision, recall, f1\n",
    "\n",
    "    def save_test_result(self):\n",
    "        print(f\"Predict and save test result...\")\n",
    "        self.model.eval()\n",
    "        guid_list = []\n",
    "        tag_list = []\n",
    "\n",
    "        with torch.no_grad():\n",
    "            for guid, txt, txt_mask, image, _ in tqdm(self.data.test_loader):\n",
    "                txt = txt.to(DEVICE)\n",
    "                txt_mask = txt_mask.to(DEVICE)\n",
    "                image = image.to(DEVICE)\n",
    "\n",
    "                self.optimizer.zero_grad()\n",
    "                output = self.model(txt, txt_mask, image, self.ablate)\n",
    "                pred = output.argmax(dim=1)\n",
    "\n",
    "                guid_list.extend(guid)\n",
    "                emotion = [ID2EMOTION[p.item()] for p in pred]\n",
    "                tag_list.extend(emotion)\n",
    "\n",
    "        df = pd.DataFrame({\"guid\": guid_list, \"tag\": tag_list})\n",
    "        df.to_csv(f\"{output_loc}/test_with_predict.txt\", index=False)\n",
    "        print(\"Save successfully!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fafd659",
   "metadata": {
    "papermill": {
     "duration": 0.004477,
     "end_time": "2024-01-31T10:29:12.470780",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.466303",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## 4. 开始训练"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0994f1d3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T10:29:12.481220Z",
     "iopub.status.busy": "2024-01-31T10:29:12.480873Z",
     "iopub.status.idle": "2024-01-31T11:35:59.886463Z",
     "shell.execute_reply": "2024-01-31T11:35:59.885501Z"
    },
    "papermill": {
     "duration": 4007.412976,
     "end_time": "2024-01-31T11:35:59.888455",
     "exception": false,
     "start_time": "2024-01-31T10:29:12.475479",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe7f7bfd07c84e4dac299a1af8d3749f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "config.json:   0%|          | 0.00/570 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "773c7b4b21234c01beee58f5de81b8cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.safetensors:   0%|          | 0.00/440M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n",
      "100%|██████████| 44.7M/44.7M [00:03<00:00, 14.8MB/s]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c2974b1b71c64ec188899b8dfdfb51c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/28.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "38a34fb31eab4d989caabd3e9a02ea09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1f67f703912f418eb23ba9201c725f1e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json:   0%|          | 0.00/466k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: BertResnetWithConcat\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:56<00:00,  1.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8481150774657726, Accuracy: 0.62375\n",
      "Valid set: \n",
      "Loss: 0.7514602029323578, Accuracy: 0.6825, Precision: 0.603969790031171, Recall: 0.5570725861453991, F1: 0.555344322998161\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:53<00:00,  1.76it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.47838540211319924, Accuracy: 0.8065625\n",
      "Valid set: \n",
      "Loss: 0.7379146856069565, Accuracy: 0.68875, Precision: 0.5888398848223476, Recall: 0.5602198569720833, F1: 0.5637548143465808\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:52<00:00,  1.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.28736488822847606, Accuracy: 0.9075\n",
      "Valid set: \n",
      "Loss: 0.8212919801473617, Accuracy: 0.685, Precision: 0.5908649100049468, Recall: 0.56186384866322, F1: 0.5672977486686953\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:52<00:00,  1.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.1549170281365514, Accuracy: 0.960625\n",
      "Valid set: \n",
      "Loss: 0.8840981182456017, Accuracy: 0.68625, Precision: 0.584869950667536, Recall: 0.5471322034967923, F1: 0.5593031999710227\n",
      "Early stop...\n",
      "The training epoch is 4\n",
      "Choose model with best accuracy: 0.68875, with:\n",
      "Precision: 0.5888398848223476\n",
      "Recall: 0.5602198569720833\n",
      "F1: 0.5637548143465808\n",
      "Training time: 545.4064795970917\n",
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: BertResnetWithCrossAttention\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:54<00:00,  1.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8277765029668808, Accuracy: 0.6328125\n",
      "Valid set: \n",
      "Loss: 0.7208098298311234, Accuracy: 0.69, Precision: 0.6480346187043646, Recall: 0.5593136467940031, F1: 0.5527689338966122\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:52<00:00,  1.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.5352846904098988, Accuracy: 0.7871875\n",
      "Valid set: \n",
      "Loss: 0.7379485428333282, Accuracy: 0.69625, Precision: 0.6165131091907264, Recall: 0.5536484740256349, F1: 0.5607703728815687\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:51<00:00,  1.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.3386993007734418, Accuracy: 0.8825\n",
      "Valid set: \n",
      "Loss: 0.8352116751670837, Accuracy: 0.67, Precision: 0.5824309590309621, Recall: 0.5594147039930172, F1: 0.5552497317099085\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:50<00:00,  1.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.1918549701757729, Accuracy: 0.94375\n",
      "Valid set: \n",
      "Loss: 0.9013938593864441, Accuracy: 0.69125, Precision: 0.6024421608079197, Recall: 0.5613392679919658, F1: 0.5750390773459982\n",
      "Early stop...\n",
      "The training epoch is 4\n",
      "Choose model with best accuracy: 0.69625, with:\n",
      "Precision: 0.6165131091907264\n",
      "Recall: 0.5536484740256349\n",
      "F1: 0.5607703728815687\n",
      "Training time: 540.4838178157806\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/densenet121-a639ec97.pth\" to /root/.cache/torch/hub/checkpoints/densenet121-a639ec97.pth\n",
      "100%|██████████| 30.8M/30.8M [00:01<00:00, 17.4MB/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: BertDensenetWithConcat\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [02:00<00:00,  1.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8704016382992268, Accuracy: 0.619375\n",
      "Valid set: \n",
      "Loss: 0.7152319985628128, Accuracy: 0.71, Precision: 0.6208239531531023, Recall: 0.5582173680654561, F1: 0.5564080103313599\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [02:00<00:00,  1.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.5019320581853389, Accuracy: 0.8028125\n",
      "Valid set: \n",
      "Loss: 0.700513830780983, Accuracy: 0.70625, Precision: 0.6129394682247603, Recall: 0.5624954561570591, F1: 0.5705689666795364\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.30314570002257824, Accuracy: 0.8915625\n",
      "Valid set: \n",
      "Loss: 0.7601141846179962, Accuracy: 0.70125, Precision: 0.5938253171048067, Recall: 0.5659983459878694, F1: 0.5692708333333334\n",
      "Early stop...\n",
      "The training epoch is 3\n",
      "Choose model with best accuracy: 0.71, with:\n",
      "Precision: 0.6208239531531023\n",
      "Recall: 0.5582173680654561\n",
      "F1: 0.5564080103313599\n",
      "Training time: 429.5694832801819\n",
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: BertDensenetWithCrossAttention\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8182583798468113, Accuracy: 0.6365625\n",
      "Valid set: \n",
      "Loss: 0.7047787576913833, Accuracy: 0.71, Precision: 0.7028286432737554, Recall: 0.5622351379292186, F1: 0.5573259647448802\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.5425574917346239, Accuracy: 0.7828125\n",
      "Valid set: \n",
      "Loss: 0.7118367427587509, Accuracy: 0.71875, Precision: 0.643425979876645, Recall: 0.5804123219000118, F1: 0.584184987464747\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [02:00<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.3441022937744856, Accuracy: 0.874375\n",
      "Valid set: \n",
      "Loss: 0.8002323782444001, Accuracy: 0.71375, Precision: 0.6378357020303, Recall: 0.5884065006642273, F1: 0.5965659441596906\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.18960339413955807, Accuracy: 0.940625\n",
      "Valid set: \n",
      "Loss: 0.8699598011374473, Accuracy: 0.705, Precision: 0.6143803026051361, Recall: 0.5707453693310163, F1: 0.5840671301984707\n",
      "Early stop...\n",
      "The training epoch is 4\n",
      "Choose model with best accuracy: 0.71875, with:\n",
      "Precision: 0.643425979876645\n",
      "Recall: 0.5804123219000118\n",
      "F1: 0.584184987464747\n",
      "Training time: 572.3940951824188\n",
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: Resnet Only\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:28<00:00,  2.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.9368028110265731, Accuracy: 0.579375\n",
      "Valid set: \n",
      "Loss: 0.895398154258728, Accuracy: 0.5725, Precision: 0.47078979843685725, Recall: 0.4203124543616948, F1: 0.43127067198439767\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:28<00:00,  2.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.6585789713263511, Accuracy: 0.7140625\n",
      "Valid set: \n",
      "Loss: 0.8761020642518997, Accuracy: 0.5825, Precision: 0.4760442487173153, Recall: 0.434645946956057, F1: 0.44475511594604583\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:28<00:00,  2.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.4917768894135952, Accuracy: 0.8196875\n",
      "Valid set: \n",
      "Loss: 0.8894303661584854, Accuracy: 0.5975, Precision: 0.49791050130033176, Recall: 0.45692308971512735, F1: 0.46825335944520524\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:28<00:00,  2.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.34432803846895693, Accuracy: 0.9134375\n",
      "Valid set: \n",
      "Loss: 0.9255641710758209, Accuracy: 0.59375, Precision: 0.4981060145466009, Recall: 0.4622389329042027, F1: 0.47348745119566754\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:29<00:00,  2.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 5\n",
      "Train set: \n",
      "Loss: 0.21911156602203846, Accuracy: 0.9709375\n",
      "Valid set: \n",
      "Loss: 0.98116266310215, Accuracy: 0.5975, Precision: 0.49961157085839014, Recall: 0.46836962971064594, F1: 0.47893013302663995\n",
      "Early stop...\n",
      "The training epoch is 5\n",
      "Choose model with best accuracy: 0.5975, with:\n",
      "Precision: 0.49791050130033176\n",
      "Recall: 0.45692308971512735\n",
      "F1: 0.46825335944520524\n",
      "Training time: 546.778391122818\n",
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: Bert only\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:39<00:00,  2.01it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8073436160385609, Accuracy: 0.6475\n",
      "Valid set: \n",
      "Loss: 0.7088299888372421, Accuracy: 0.68125, Precision: 0.6489290478382804, Recall: 0.5409181451612044, F1: 0.5463500191084896\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:38<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.6280792745947837, Accuracy: 0.7425\n",
      "Valid set: \n",
      "Loss: 0.6900928324460983, Accuracy: 0.71125, Precision: 0.6691808467761754, Recall: 0.5611956772250118, F1: 0.5740785126312759\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:38<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.4697794611752033, Accuracy: 0.819375\n",
      "Valid set: \n",
      "Loss: 0.7664698773622513, Accuracy: 0.68125, Precision: 0.6113607267336799, Recall: 0.5479584633591967, F1: 0.5549737713694421\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:39<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.333427894897759, Accuracy: 0.8784375\n",
      "Valid set: \n",
      "Loss: 0.8845172747969627, Accuracy: 0.675, Precision: 0.572603891895624, Recall: 0.5515055923646809, F1: 0.5511787505503137\n",
      "Early stop...\n",
      "The training epoch is 4\n",
      "Choose model with best accuracy: 0.71125, with:\n",
      "Precision: 0.6691808467761754\n",
      "Recall: 0.5611956772250118\n",
      "F1: 0.5740785126312759\n",
      "Training time: 483.3071668148041\n",
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: Densenet Only\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:39<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.9690352302789688, Accuracy: 0.5596875\n",
      "Valid set: \n",
      "Loss: 0.8907674396038056, Accuracy: 0.60375, Precision: 0.4668176514800935, Recall: 0.42568820698731646, F1: 0.43016133442712606\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:38<00:00,  2.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.6779980766773224, Accuracy: 0.7175\n",
      "Valid set: \n",
      "Loss: 0.8675226849317551, Accuracy: 0.615, Precision: 0.4820095605267145, Recall: 0.4466293266712334, F1: 0.453219457666463\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:39<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.5173545356094837, Accuracy: 0.8175\n",
      "Valid set: \n",
      "Loss: 0.8680182683467865, Accuracy: 0.6325, Precision: 0.5071852557095338, Recall: 0.47214392421307033, F1: 0.48102421402875195\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:38<00:00,  2.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.3854733154177666, Accuracy: 0.89125\n",
      "Valid set: \n",
      "Loss: 0.8815545570850373, Accuracy: 0.62625, Precision: 0.5044052225644265, Recall: 0.4772023273856694, F1: 0.4857964344579962\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:38<00:00,  2.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 5\n",
      "Train set: \n",
      "Loss: 0.27447510592639446, Accuracy: 0.953125\n",
      "Valid set: \n",
      "Loss: 0.9061819243431092, Accuracy: 0.62625, Precision: 0.5044015511781458, Recall: 0.4779366976852571, F1: 0.486341764779525\n",
      "Early stop...\n",
      "The training epoch is 5\n",
      "Choose model with best accuracy: 0.6325, with:\n",
      "Precision: 0.5071852557095338\n",
      "Recall: 0.47214392421307033\n",
      "F1: 0.48102421402875195\n",
      "Training time: 603.2657871246338\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "if __name__ == \"__main__\":\n",
    "    bert_resnet_with_concat = Model(model=0)\n",
    "\n",
    "    start_time = time.time()\n",
    "    bert_resnet_with_concat.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    bert_resnet_with_attention = Model(model=1)\n",
    "\n",
    "    start_time = time.time()\n",
    "    bert_resnet_with_attention.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    bert_densenet_with_concat = Model(model=2)\n",
    "\n",
    "    start_time = time.time()\n",
    "    bert_densenet_with_concat.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    bert_densenet_with_attention = Model(model=3)\n",
    "\n",
    "    start_time = time.time()\n",
    "    bert_densenet_with_attention.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    resnet = Model(model=0, ablate=1)\n",
    "\n",
    "    start_time = time.time()\n",
    "    resnet.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    bert = Model(model=0, ablate=2)\n",
    "\n",
    "    start_time = time.time()\n",
    "    bert.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    densenet = Model(model=2, ablate=1)\n",
    "\n",
    "    start_time = time.time()\n",
    "    densenet.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "\n",
    "    plt.plot(bert_resnet_with_concat.train_loss, label=\"BertResnetWithConcat\")\n",
    "    plt.plot(bert_resnet_with_attention.train_loss, label=\"BertResnetWithAttention\")\n",
    "    plt.plot(bert_densenet_with_concat.train_loss, label=\"BertDensenetWithConcat\")\n",
    "    plt.plot(bert_densenet_with_attention.train_loss, label=\"BertDensenetWithAttention\")\n",
    "    plt.plot(resnet.train_loss, label=\"Resnet Only\")\n",
    "    plt.plot(bert.train_loss, label=\"Bert Only\")\n",
    "    plt.plot(densenet.train_loss, label=\"Densenet Only\")\n",
    "    plt.title(\"Train Loss\")\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "    plt.plot(bert_resnet_with_concat.val_accuracy, label=\"BertResnetWithConcat\")\n",
    "    plt.plot(bert_resnet_with_attention.val_accuracy, label=\"BertResnetWithAttention\")\n",
    "    plt.plot(bert_densenet_with_concat.val_accuracy, label=\"BertDensenetWithConcat\")\n",
    "    plt.plot(\n",
    "        bert_densenet_with_attention.val_accuracy, label=\"BertDensenetWithAttention\"\n",
    "    )\n",
    "    plt.plot(resnet.val_accuracy, label=\"Resnet Only\")\n",
    "    plt.plot(bert.val_accuracy, label=\"Bert Only\")\n",
    "    plt.plot(densenet.val_accuracy, label=\"Densenet Only\")\n",
    "    plt.title(\"Validation Accuracy\")\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "401a6815",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-01-31T11:36:00.936193Z",
     "iopub.status.busy": "2024-01-31T11:36:00.935814Z",
     "iopub.status.idle": "2024-01-31T11:46:22.563692Z",
     "shell.execute_reply": "2024-01-31T11:46:22.562740Z"
    },
    "papermill": {
     "duration": 622.132801,
     "end_time": "2024-01-31T11:46:22.565859",
     "exception": false,
     "start_time": "2024-01-31T11:36:00.433058",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===== Data Info =====\n",
      "tag\n",
      "positive    2388\n",
      "negative    1193\n",
      "neutral      419\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Train size: 3200\n",
      "Val size: 800\n",
      "Test size: 511 \n",
      "\n",
      "Loading dataset...\n",
      "\n",
      "===== Traning Info =====\n",
      "Device: cuda\n",
      "Batch size: 16\n",
      "Model: BertDensenetWithCrossAttention\n",
      "\n",
      "==== Starting Train ====\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [02:00<00:00,  1.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 1\n",
      "Train set: \n",
      "Loss: 0.8182583798468113, Accuracy: 0.6365625\n",
      "Valid set: \n",
      "Loss: 0.7047787576913833, Accuracy: 0.71, Precision: 0.7028286432737554, Recall: 0.5622351379292186, F1: 0.5573259647448802\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 2\n",
      "Train set: \n",
      "Loss: 0.5425574917346239, Accuracy: 0.7828125\n",
      "Valid set: \n",
      "Loss: 0.7118367427587509, Accuracy: 0.71875, Precision: 0.643425979876645, Recall: 0.5804123219000118, F1: 0.584184987464747\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 3\n",
      "Train set: \n",
      "Loss: 0.3441022937744856, Accuracy: 0.874375\n",
      "Valid set: \n",
      "Loss: 0.8002323782444001, Accuracy: 0.71375, Precision: 0.6378357020303, Recall: 0.5884065006642273, F1: 0.5965659441596906\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [01:59<00:00,  1.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch 4\n",
      "Train set: \n",
      "Loss: 0.18960339413955807, Accuracy: 0.940625\n",
      "Valid set: \n",
      "Loss: 0.8699598011374473, Accuracy: 0.705, Precision: 0.6143803026051361, Recall: 0.5707453693310163, F1: 0.5840671301984707\n",
      "Early stop...\n",
      "The training epoch is 4\n",
      "Choose model with best accuracy: 0.71875, with:\n",
      "Precision: 0.643425979876645\n",
      "Recall: 0.5804123219000118\n",
      "F1: 0.584184987464747\n",
      "Training time: 573.195264339447\n",
      "Predict and save test result...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 32/32 [00:14<00:00,  2.20it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Save successfully!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "if __name__ == \"__main__\":\n",
    "    model = Model(model=3)\n",
    "\n",
    "    start_time = time.time()\n",
    "    model.train()\n",
    "    end_time = time.time()\n",
    "\n",
    "    print(\"Training time:\", end_time - start_time)\n",
    "    model.save_test_result()"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "gpu",
   "dataSources": [
    {
     "datasetId": 4372797,
     "sourceId": 7508232,
     "sourceType": "datasetVersion"
    }
   ],
   "dockerImageVersionId": 30636,
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.10.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 4645.977891,
   "end_time": "2024-01-31T11:46:27.497060",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-01-31T10:29:01.519169",
   "version": "2.4.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "022506a0337a43b982732d5879779e66": {
      "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": ""
      }
     },
     "04250c5c949247ef83557f451c6ac6cb": {
      "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
      }
     },
     "0d448641d5b6444d9a2e212996df3a0b": {
      "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_e7b950f0a80245dbba49a512b346452d",
       "max": 231508.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_022506a0337a43b982732d5879779e66",
       "value": 231508.0
      }
     },
     "0e4fc9ae28a84136a3fe9c9c5d8ced48": {
      "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
      }
     },
     "172f9854d9f84169b093daae19a25a55": {
      "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
      }
     },
     "1c2cecd8527249159a8682b10b803224": {
      "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_9f59efb74e694f7891d6753a4407e44b",
       "placeholder": "​",
       "style": "IPY_MODEL_497c35be7e9d4c618b969de1b7e7016c",
       "value": "vocab.txt: 100%"
      }
     },
     "1ef045defad948679133a68926de5d51": {
      "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
      }
     },
     "1f67f703912f418eb23ba9201c725f1e": {
      "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_32638c8736a54b689862e84535b98625",
        "IPY_MODEL_74e1ee5c5a1445f58c84fb54a2052757",
        "IPY_MODEL_21ed6d98866b4186bf583dda3156066d"
       ],
       "layout": "IPY_MODEL_6590c324b8884fb983d91e8246760bbc"
      }
     },
     "21ed6d98866b4186bf583dda3156066d": {
      "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_d2127d8afef842488f1dd08d6c8eceff",
       "placeholder": "​",
       "style": "IPY_MODEL_3d857f2a4fa24921b1e837a90f4d9424",
       "value": " 466k/466k [00:00&lt;00:00, 794kB/s]"
      }
     },
     "28cae49563554b98b087c8b8ede67c19": {
      "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_47d058e380484984a2804454c0c0e449",
       "placeholder": "​",
       "style": "IPY_MODEL_d42d9ed3c68743cebdc11744d26a630d",
       "value": " 28.0/28.0 [00:00&lt;00:00, 2.30kB/s]"
      }
     },
     "2bd97b44ba5a4cd1bf651a9207a4315b": {
      "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": ""
      }
     },
     "32638c8736a54b689862e84535b98625": {
      "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_d853058c93bd430891fdb48a105671aa",
       "placeholder": "​",
       "style": "IPY_MODEL_e32e28be16174075b2451cb7a72483a6",
       "value": "tokenizer.json: 100%"
      }
     },
     "37925121838e4205ba310d9fd56bedba": {
      "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": ""
      }
     },
     "38a34fb31eab4d989caabd3e9a02ea09": {
      "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_1c2cecd8527249159a8682b10b803224",
        "IPY_MODEL_0d448641d5b6444d9a2e212996df3a0b",
        "IPY_MODEL_676715086e1444b9845bd515587c8699"
       ],
       "layout": "IPY_MODEL_c70e11a9f4934028869d045673192678"
      }
     },
     "392339b6773d471c937271c8776d7c1b": {
      "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_0e4fc9ae28a84136a3fe9c9c5d8ced48",
       "placeholder": "​",
       "style": "IPY_MODEL_59556bd9817b4c8fb9a5c2c2dba71f48",
       "value": " 440M/440M [00:11&lt;00:00, 42.4MB/s]"
      }
     },
     "3d857f2a4fa24921b1e837a90f4d9424": {
      "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": ""
      }
     },
     "407d761ff9ea4d1daaf563d2b3816bac": {
      "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
      }
     },
     "47d058e380484984a2804454c0c0e449": {
      "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
      }
     },
     "497c35be7e9d4c618b969de1b7e7016c": {
      "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": ""
      }
     },
     "4d7c0d11c59b4594a0d37c8f34a2c153": {
      "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": ""
      }
     },
     "529dedb8ddc54caba98e1128d7b4a351": {
      "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_ced59f85c861411ebdc54f243f50258b",
       "placeholder": "​",
       "style": "IPY_MODEL_750c2c8c9aa4492ebd4ffcabfffd4a93",
       "value": "tokenizer_config.json: 100%"
      }
     },
     "55020898305c4515a5b4ac9232820a98": {
      "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_5c02937e1f67485b8179dcc1674a793c",
       "placeholder": "​",
       "style": "IPY_MODEL_b6209ab5a6d043e19f124deeb78161b5",
       "value": " 570/570 [00:00&lt;00:00, 42.5kB/s]"
      }
     },
     "59556bd9817b4c8fb9a5c2c2dba71f48": {
      "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": ""
      }
     },
     "5c02937e1f67485b8179dcc1674a793c": {
      "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
      }
     },
     "5d073ae86f5049a0826750fbc609e473": {
      "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
      }
     },
     "6590c324b8884fb983d91e8246760bbc": {
      "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
      }
     },
     "676715086e1444b9845bd515587c8699": {
      "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_dcf4860a631642e28c362cde4122c449",
       "placeholder": "​",
       "style": "IPY_MODEL_4d7c0d11c59b4594a0d37c8f34a2c153",
       "value": " 232k/232k [00:00&lt;00:00, 1.13MB/s]"
      }
     },
     "74e1ee5c5a1445f58c84fb54a2052757": {
      "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_8fd7766366c14c8dbafe100d93cdec5d",
       "max": 466062.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_37925121838e4205ba310d9fd56bedba",
       "value": 466062.0
      }
     },
     "750c2c8c9aa4492ebd4ffcabfffd4a93": {
      "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": ""
      }
     },
     "773c7b4b21234c01beee58f5de81b8cf": {
      "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_921b74d3056d41dbb67b274a8eec8fe1",
        "IPY_MODEL_bc60327b8d624b85ae6a6941ae749745",
        "IPY_MODEL_392339b6773d471c937271c8776d7c1b"
       ],
       "layout": "IPY_MODEL_04250c5c949247ef83557f451c6ac6cb"
      }
     },
     "8b050db8c4174b1bb69978c01a263aa6": {
      "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": ""
      }
     },
     "8fd7766366c14c8dbafe100d93cdec5d": {
      "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
      }
     },
     "921b74d3056d41dbb67b274a8eec8fe1": {
      "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_b529195ffb3f4dd0997b358c12ad2a0a",
       "placeholder": "​",
       "style": "IPY_MODEL_f8081a7438db41e7932f72eea21363c9",
       "value": "model.safetensors: 100%"
      }
     },
     "97c5c561d73e4f5faf38dc7b628ed9cb": {
      "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
      }
     },
     "998fa7870c724797ba4912c299fae882": {
      "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_5d073ae86f5049a0826750fbc609e473",
       "max": 570.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_b640ec7c47b840db93f2f7889a2c9791",
       "value": 570.0
      }
     },
     "9f59efb74e694f7891d6753a4407e44b": {
      "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
      }
     },
     "b529195ffb3f4dd0997b358c12ad2a0a": {
      "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
      }
     },
     "b6209ab5a6d043e19f124deeb78161b5": {
      "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": ""
      }
     },
     "b640ec7c47b840db93f2f7889a2c9791": {
      "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": ""
      }
     },
     "bc60327b8d624b85ae6a6941ae749745": {
      "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_172f9854d9f84169b093daae19a25a55",
       "max": 440449768.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_2bd97b44ba5a4cd1bf651a9207a4315b",
       "value": 440449768.0
      }
     },
     "bf0cd6ef8c85401e84694e7b9d542181": {
      "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_97c5c561d73e4f5faf38dc7b628ed9cb",
       "placeholder": "​",
       "style": "IPY_MODEL_d9c14c8288b545259793d323ceb11376",
       "value": "config.json: 100%"
      }
     },
     "c2974b1b71c64ec188899b8dfdfb51c0": {
      "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_529dedb8ddc54caba98e1128d7b4a351",
        "IPY_MODEL_da7a0cd47b7848e0988d235a4c808fa8",
        "IPY_MODEL_28cae49563554b98b087c8b8ede67c19"
       ],
       "layout": "IPY_MODEL_c4c31e77cf1944148ebec7d064a8b1ab"
      }
     },
     "c4c31e77cf1944148ebec7d064a8b1ab": {
      "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
      }
     },
     "c70e11a9f4934028869d045673192678": {
      "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
      }
     },
     "ced59f85c861411ebdc54f243f50258b": {
      "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
      }
     },
     "d2127d8afef842488f1dd08d6c8eceff": {
      "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
      }
     },
     "d42d9ed3c68743cebdc11744d26a630d": {
      "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": ""
      }
     },
     "d853058c93bd430891fdb48a105671aa": {
      "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
      }
     },
     "d9c14c8288b545259793d323ceb11376": {
      "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": ""
      }
     },
     "da7a0cd47b7848e0988d235a4c808fa8": {
      "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_407d761ff9ea4d1daaf563d2b3816bac",
       "max": 28.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_8b050db8c4174b1bb69978c01a263aa6",
       "value": 28.0
      }
     },
     "dcf4860a631642e28c362cde4122c449": {
      "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
      }
     },
     "e32e28be16174075b2451cb7a72483a6": {
      "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": ""
      }
     },
     "e7b950f0a80245dbba49a512b346452d": {
      "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
      }
     },
     "f8081a7438db41e7932f72eea21363c9": {
      "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": ""
      }
     },
     "fe7f7bfd07c84e4dac299a1af8d3749f": {
      "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_bf0cd6ef8c85401e84694e7b9d542181",
        "IPY_MODEL_998fa7870c724797ba4912c299fae882",
        "IPY_MODEL_55020898305c4515a5b4ac9232820a98"
       ],
       "layout": "IPY_MODEL_1ef045defad948679133a68926de5d51"
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}