tuning_hyperparameter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0eabf088",
   "metadata": {},
   "source": [
    "# Import Libs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "id": "a164957a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torchvision\n",
    "from torch.optim import Adam, lr_scheduler\n",
    "from torch.utils.data import DataLoader, Dataset\n",
    "from torch.utils.data.sampler import SubsetRandomSampler\n",
    "from torchvision import models, transforms\n",
    "from torch.utils.tensorboard import SummaryWriter\n",
    "\n",
    "from PIL import Image\n",
    "import pandas as pd\n",
    "import numpy  as np\n",
    "from datetime import datetime\n",
    "import time\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc15f8b7",
   "metadata": {},
   "source": [
    "# Configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "id": "a277bc02",
   "metadata": {},
   "outputs": [],
   "source": [
    "classes = [\n",
    "    \"angry\", \n",
    "    \"disgust\", \n",
    "    \"fear\", \n",
    "    \"happy\", \n",
    "    \"sad\", \n",
    "    \"surprise\", \n",
    "    \"neutral\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "id": "85ac1f95",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Config:\n",
    "    # dataset\n",
    "    TRAIN_DS_PATH = './dataset/train.csv'\n",
    "    VAL_DS_PATH = './dataset/val.csv'\n",
    "    TEST_DS_PATH = './dataset/finaltest.csv'\n",
    "    \n",
    "    # images dir\n",
    "    TRAIN_IMG_DIR = './dataset/train/'\n",
    "    VAL_IMG_DIR  = './dataset/val/'\n",
    "    TEST_IMG_DIR  = './dataset/finaltest/'\n",
    "    \n",
    "    # training hyperparams\n",
    "    EPOCHS = 25\n",
    "    LR = 1e-3\n",
    "    BATCH_SIZE = 8\n",
    "    NUM_WORKERS = 0\n",
    "    SHUFFLE = True\n",
    "    \n",
    "    # saved model path\n",
    "    MODEL_DIR = './model/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "id": "3a43450f",
   "metadata": {},
   "outputs": [],
   "source": [
    "cfg = Config()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "id": "d1a2325d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tensorboard writer\n",
    "current_time = datetime.now().strftime('%b%d_%H-%M-%S')\n",
    "writer = SummaryWriter(f\"runs/{current_time}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd927d2d",
   "metadata": {},
   "source": [
    "# Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "id": "2e5d1af9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# reference: https://github.com/omarsayed7/Deep-Emotion/blob/master/data_loaders.py\n",
    "class EmotionDataset(Dataset):\n",
    "    def __init__(self, csv_file, img_dir, datatype, transform):\n",
    "        '''\n",
    "        Pytorch Dataset class\n",
    "        params:-\n",
    "                 csv_file : the path of the csv file    (train, validation, test)\n",
    "                 img_dir  : the directory of the images (train, validation, test)\n",
    "                 datatype : data type for the dataset   (train, val, test)\n",
    "                 transform: pytorch transformation over the data\n",
    "        return :-\n",
    "                 image, labels\n",
    "        '''\n",
    "        self.csv_file = pd.read_csv(csv_file)\n",
    "        self.labels = self.csv_file['emotion']\n",
    "        self.img_dir = img_dir\n",
    "        self.datatype = datatype\n",
    "        self.transform = transform\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.csv_file)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        if torch.is_tensor(idx):\n",
    "            idx = idx.tolist()\n",
    "        img = Image.open(self.img_dir + self.datatype + str(idx) + '.jpg').convert('RGB')\n",
    "        labels = np.array(self.labels[idx])\n",
    "        labels = torch.from_numpy(labels).long()\n",
    "\n",
    "        if self.transform :\n",
    "            img = self.transform(img)\n",
    "        return img, labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "id": "499e6487",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tranformations\n",
    "transformation= transforms.Compose([\n",
    "    transforms.Resize(256),\n",
    "#     transforms.Grayscale(3), # no need this since you .convert(\"RGB\") at __getitem__\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "id": "c8f25683",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create datasets\n",
    "train_dataset = EmotionDataset(\n",
    "    csv_file=cfg.TRAIN_DS_PATH, \n",
    "    img_dir=cfg.TRAIN_IMG_DIR,\n",
    "    datatype='train',\n",
    "    transform=transformation\n",
    ")\n",
    "\n",
    "validation_dataset = EmotionDataset(\n",
    "    csv_file=cfg.VAL_DS_PATH, \n",
    "    img_dir=cfg.VAL_IMG_DIR,\n",
    "    datatype='val',\n",
    "    transform = transformation\n",
    ")\n",
    "\n",
    "\n",
    "test_dataset = EmotionDataset(\n",
    "    csv_file=cfg.TEST_DS_PATH, \n",
    "    img_dir=cfg.TEST_IMG_DIR,\n",
    "    datatype='finaltest',\n",
    "    transform = transformation\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "id": "460c8bf9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create data loaders\n",
    "train_loader = DataLoader(\n",
    "    train_dataset, \n",
    "    batch_size=cfg.BATCH_SIZE, \n",
    "    shuffle =cfg.SHUFFLE, \n",
    "    num_workers=cfg.NUM_WORKERS\n",
    ")\n",
    "\n",
    "val_loader = DataLoader(\n",
    "    validation_dataset, \n",
    "    batch_size=cfg.BATCH_SIZE, \n",
    "    shuffle =cfg.SHUFFLE, \n",
    "    num_workers=cfg.NUM_WORKERS\n",
    ")\n",
    "\n",
    "test_loader = DataLoader(\n",
    "    test_dataset, \n",
    "    batch_size=cfg.BATCH_SIZE, \n",
    "    shuffle =cfg.SHUFFLE, \n",
    "    num_workers=cfg.NUM_WORKERS\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1d2f276",
   "metadata": {},
   "source": [
    "# Helpers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "id": "39b77d7e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_epoch(epoch, model, train_loader, criterion, optimizer, device, log_every_n_batches=100, writer=None):\n",
    "    \"\"\"Train the model for 1 epoch\n",
    "    Args:\n",
    "        epoch: Current training epoch\n",
    "        model: nn.Module\n",
    "        train_loader: train DataLoader\n",
    "        criterion: callable loss function\n",
    "        optimizer: pytorch optimizer\n",
    "        device: torch.device\n",
    "        log_every_n_batches: Metrics will log every n batches\n",
    "        writer: Tensorboard SummaryWriter\n",
    "    Returns\n",
    "    -------\n",
    "    Tuple[Float, Float]\n",
    "        average train loss and average train accuracy for current epoch\n",
    "    \"\"\"\n",
    "    \n",
    "    train_losses = []\n",
    "    train_corrects = []\n",
    "    running_losses = 0.0\n",
    "    running_corrects = 0.0\n",
    "    total = 0\n",
    "    model.train()\n",
    "\n",
    "    # Iterate over data.\n",
    "    for batch_idx, (inputs, labels) in enumerate(train_loader):\n",
    "        inputs = inputs.to(device)\n",
    "        labels = labels.to(device)\n",
    "\n",
    "        # prediction\n",
    "        outputs = model(inputs)\n",
    "\n",
    "        # calculate loss & number of corrects for current batch\n",
    "        _, preds = torch.max(outputs, 1)\n",
    "        loss = criterion(outputs, labels)\n",
    "        n_corrects = torch.sum(preds == labels.data).item()\n",
    "\n",
    "        # backprop\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        # calculate runnings_loss, running_correct and total if log_every_n_batches > 0\n",
    "        if log_every_n_batches:\n",
    "            running_losses += loss.item()\n",
    "            running_corrects += n_corrects\n",
    "            total += labels.size(0)\n",
    "            \n",
    "            if batch_idx % log_every_n_batches == 0:\n",
    "                print(f\"loss: {loss.item():>7f}  [{batch_idx:>5d}/{len(train_loader):>5d}]\")\n",
    "                \n",
    "                # write to tensorboard\n",
    "                if writer:\n",
    "                    writer.add_scalar(\"train_loss\", \n",
    "                                      running_losses/log_every_n_batches,\n",
    "                                      global_step=epoch*len(train_loader)+batch_idx)\n",
    "                    writer.add_scalar(\"train_accuracy\", \n",
    "                                      running_corrects/total,\n",
    "                                      global_step=epoch*len(train_loader)+batch_idx)\n",
    "                \n",
    "                # reset to zero values for next logging\n",
    "                running_losses = 0.0\n",
    "                running_corrects = 0.0\n",
    "                total = 0\n",
    "            \n",
    "        # accumulate loss and n_corrects for current batch for later average metric calculation\n",
    "        train_losses.append(loss.item())\n",
    "        train_corrects.append(n_corrects)\n",
    "        \n",
    "    ave_train_loss = sum(train_losses)/len(train_losses)\n",
    "    ave_train_accuracy = sum(train_corrects)/len(train_loader.dataset)      \n",
    "\n",
    "    return ave_train_loss, ave_train_accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "id": "1afae531",
   "metadata": {},
   "outputs": [],
   "source": [
    "def val_epoch(epoch, model, val_loader, criterion, device, log_every_n_batches=100, writer=None):\n",
    "    \"\"\"Validate the model for 1 epoch\n",
    "    Args:\n",
    "        epoch: Current validation epoch\n",
    "        model: nn.Module\n",
    "        val_loader: val DataLoader\n",
    "        criterion: callable loss function\n",
    "        device: torch.device\n",
    "        log_every_n_batches: Metrics will log every n batches\n",
    "        writer: Tensorboard SummaryWriter\n",
    "    Returns\n",
    "    -------\n",
    "    Tuple[Float, Float]\n",
    "        average val loss and average val accuracy for current epoch\n",
    "    \"\"\"\n",
    "\n",
    "    val_losses = []\n",
    "    val_corrects = []\n",
    "    running_losses = 0.0\n",
    "    running_corrects = 0.0\n",
    "    total = 0\n",
    "    model.eval()\n",
    "\n",
    "    # Iterate over data\n",
    "    with torch.no_grad():\n",
    "        for batch_idx, (inputs, labels) in enumerate(val_loader):\n",
    "            inputs = inputs.to(device)\n",
    "            labels = labels.to(device)\n",
    "\n",
    "            # prediction\n",
    "            outputs = model(inputs)\n",
    "\n",
    "            # calculate loss & number of corrects for current batch\n",
    "            _, preds = torch.max(outputs, 1)\n",
    "            loss = criterion(outputs, labels)\n",
    "            n_corrects = torch.sum(preds == labels.data).item()\n",
    "            \n",
    "            # calculate runnings_loss, running_correct and total if log_every_n_batches > 0\n",
    "            if log_every_n_batches:\n",
    "                running_losses += loss.item()\n",
    "                running_corrects += n_corrects\n",
    "                total += labels.size(0)\n",
    "\n",
    "                if batch_idx % log_every_n_batches == 0:\n",
    "                    if writer:\n",
    "                        writer.add_scalar(\"val_loss\", \n",
    "                                          running_losses/log_every_n_batches,\n",
    "                                          global_step=epoch*len(val_loader)+batch_idx)\n",
    "                        writer.add_scalar(\"val_accuracy\", \n",
    "                                          running_corrects/total,\n",
    "                                          global_step=epoch*len(val_loader)+batch_idx)\n",
    "\n",
    "                    # reset to zero values for next logging\n",
    "                    running_losses = 0.0\n",
    "                    running_corrects = 0.0\n",
    "                    total = 0\n",
    "                \n",
    "            # accumulate loss and n_corrects for current batch for later average metric calculation\n",
    "            val_losses.append(loss.item())\n",
    "            val_corrects.append(n_corrects)\n",
    "            \n",
    "        ave_val_loss = sum(val_losses)/len(val_losses)\n",
    "        ave_val_accuracy = sum(val_corrects)/len(val_loader.dataset)\n",
    "\n",
    "    return ave_val_loss, ave_val_accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03aa817a",
   "metadata": {},
   "source": [
    "# Training and Evaluating Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "id": "19b36628",
   "metadata": {},
   "outputs": [],
   "source": [
    "# load pretrained model\n",
    "model = models.mobilenet.mobilenet_v2(pretrained=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "id": "888b5bca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# freeze all layers\n",
    "for param in model.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "num_features = model.classifier[1].in_features\n",
    "model.classifier[1] = nn.Linear(num_features, len(classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "id": "efe3cfd0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using cuda\n"
     ]
    }
   ],
   "source": [
    "# transfer to cuda device if any\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = model.to(device)\n",
    "print(f\"Using {device}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "id": "1c76f4fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# loss, optimizer and scheduler \n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = Adam(model.parameters(), lr=cfg.LR)\n",
    "scheduler = lr_scheduler.ReduceLROnPlateau(optimizer, mode='min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "id": "a3a93309",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1\n",
      "-------------------------------\n",
      "loss: 2.197984  [    0/ 3589]\n",
      "8\n",
      "loss: 1.622092  [  100/ 3589]\n",
      "800\n",
      "loss: 2.288800  [  200/ 3589]\n",
      "800\n",
      "loss: 1.854156  [  300/ 3589]\n",
      "800\n",
      "loss: 1.427849  [  400/ 3589]\n",
      "800\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Input \u001b[1;32mIn [309]\u001b[0m, in \u001b[0;36m<cell line: 4>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mEpoch \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mepoch\u001b[38;5;241m+\u001b[39m\u001b[38;5;241m1\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m-------------------------------\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m      7\u001b[0m \u001b[38;5;66;03m# Train\u001b[39;00m\n\u001b[1;32m----> 8\u001b[0m ave_train_loss, ave_train_acc \u001b[38;5;241m=\u001b[39m \u001b[43mtrain_epoch\u001b[49m\u001b[43m(\u001b[49m\u001b[43mepoch\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrain_loader\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcriterion\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moptimizer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m100\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mwriter\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m      9\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mAverage TrainLoss: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mave_train_loss\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m.4f\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;130;01m\\t\u001b[39;00m\u001b[38;5;124mAverage TrainAcc: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mave_train_acc\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m.4f\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m     11\u001b[0m \u001b[38;5;66;03m# Evaluation\u001b[39;00m\n",
      "Input \u001b[1;32mIn [303]\u001b[0m, in \u001b[0;36mtrain_epoch\u001b[1;34m(epoch, model, train_loader, criterion, optimizer, device, log_every_n_batches, writer)\u001b[0m\n\u001b[0;32m     23\u001b[0m model\u001b[38;5;241m.\u001b[39mtrain()\n\u001b[0;32m     25\u001b[0m \u001b[38;5;66;03m# Iterate over data.\u001b[39;00m\n\u001b[1;32m---> 26\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m batch_idx, (inputs, labels) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(train_loader):\n\u001b[0;32m     27\u001b[0m     inputs \u001b[38;5;241m=\u001b[39m inputs\u001b[38;5;241m.\u001b[39mto(device)\n\u001b[0;32m     28\u001b[0m     labels \u001b[38;5;241m=\u001b[39m labels\u001b[38;5;241m.\u001b[39mto(device)\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torch\\utils\\data\\dataloader.py:530\u001b[0m, in \u001b[0;36m_BaseDataLoaderIter.__next__\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    528\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_sampler_iter \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m    529\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_reset()\n\u001b[1;32m--> 530\u001b[0m data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_next_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    531\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_num_yielded \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[0;32m    532\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_dataset_kind \u001b[38;5;241m==\u001b[39m _DatasetKind\u001b[38;5;241m.\u001b[39mIterable \u001b[38;5;129;01mand\u001b[39;00m \\\n\u001b[0;32m    533\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_IterableDataset_len_called \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m \\\n\u001b[0;32m    534\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_num_yielded \u001b[38;5;241m>\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_IterableDataset_len_called:\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torch\\utils\\data\\dataloader.py:570\u001b[0m, in \u001b[0;36m_SingleProcessDataLoaderIter._next_data\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    568\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_next_data\u001b[39m(\u001b[38;5;28mself\u001b[39m):\n\u001b[0;32m    569\u001b[0m     index \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_next_index()  \u001b[38;5;66;03m# may raise StopIteration\u001b[39;00m\n\u001b[1;32m--> 570\u001b[0m     data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_dataset_fetcher\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfetch\u001b[49m\u001b[43m(\u001b[49m\u001b[43mindex\u001b[49m\u001b[43m)\u001b[49m  \u001b[38;5;66;03m# may raise StopIteration\u001b[39;00m\n\u001b[0;32m    571\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_pin_memory:\n\u001b[0;32m    572\u001b[0m         data \u001b[38;5;241m=\u001b[39m _utils\u001b[38;5;241m.\u001b[39mpin_memory\u001b[38;5;241m.\u001b[39mpin_memory(data)\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torch\\utils\\data\\_utils\\fetch.py:49\u001b[0m, in \u001b[0;36m_MapDatasetFetcher.fetch\u001b[1;34m(self, possibly_batched_index)\u001b[0m\n\u001b[0;32m     47\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mfetch\u001b[39m(\u001b[38;5;28mself\u001b[39m, possibly_batched_index):\n\u001b[0;32m     48\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mauto_collation:\n\u001b[1;32m---> 49\u001b[0m         data \u001b[38;5;241m=\u001b[39m [\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdataset[idx] \u001b[38;5;28;01mfor\u001b[39;00m idx \u001b[38;5;129;01min\u001b[39;00m possibly_batched_index]\n\u001b[0;32m     50\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m     51\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdataset[possibly_batched_index]\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torch\\utils\\data\\_utils\\fetch.py:49\u001b[0m, in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m     47\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mfetch\u001b[39m(\u001b[38;5;28mself\u001b[39m, possibly_batched_index):\n\u001b[0;32m     48\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mauto_collation:\n\u001b[1;32m---> 49\u001b[0m         data \u001b[38;5;241m=\u001b[39m [\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdataset\u001b[49m\u001b[43m[\u001b[49m\u001b[43midx\u001b[49m\u001b[43m]\u001b[49m \u001b[38;5;28;01mfor\u001b[39;00m idx \u001b[38;5;129;01min\u001b[39;00m possibly_batched_index]\n\u001b[0;32m     50\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m     51\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdataset[possibly_batched_index]\n",
      "Input \u001b[1;32mIn [299]\u001b[0m, in \u001b[0;36mEmotionDataset.__getitem__\u001b[1;34m(self, idx)\u001b[0m\n\u001b[0;32m     28\u001b[0m labels \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mfrom_numpy(labels)\u001b[38;5;241m.\u001b[39mlong()\n\u001b[0;32m     30\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtransform :\n\u001b[1;32m---> 31\u001b[0m     img \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtransform\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimg\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     32\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m img, labels\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torchvision\\transforms\\transforms.py:95\u001b[0m, in \u001b[0;36mCompose.__call__\u001b[1;34m(self, img)\u001b[0m\n\u001b[0;32m     93\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__call__\u001b[39m(\u001b[38;5;28mself\u001b[39m, img):\n\u001b[0;32m     94\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m t \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtransforms:\n\u001b[1;32m---> 95\u001b[0m         img \u001b[38;5;241m=\u001b[39m \u001b[43mt\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimg\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     96\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m img\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torchvision\\transforms\\transforms.py:135\u001b[0m, in \u001b[0;36mToTensor.__call__\u001b[1;34m(self, pic)\u001b[0m\n\u001b[0;32m    127\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__call__\u001b[39m(\u001b[38;5;28mself\u001b[39m, pic):\n\u001b[0;32m    128\u001b[0m     \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m    129\u001b[0m \u001b[38;5;124;03m    Args:\u001b[39;00m\n\u001b[0;32m    130\u001b[0m \u001b[38;5;124;03m        pic (PIL Image or numpy.ndarray): Image to be converted to tensor.\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    133\u001b[0m \u001b[38;5;124;03m        Tensor: Converted image.\u001b[39;00m\n\u001b[0;32m    134\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[1;32m--> 135\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mF\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mto_tensor\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpic\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mc:\\users\\ivand\\projects\\ai\\emotion-recognition\\venv\\lib\\site-packages\\torchvision\\transforms\\functional.py:155\u001b[0m, in \u001b[0;36mto_tensor\u001b[1;34m(pic)\u001b[0m\n\u001b[0;32m    153\u001b[0m img \u001b[38;5;241m=\u001b[39m img\u001b[38;5;241m.\u001b[39mpermute((\u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m0\u001b[39m, \u001b[38;5;241m1\u001b[39m))\u001b[38;5;241m.\u001b[39mcontiguous()\n\u001b[0;32m    154\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(img, torch\u001b[38;5;241m.\u001b[39mByteTensor):\n\u001b[1;32m--> 155\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mimg\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mto\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdtype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdefault_float_dtype\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdiv\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m255\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[0;32m    156\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m    157\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m img\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# Start training and val loops\n",
    "\n",
    "since = time.time()\n",
    "for epoch in range(cfg.EPOCHS):\n",
    "    print(f\"Epoch {epoch+1}\\n-------------------------------\")\n",
    "    \n",
    "    # Train\n",
    "    ave_train_loss, ave_train_acc = train_epoch(epoch, model, train_loader, criterion, optimizer, device, 100, writer)\n",
    "    print(f'Average TrainLoss: {ave_train_loss:.4f} \\tAverage TrainAcc: {ave_train_acc:.4f}')\n",
    "    \n",
    "    # Evaluation\n",
    "    print(\"Evaluation:\")\n",
    "    ave_val_loss, ave_val_acc = val_epoch(epoch, model, val_loader, criterion, device, 100, writer)\n",
    "    print(f'Average ValLoss: {ave_val_loss:.4f} \\tAverage ValAcc: {ave_val_acc:.4f}')\n",
    "    \n",
    "    # log metrics at every epoch\n",
    "#     if writer:\n",
    "#         writer.add_scalar(\"train_loss\", ave_train_loss, epoch)\n",
    "#         writer.add_scalar(\"train_accuracy\", ave_train_acc, epoch)\n",
    "#         writer.add_scalar(\"val_loss\", ave_val_loss, epoch)\n",
    "#         writer.add_scalar(\"val_accuracy\", ave_val_acc, epoch)\n",
    "    \n",
    "    # schedule lr\n",
    "    scheduler.step(ave_val_loss)\n",
    "    print(\"\\n\")\n",
    "        \n",
    "time_elapsed = time.time() - since\n",
    "print(f\"Training complete in {time_elapsed//60:.0f}m {time_elapsed%60:.0f}s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1c2785c1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f5aff473",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}