SCNN_MNIST.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "0ZnKUkkN52Ua"
      },
      "source": [
        "<font color=\"#e32b3d\"><h1>Shallow Neural Networks for Classification</h2></font>\n",
        "\n",
        "<font color=\"#822b34\"><h2>*MNIST Dataset*</h2></font>"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "c3BDQvldVyPH"
      },
      "source": [
        "<font color=\"#2ff9b3d\"><h1>Part 1: Importing Datasets and Pre-processing </h1></font>\n",
        "\n",
        "required library and packages are imported."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "-2FMOPeHVyPI"
      },
      "outputs": [],
      "source": [
        "# Import library\n",
        "import torch\n",
        "from torch import nn\n",
        "import torch.nn.functional as F\n",
        "from torch.utils.data import DataLoader, Dataset\n",
        "\n",
        "import torchvision\n",
        "from torchvision import transforms\n",
        "from torchvision.datasets import ImageFolder\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from tqdm import tqdm\n",
        "from PIL import Image\n",
        "import cv2\n",
        "import glob\n",
        "import os"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "xsV0Pa_VQemk"
      },
      "outputs": [],
      "source": [
        "##Constants\n",
        "batch_size = 128\n",
        "num_of_classes = 10\n",
        "dim_MNIST = 5\n",
        "dim_CIFAR = 6\n",
        "in_shape_MNIST = 1\n",
        "in_shape_CIFAR = 3"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "A7pCgzj2hMwI"
      },
      "source": [
        "Importing the MNIST dataset into train and test sets.\n",
        "\n",
        "\n",
        "note that using torch.vision requiers to transform dataset into tensors using *ToTensor()*, this function normalizes all the features in dataset and no further normalization is needed."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 821
        },
        "id": "uoSSTRdWVyPK",
        "outputId": "dc6373cb-1ef5-46b5-f51f-239fe34d7a61"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to ./MNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 9912422/9912422 [00:00<00:00, 226843835.05it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Extracting ./MNIST/raw/train-images-idx3-ubyte.gz to ./MNIST/raw\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to ./MNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 28881/28881 [00:00<00:00, 95683802.39it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Extracting ./MNIST/raw/train-labels-idx1-ubyte.gz to ./MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to ./MNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 1648877/1648877 [00:00<00:00, 66504071.43it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Extracting ./MNIST/raw/t10k-images-idx3-ubyte.gz to ./MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to ./MNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 4542/4542 [00:00<00:00, 4549923.28it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Extracting ./MNIST/raw/t10k-labels-idx1-ubyte.gz to ./MNIST/raw\n",
            "\n",
            "Image size: torch.Size([1, 28, 28])\n",
            "Label: 1\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "## importing MNIST\n",
        "MNIST_Train_set = torchvision.datasets.MNIST(root=\"./\", download=True,train=True, transform=transforms.Compose([transforms.ToTensor(), transforms.RandomHorizontalFlip(p=0.5)]))\n",
        "MNIST_test_set = torchvision.datasets.MNIST(root=\"./\", download=True,train=False, transform=transforms.Compose([transforms.ToTensor(), transforms.RandomHorizontalFlip(p=0.5)]))\n",
        "dff = pd.DataFrame(MNIST_Train_set.targets.numpy(), columns=[\"labels\"])\n",
        "MNIST_classes = list(set(dff['labels'].values))\n",
        "##Creating DataLoaders\n",
        "MNIST_train_dataloader = DataLoader(\n",
        "    dataset = MNIST_Train_set,\n",
        "    batch_size = batch_size,\n",
        "    shuffle = True\n",
        ")\n",
        "\n",
        "MNIST_test_dataloader = DataLoader(\n",
        "    dataset = MNIST_test_set,\n",
        "    batch_size = batch_size\n",
        ")\n",
        "##Visualizing a sample\n",
        "MNIST_train_features_batch, MNIST_train_labels_batch = next(iter(MNIST_train_dataloader))\n",
        "MNIST_train_features_batch.shape, MNIST_train_labels_batch.shape\n",
        "img, label = MNIST_train_features_batch[0], MNIST_train_labels_batch[0]\n",
        "mean, std = 0.5, 0.5\n",
        "img = std * img + mean\n",
        "plt.imshow(img.permute(1, 2, 0), cmap='gray')\n",
        "plt.title(MNIST_classes[label])\n",
        "plt.axis(\"Off\");\n",
        "print(f\"Image size: {img.shape}\")\n",
        "print(f\"Label: {label}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3DJVwaToXIu_",
        "outputId": "54b56264-ce37-46c1-e141-0a4216da2b63"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "cuda\n"
          ]
        }
      ],
      "source": [
        "# We move our tensor to the GPU if available\n",
        "if (torch.cuda.is_available()):\n",
        "    DEVICE = 'cuda'\n",
        "else:\n",
        "    DEVICE = 'cpu'\n",
        "print(DEVICE)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "rqi6Fek8hMwW"
      },
      "source": [
        "<font color=\"#2ff9b3d\"><h1>Part 3: SCNNB Implementation </h1></font>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "6evo0QhzVyPT"
      },
      "outputs": [],
      "source": [
        "# Create a convolutional neural network : SCNNB\n",
        "class SCNNB(nn.Module):\n",
        "    def __init__(self, input_shape, dim,  output_shape):\n",
        "        super().__init__()\n",
        "        \n",
        "        self.CNN_block = nn.Sequential(\n",
        "            nn.Conv2d(input_shape, 32, kernel_size = 3), \n",
        "            nn.BatchNorm2d(num_features=32),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2),\n",
        "            nn.Conv2d(32, 64, kernel_size = 3),\n",
        "            nn.BatchNorm2d(num_features=64),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2)\n",
        "\n",
        "        )\n",
        "\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Linear(in_features = 64*dim*dim, out_features = 1280),\n",
        "            nn.ReLU(),\n",
        "            nn.Dropout(p=0.5),\n",
        "            nn.Linear(in_features = 1280, out_features = output_shape),\n",
        "            nn.Softmax(dim=1)\n",
        "        )\n",
        "    \n",
        "    def forward(self, x: torch.Tensor):\n",
        "        x = self.CNN_block(x)\n",
        "        # print(x.shape)\n",
        "        x = self.classifier(x)\n",
        "        # print(x.shape)\n",
        "        return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "MmQMVySpzeay"
      },
      "outputs": [],
      "source": [
        "# Create a convolutional neural network : SCNNB-a\n",
        "class SCNNB_a(nn.Module):\n",
        "    def __init__(self, input_shape, dim, output_shape):\n",
        "        super().__init__()\n",
        "        \n",
        "        self.CNN_block = nn.Sequential(\n",
        "            nn.Conv2d(input_shape, 32, kernel_size = 3), \n",
        "            # nn.BatchNorm2d(num_features=32),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2),\n",
        "            nn.Conv2d(32, 64, kernel_size = 3),\n",
        "            nn.BatchNorm2d(num_features=64),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2)\n",
        "\n",
        "        )\n",
        "\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Linear(in_features = 64*dim*dim, out_features = 1280),\n",
        "            nn.ReLU(),\n",
        "            nn.Dropout(p=0.5),\n",
        "            nn.Linear(in_features = 1280, out_features = output_shape),\n",
        "            nn.Softmax(dim=1)\n",
        "        )\n",
        "    \n",
        "    def forward(self, x: torch.Tensor):\n",
        "        x = self.CNN_block(x)\n",
        "        # print(x.shape)\n",
        "        x = self.classifier(x)\n",
        "        # print(x.shape)\n",
        "        return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "Vo611YYBzj6j"
      },
      "outputs": [],
      "source": [
        "# Create a convolutional neural network : SCNNB-b\n",
        "class SCNNB_b(nn.Module):\n",
        "    def __init__(self, input_shape, dim, output_shape):\n",
        "        super().__init__()\n",
        "        \n",
        "        self.CNN_block = nn.Sequential(\n",
        "            nn.Conv2d(input_shape, 32, kernel_size = 3), \n",
        "            # nn.BatchNorm2d(num_features=32),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2),\n",
        "            nn.Conv2d(32, 64, kernel_size = 3),\n",
        "            # nn.BatchNorm2d(num_features=64),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2)\n",
        "\n",
        "        )\n",
        "\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Linear(in_features = 64*dim*dim, out_features = 1280),\n",
        "            nn.ReLU(),\n",
        "            nn.Dropout(p=0.5),\n",
        "            nn.Linear(in_features = 1280, out_features = output_shape),\n",
        "            nn.Softmax(dim=1)\n",
        "        )\n",
        "    \n",
        "    def forward(self, x: torch.Tensor):\n",
        "        x = self.CNN_block(x)\n",
        "        # print(x.shape)\n",
        "        x = self.classifier(x)\n",
        "        # print(x.shape)\n",
        "        return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1olTY3XyhMwX",
        "outputId": "02fcf591-384d-4ad1-a3da-0fd7dbd15fb2"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "SCNNB(\n",
              "  (CNN_block): Sequential(\n",
              "    (0): Conv2d(1, 32, kernel_size=(3, 3), stride=(1, 1))\n",
              "    (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "    (3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (4): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1))\n",
              "    (5): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (6): ReLU()\n",
              "    (7): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (classifier): Sequential(\n",
              "    (0): Flatten(start_dim=1, end_dim=-1)\n",
              "    (1): Linear(in_features=1600, out_features=1280, bias=True)\n",
              "    (2): ReLU()\n",
              "    (3): Dropout(p=0.5, inplace=False)\n",
              "    (4): Linear(in_features=1280, out_features=10, bias=True)\n",
              "    (5): Softmax(dim=1)\n",
              "  )\n",
              ")"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "## Sample Model Description\n",
        "model_SCNNB = SCNNB(input_shape=1, dim=5, output_shape=num_of_classes)\n",
        "# model_SCNNB_a = SCNNB_a(input_shape=1, output_shape=len(num_of_classes))\n",
        "# model_SCNNB_b = SCNNB_b(input_shape=1, output_shape=len(num_of_classes))\n",
        "\n",
        "model_SCNNB"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "3RJROgg9hMwZ"
      },
      "outputs": [],
      "source": [
        "def train_step(model,\n",
        "               data_loader,\n",
        "               loss_fn,\n",
        "               optimizer,\n",
        "               accuracy_fn,\n",
        "               device):\n",
        "    \n",
        "    # Send model to GPU\n",
        "    model.train()\n",
        "    model = model.to(device)\n",
        "    train_loss, train_acc, counter = 0, 0, 0\n",
        "    with tqdm(data_loader, desc =\"  train\") as train_tqdm:\n",
        "        for X, y in train_tqdm:\n",
        "            # Send data to GPU\n",
        "            X, y = X.to(device), y.to(device)\n",
        "\n",
        "            # 1. Forward pass\n",
        "            y_pred = model(X)\n",
        "\n",
        "            # 2. Calculate loss\n",
        "            loss = loss_fn(y_pred, y)\n",
        "            train_loss += loss\n",
        "            train_acc += accuracy_fn(true=y.cpu(),\n",
        "                                     pred=y_pred.cpu())\n",
        "            counter += 1\n",
        "            train_tqdm.set_postfix(train_acc=\"{:.3f}\".format(train_acc/counter), train_loss=train_loss.item()/counter, refresh=False)\n",
        "\n",
        "            # 3. Optimizer zero grad\n",
        "            optimizer.zero_grad()\n",
        "\n",
        "            # 4. Loss backward\n",
        "            loss.backward()\n",
        "\n",
        "            # 5. Optimizer step\n",
        "            optimizer.step()\n",
        "\n",
        "        # Calculate loss and accuracy per epoch and print out what's happening\n",
        "        train_loss /= len(data_loader)\n",
        "        train_acc /= len(data_loader)\n",
        "    return train_loss.item(), train_acc\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "YQft1sVthMwa"
      },
      "outputs": [],
      "source": [
        "def test_step(model,\n",
        "              data_loader,\n",
        "              loss_fn,\n",
        "              accuracy_fn,\n",
        "              device):\n",
        "    \n",
        "    # Send model to GPU\n",
        "    model.to(device)\n",
        "\n",
        "    # put model in eval mode\n",
        "    model.eval() \n",
        "    test_loss, test_acc, counter = 0, 0, 0\n",
        "    with tqdm(data_loader, desc =\"   test\") as test_tqdm:\n",
        "        for X, y in test_tqdm:\n",
        "            # Send data to GPU\n",
        "            X, y = X.to(device), y.to(device)\n",
        "            \n",
        "            # 1. Forward pass\n",
        "            test_pred = model(X)\n",
        "            \n",
        "            # 2. Calculate loss and accuracy\n",
        "            test_loss += loss_fn(test_pred, y)\n",
        "            test_acc += accuracy_fn(true=y.cpu(),\n",
        "                                    pred=test_pred.cpu())\n",
        "            counter += 1\n",
        "            test_tqdm.set_postfix(test_acc=\"{:.3f}\".format(test_acc/counter), test_loss=test_loss.item()/counter, refresh=True)\n",
        "\n",
        "        # Adjust metrics and print out\n",
        "        test_loss /= len(data_loader)\n",
        "        test_acc  /= len(data_loader)\n",
        "            \n",
        "    return test_loss.item(), test_acc"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "uNlaMz7EhMwb"
      },
      "outputs": [],
      "source": [
        "def train(hparams, train_data_loader, test_data_loader):\n",
        "  # Datasets: \n",
        "  if (hparams[0] == 'MNIST' or hparams[0] == 'FMNIST' ):\n",
        "    in_shape = in_shape_MNIST\n",
        "    dim = dim_MNIST\n",
        "  elif hparams[0] == 'CIFAR' :\n",
        "    in_shape = in_shape_CIFAR\n",
        "    dim = dim_CIFAR\n",
        "\n",
        "  # Models:\n",
        "  if hparams[1] == 'SCNNB':\n",
        "    model = SCNNB(input_shape=in_shape, dim=dim, output_shape=num_of_classes)\n",
        "  elif hparams[1] == 'SCNNB-a':\n",
        "    model = SCNNB_a(input_shape=in_shape, dim=dim, output_shape=num_of_classes)\n",
        "  elif hparams[1] == 'SCNNB-b':\n",
        "    model = SCNNB_b(input_shape=in_shape, dim=dim, output_shape=num_of_classes)\n",
        "\n",
        "  loss_fn = nn.CrossEntropyLoss()\n",
        "  optimizer = torch.optim.SGD(model.parameters(), lr = 0.02, momentum=0.9, weight_decay=0.000005)\n",
        "  device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
        "  def accuracy_fn(true, pred):\n",
        "      pred = F.softmax(pred, dim = 1)\n",
        "      true = torch.zeros(pred.shape[0], pred.shape[1]).scatter_(1, true.unsqueeze(1), 1.)\n",
        "      acc = (true.argmax(-1) == pred.argmax(-1)).float().detach().numpy()\n",
        "      acc = float((100 * acc.sum()) / len(acc))\n",
        "      return round(acc, 4)\n",
        "  # Measure time\n",
        "  from timeit import default_timer as timer\n",
        "\n",
        "  train_epoch_loss = []\n",
        "  train_epoch_accuracy = []\n",
        "  test_epoch_loss = []\n",
        "  test_epoch_accuracy = []\n",
        "\n",
        "  train_time_start = timer()\n",
        "\n",
        "  epochs = hparams[2]\n",
        "  for epoch in range(epochs):\n",
        "      print(\"EPOCH {}/{} :\".format(epoch +1, epochs))\n",
        "      train_loss, train_acc = train_step(data_loader=train_data_loader,\n",
        "                                                model=model, \n",
        "                                                loss_fn=loss_fn,\n",
        "                                                optimizer=optimizer,\n",
        "                                                accuracy_fn=accuracy_fn,\n",
        "                                                device=device\n",
        "                                                )\n",
        "      test_loss, test_acc = test_step(data_loader=test_data_loader,\n",
        "                                      model=model,\n",
        "                                      loss_fn=loss_fn,\n",
        "                                      accuracy_fn=accuracy_fn,\n",
        "                                      device=device\n",
        "                                      )\n",
        "      \n",
        "      train_epoch_loss.append(train_loss)\n",
        "      train_epoch_accuracy.append(train_acc)\n",
        "      test_epoch_loss.append(test_loss)\n",
        "      test_epoch_accuracy.append(test_acc)\n",
        "\n",
        "  train_time_end = timer()\n",
        "  total_time = train_time_end - train_time_start\n",
        "  print(f\"\\n\\nTrain time: {total_time:.3f} seconds\")\n",
        "  return model, train_epoch_loss, train_epoch_accuracy, test_epoch_loss, test_epoch_accuracy"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m_52gQO5bB3s",
        "outputId": "6b0fb5f5-9c50-4e8d-8f3f-14ffdff50505"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 1/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:20<00:00, 22.84it/s, train_acc=73.374, train_loss=1.74]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.27it/s, test_acc=81.922, test_loss=1.65]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 2/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.28it/s, train_acc=89.878, train_loss=1.57]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.38it/s, test_acc=95.945, test_loss=1.51]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 3/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.50it/s, train_acc=96.260, train_loss=1.5]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.41it/s, test_acc=96.509, test_loss=1.5]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 4/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.22it/s, train_acc=97.090, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.51it/s, test_acc=97.903, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 5/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.32it/s, train_acc=97.420, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.91it/s, test_acc=97.992, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 6/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.48it/s, train_acc=97.881, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.60it/s, test_acc=97.903, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 7/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.87it/s, train_acc=97.957, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.31it/s, test_acc=98.141, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 8/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.97it/s, train_acc=98.173, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.27it/s, test_acc=98.497, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 9/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.58it/s, train_acc=98.331, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.40it/s, test_acc=98.447, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 10/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.39it/s, train_acc=98.389, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.73it/s, test_acc=98.447, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 11/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.96it/s, train_acc=98.559, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.41it/s, test_acc=98.675, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 12/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.60it/s, train_acc=98.648, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.24it/s, test_acc=98.586, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 13/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.29it/s, train_acc=98.694, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.62it/s, test_acc=98.586, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 14/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.29it/s, train_acc=98.716, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.87it/s, test_acc=98.586, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 15/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.31it/s, train_acc=98.738, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.25it/s, test_acc=98.714, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 16/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.27it/s, train_acc=98.834, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.14it/s, test_acc=98.774, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 17/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.30it/s, train_acc=98.897, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.61it/s, test_acc=98.655, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 18/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.11it/s, train_acc=98.957, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.75it/s, test_acc=98.853, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 19/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.78it/s, train_acc=99.001, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.97it/s, test_acc=98.922, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 20/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.77it/s, train_acc=99.022, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.90it/s, test_acc=98.705, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 21/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.50it/s, train_acc=99.067, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.99it/s, test_acc=98.833, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 22/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.14it/s, train_acc=99.033, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.04it/s, test_acc=98.873, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 23/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.85it/s, train_acc=99.069, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.33it/s, test_acc=98.823, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 24/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.01it/s, train_acc=99.159, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.79it/s, test_acc=98.843, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 25/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.89it/s, train_acc=99.158, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.04it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 26/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.91it/s, train_acc=99.255, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.61it/s, test_acc=98.922, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 27/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.97it/s, train_acc=99.220, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.78it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 28/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.20it/s, train_acc=99.190, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.14it/s, test_acc=98.912, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 29/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.14it/s, train_acc=99.285, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.79it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 30/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.45it/s, train_acc=99.295, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.26it/s, test_acc=98.803, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 31/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.88it/s, train_acc=99.315, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.59it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 32/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.76it/s, train_acc=99.329, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 33/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.23it/s, train_acc=99.285, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:03<00:00, 20.93it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 34/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 34.59it/s, train_acc=99.325, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.25it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 35/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.69it/s, train_acc=99.370, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.02it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 36/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.36it/s, train_acc=99.377, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.49it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 37/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.17it/s, train_acc=99.423, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.09it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 38/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.74it/s, train_acc=99.342, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.22it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 39/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.29it/s, train_acc=99.434, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.02it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 40/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.29it/s, train_acc=99.424, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.28it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 41/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.59it/s, train_acc=99.455, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.84it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 42/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.11it/s, train_acc=99.470, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 43/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.42it/s, train_acc=99.499, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.28it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 44/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.01it/s, train_acc=99.487, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.05it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 45/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.70it/s, train_acc=99.515, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.97it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 46/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.28it/s, train_acc=99.493, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.03it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 47/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.49it/s, train_acc=99.506, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.34it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 48/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.40it/s, train_acc=99.555, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.81it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 49/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.08it/s, train_acc=99.556, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.98it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 50/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.28it/s, train_acc=99.526, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.71it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 51/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.20it/s, train_acc=99.565, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.62it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 52/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.60it/s, train_acc=99.575, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.66it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 53/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.00it/s, train_acc=99.551, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.22it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 54/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.65it/s, train_acc=99.582, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.70it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 55/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.90it/s, train_acc=99.595, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.06it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 56/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.67it/s, train_acc=99.612, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.52it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 57/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.92it/s, train_acc=99.590, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.68it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 58/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.03it/s, train_acc=99.605, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.40it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 59/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.90it/s, train_acc=99.641, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.44it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 60/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.53it/s, train_acc=99.660, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.24it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 61/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.73it/s, train_acc=99.663, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.93it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 62/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.97it/s, train_acc=99.669, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 63/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.87it/s, train_acc=99.667, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.55it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 64/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.52it/s, train_acc=99.687, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.68it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 65/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.94it/s, train_acc=99.680, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.14it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 66/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.91it/s, train_acc=99.686, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.32it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 67/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.46it/s, train_acc=99.667, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.62it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 68/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.09it/s, train_acc=99.670, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.69it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 69/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.78it/s, train_acc=99.723, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.26it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 70/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.91it/s, train_acc=99.713, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.11it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 71/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.83it/s, train_acc=99.710, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.51it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 72/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.24it/s, train_acc=99.740, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.54it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 73/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.87it/s, train_acc=99.705, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.07it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 74/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.55it/s, train_acc=99.732, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.88it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 75/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.73it/s, train_acc=99.737, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.07it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 76/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.30it/s, train_acc=99.750, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.55it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 77/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.04it/s, train_acc=99.770, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.41it/s, test_acc=99.288, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 78/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.05it/s, train_acc=99.750, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.99it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 79/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.82it/s, train_acc=99.747, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.06it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 80/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.72it/s, train_acc=99.753, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.05it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 81/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.51it/s, train_acc=99.763, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.64it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 82/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.12it/s, train_acc=99.742, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.92it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 83/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.47it/s, train_acc=99.783, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.94it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 84/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.92it/s, train_acc=99.762, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.86it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 85/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.19it/s, train_acc=99.763, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.68it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 86/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.42it/s, train_acc=99.788, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.12it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 87/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.62it/s, train_acc=99.778, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.77it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 88/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.37it/s, train_acc=99.803, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.73it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 89/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.21it/s, train_acc=99.785, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.02it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 90/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.93it/s, train_acc=99.787, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.98it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 91/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.07it/s, train_acc=99.785, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.44it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 92/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.93it/s, train_acc=99.800, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.61it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 93/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.35it/s, train_acc=99.815, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.37it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 94/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.19it/s, train_acc=99.812, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.66it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 95/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.45it/s, train_acc=99.800, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.63it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 96/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.37it/s, train_acc=99.802, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.79it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 97/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.34it/s, train_acc=99.832, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.95it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 98/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.31it/s, train_acc=99.828, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.07it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 99/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.58it/s, train_acc=99.843, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.31it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 100/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.13it/s, train_acc=99.838, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.48it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 101/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.93it/s, train_acc=99.828, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.82it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 102/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.18it/s, train_acc=99.805, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.60it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 103/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.19it/s, train_acc=99.813, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.17it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 104/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.33it/s, train_acc=99.833, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.81it/s, test_acc=99.248, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 105/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.55it/s, train_acc=99.847, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.57it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 106/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.68it/s, train_acc=99.840, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.59it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 107/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.20it/s, train_acc=99.822, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.35it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 108/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.87it/s, train_acc=99.840, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.28it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 109/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.35it/s, train_acc=99.833, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.78it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 110/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.42it/s, train_acc=99.835, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.67it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 111/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:19<00:00, 24.55it/s, train_acc=99.827, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.95it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 112/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.64it/s, train_acc=99.857, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.52it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 113/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.94it/s, train_acc=99.850, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.44it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 114/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.58it/s, train_acc=99.833, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.65it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 115/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.17it/s, train_acc=99.860, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.63it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 116/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 33.69it/s, train_acc=99.858, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.27it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 117/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.74it/s, train_acc=99.857, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.70it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 118/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.98it/s, train_acc=99.863, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.83it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 119/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.31it/s, train_acc=99.855, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.12it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 120/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.41it/s, train_acc=99.878, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.63it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 121/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.50it/s, train_acc=99.870, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.78it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 122/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.45it/s, train_acc=99.853, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.89it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 123/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.15it/s, train_acc=99.862, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.65it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 124/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.20it/s, train_acc=99.875, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.74it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 125/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.39it/s, train_acc=99.867, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.25it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 126/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.49it/s, train_acc=99.857, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.30it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 127/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.62it/s, train_acc=99.862, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.10it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 128/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.41it/s, train_acc=99.872, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.46it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 129/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.66it/s, train_acc=99.877, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.60it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 130/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.59it/s, train_acc=99.883, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.13it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 131/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.50it/s, train_acc=99.890, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.72it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 132/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.15it/s, train_acc=99.863, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.38it/s, test_acc=99.248, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 133/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.60it/s, train_acc=99.893, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.39it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 134/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.03it/s, train_acc=99.875, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.65it/s, test_acc=99.229, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 135/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.70it/s, train_acc=99.890, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.50it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 136/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.77it/s, train_acc=99.880, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.78it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 137/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.98it/s, train_acc=99.890, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.52it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 138/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.67it/s, train_acc=99.890, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.35it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 139/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.74it/s, train_acc=99.902, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.28it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 140/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.99it/s, train_acc=99.887, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.34it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 141/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.78it/s, train_acc=99.883, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.67it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 142/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.13it/s, train_acc=99.880, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.43it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 143/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.00it/s, train_acc=99.890, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.90it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 144/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.18it/s, train_acc=99.885, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.63it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 145/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.56it/s, train_acc=99.887, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.49it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 146/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.78it/s, train_acc=99.867, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.47it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 147/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.74it/s, train_acc=99.902, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.36it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 148/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.89it/s, train_acc=99.891, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.98it/s, test_acc=99.229, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 149/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.43it/s, train_acc=99.885, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.05it/s, test_acc=99.229, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 150/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.45it/s, train_acc=99.900, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.36it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Train time: 2327.876 seconds\n"
          ]
        }
      ],
      "source": [
        "hparams = [\"MNIST\", \"SCNNB\", 150]\n",
        "SCNNB_MNIST_model, SCNNB_MNIST_train_loss, SCNNB_MNIST_train_accuracy, SCNNB_MNIST_test_loss, SCNNB_MNIST_test_accuracy = train(hparams ,MNIST_train_dataloader, MNIST_test_dataloader)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "id": "1K4OXmd285Za"
      },
      "outputs": [],
      "source": [
        "torch.save(SCNNB_MNIST_model.state_dict(), 'SCNNB_MNIST')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UrfQCnYBQZP6",
        "outputId": "76220abe-0f56-41e3-a326-f90f514acfd1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Maximum Test Accuracy for SCNNB model is:  99.28798607594936\n"
          ]
        }
      ],
      "source": [
        "SCNNB_max_acc = max(SCNNB_MNIST_test_accuracy)\n",
        "SCNNB_MEAN = sum(SCNNB_MNIST_test_accuracy)/len(SCNNB_MNIST_test_accuracy)\n",
        "print(\"Maximum Test Accuracy for SCNNB model is: \", str(SCNNB_max_acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ag7qcRvH7-hI",
        "outputId": "15eb65bb-5312-4b32-8742-301abaec710c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 1/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.75it/s, train_acc=78.611, train_loss=1.69]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.81it/s, test_acc=87.045, test_loss=1.6]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 2/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.99it/s, train_acc=86.900, train_loss=1.6]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.65it/s, test_acc=88.321, test_loss=1.58]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 3/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.45it/s, train_acc=87.912, train_loss=1.58]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.12it/s, test_acc=88.835, test_loss=1.57]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 4/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.82it/s, train_acc=91.469, train_loss=1.55]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.13it/s, test_acc=97.172, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 5/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.29it/s, train_acc=97.033, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.71it/s, test_acc=98.032, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 6/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 34.53it/s, train_acc=97.542, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.81it/s, test_acc=98.072, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 7/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.07it/s, train_acc=97.835, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.53it/s, test_acc=98.329, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 8/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.98it/s, train_acc=98.019, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.28it/s, test_acc=98.180, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 9/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.31it/s, train_acc=98.217, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.42it/s, test_acc=98.101, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 10/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.39it/s, train_acc=98.316, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.39it/s, test_acc=98.289, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 11/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 36.07it/s, train_acc=98.446, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.27it/s, test_acc=98.339, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 12/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.65it/s, train_acc=98.523, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.11it/s, test_acc=98.536, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 13/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.97it/s, train_acc=98.560, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.00it/s, test_acc=98.497, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 14/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.89it/s, train_acc=98.697, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.42it/s, test_acc=98.487, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 15/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.49it/s, train_acc=98.782, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.19it/s, test_acc=98.655, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 16/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.68it/s, train_acc=98.756, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.28it/s, test_acc=98.803, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 17/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.12it/s, train_acc=98.836, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.23it/s, test_acc=98.566, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 18/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.72it/s, train_acc=98.892, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.73it/s, test_acc=98.685, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 19/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.65it/s, train_acc=98.936, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.78it/s, test_acc=98.645, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 20/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.03it/s, train_acc=98.952, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.60it/s, test_acc=98.794, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 21/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.10it/s, train_acc=99.059, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.15it/s, test_acc=98.616, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 22/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.38it/s, train_acc=99.057, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.50it/s, test_acc=98.566, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 23/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.41it/s, train_acc=99.070, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.03it/s, test_acc=98.863, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 24/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.83it/s, train_acc=99.093, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.02it/s, test_acc=98.833, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 25/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.87it/s, train_acc=99.107, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.70it/s, test_acc=98.833, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 26/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.48it/s, train_acc=99.190, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.04it/s, test_acc=98.863, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 27/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.68it/s, train_acc=99.228, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.18it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 28/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.02it/s, train_acc=99.223, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 29/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.93it/s, train_acc=99.185, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.92it/s, test_acc=98.873, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 30/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.14it/s, train_acc=99.275, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.64it/s, test_acc=98.902, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 31/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.40it/s, train_acc=99.281, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.57it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 32/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.80it/s, train_acc=99.257, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.80it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 33/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.75it/s, train_acc=99.335, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.24it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 34/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.73it/s, train_acc=99.316, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.98it/s, test_acc=98.823, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 35/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.22it/s, train_acc=99.365, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.40it/s, test_acc=98.823, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 36/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=99.344, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.00it/s, test_acc=98.942, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 37/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.61it/s, train_acc=99.349, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.83it/s, test_acc=98.942, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 38/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.67it/s, train_acc=99.425, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.47it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 39/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.84it/s, train_acc=99.429, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.32it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 40/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.70it/s, train_acc=99.444, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.43it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 41/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.41it/s, train_acc=99.426, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.87it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 42/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.49it/s, train_acc=99.489, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.22it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 43/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.55it/s, train_acc=99.443, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.93it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 44/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.78it/s, train_acc=99.474, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.91it/s, test_acc=98.922, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 45/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.59it/s, train_acc=99.535, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.19it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 46/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.60it/s, train_acc=99.499, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.66it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 47/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.30it/s, train_acc=99.502, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 26.98it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 48/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.17it/s, train_acc=99.551, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.21it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 49/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.10it/s, train_acc=99.552, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.18it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 50/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.47it/s, train_acc=99.534, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.41it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 51/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.34it/s, train_acc=99.592, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.24it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 52/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.97it/s, train_acc=99.585, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.75it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 53/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.26it/s, train_acc=99.586, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.92it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 54/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.20it/s, train_acc=99.582, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.83it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 55/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.57it/s, train_acc=99.613, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.75it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 56/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.33it/s, train_acc=99.600, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.18it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 57/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.71it/s, train_acc=99.653, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.43it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 58/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.51it/s, train_acc=99.659, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.50it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 59/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.11it/s, train_acc=99.642, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.63it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 60/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.36it/s, train_acc=99.659, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.66it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 61/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.46it/s, train_acc=99.665, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.92it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 62/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.25it/s, train_acc=99.629, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.84it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 63/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.76it/s, train_acc=99.663, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.36it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 64/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.63it/s, train_acc=99.680, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.49it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 65/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.99it/s, train_acc=99.689, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.97it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 66/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.73it/s, train_acc=99.692, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.82it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 67/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.87it/s, train_acc=99.692, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.96it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 68/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.30it/s, train_acc=99.715, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.24it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 69/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.61it/s, train_acc=99.704, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.97it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 70/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.82it/s, train_acc=99.683, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.88it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 71/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.86it/s, train_acc=99.723, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.62it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 72/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.60it/s, train_acc=99.707, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.02it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 73/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.35it/s, train_acc=99.728, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.69it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 74/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.90it/s, train_acc=99.748, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.21it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 75/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.16it/s, train_acc=99.755, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.76it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 76/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.28it/s, train_acc=99.737, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.03it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 77/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.30it/s, train_acc=99.763, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.18it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 78/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.76it/s, train_acc=99.765, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.14it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 79/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.31it/s, train_acc=99.783, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.88it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 80/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.85it/s, train_acc=99.773, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.75it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 81/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.67it/s, train_acc=99.778, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.07it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 82/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.56it/s, train_acc=99.763, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.63it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 83/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.48it/s, train_acc=99.783, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.21it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 84/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.92it/s, train_acc=99.795, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.21it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 85/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.67it/s, train_acc=99.777, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.60it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 86/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.09it/s, train_acc=99.805, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.57it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 87/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.47it/s, train_acc=99.766, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.13it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 88/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.26it/s, train_acc=99.797, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.98it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 89/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.16it/s, train_acc=99.787, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.24it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 90/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.00it/s, train_acc=99.808, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.25it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 91/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.04it/s, train_acc=99.792, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.00it/s, test_acc=99.179, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 92/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.24it/s, train_acc=99.818, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.42it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 93/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.25it/s, train_acc=99.805, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.19it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 94/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.48it/s, train_acc=99.817, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.15it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 95/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.61it/s, train_acc=99.820, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.70it/s, test_acc=99.239, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 96/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.96it/s, train_acc=99.820, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.06it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 97/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.59it/s, train_acc=99.813, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.83it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 98/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.18it/s, train_acc=99.832, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.71it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 99/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.00it/s, train_acc=99.820, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.34it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 100/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.83it/s, train_acc=99.842, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.26it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 101/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.24it/s, train_acc=99.830, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.50it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 102/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.69it/s, train_acc=99.842, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.75it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 103/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.38it/s, train_acc=99.840, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.86it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 104/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.39it/s, train_acc=99.838, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.33it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 105/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.42it/s, train_acc=99.825, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.99it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 106/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.16it/s, train_acc=99.842, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.93it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 107/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.39it/s, train_acc=99.858, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.34it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 108/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.46it/s, train_acc=99.855, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.86it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 109/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.39it/s, train_acc=99.833, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.39it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 110/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.21it/s, train_acc=99.852, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.98it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 111/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.45it/s, train_acc=99.863, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.80it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 112/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.75it/s, train_acc=99.850, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.27it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 113/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.36it/s, train_acc=99.867, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.22it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 114/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.17it/s, train_acc=99.850, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.73it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 115/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.43it/s, train_acc=99.863, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.43it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 116/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.29it/s, train_acc=99.853, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.82it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 117/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.64it/s, train_acc=99.862, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.40it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 118/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.21it/s, train_acc=99.850, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.77it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 119/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.18it/s, train_acc=99.870, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.69it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 120/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.93it/s, train_acc=99.853, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 121/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.51it/s, train_acc=99.858, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.55it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 122/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.35it/s, train_acc=99.865, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.88it/s, test_acc=99.169, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 123/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.63it/s, train_acc=99.873, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.58it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 124/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.21it/s, train_acc=99.867, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.47it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 125/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.83it/s, train_acc=99.870, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.58it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 126/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.80it/s, train_acc=99.873, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.24it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 127/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.58it/s, train_acc=99.858, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.67it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 128/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.85it/s, train_acc=99.882, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.80it/s, test_acc=99.159, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 129/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.38it/s, train_acc=99.865, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.01it/s, test_acc=99.239, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 130/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.79it/s, train_acc=99.876, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.57it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 131/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.67it/s, train_acc=99.868, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.11it/s, test_acc=99.278, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 132/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.26it/s, train_acc=99.872, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.69it/s, test_acc=99.248, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 133/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.57it/s, train_acc=99.870, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.35it/s, test_acc=99.239, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 134/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.83it/s, train_acc=99.893, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.35it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 135/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.79it/s, train_acc=99.887, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.49it/s, test_acc=99.248, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 136/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.59it/s, train_acc=99.893, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.65it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 137/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.70it/s, train_acc=99.881, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.53it/s, test_acc=99.248, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 138/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.70it/s, train_acc=99.872, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.08it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 139/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.87it/s, train_acc=99.877, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.94it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 140/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.58it/s, train_acc=99.873, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.70it/s, test_acc=99.130, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 141/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.36it/s, train_acc=99.878, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.35it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 142/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.45it/s, train_acc=99.885, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.25it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 143/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.67it/s, train_acc=99.883, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.17it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 144/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.74it/s, train_acc=99.877, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.43it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 145/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.31it/s, train_acc=99.897, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.26it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 146/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.06it/s, train_acc=99.892, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.39it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 147/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.69it/s, train_acc=99.887, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.78it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 148/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.18it/s, train_acc=99.888, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.35it/s, test_acc=99.209, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 149/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.17it/s, train_acc=99.885, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.97it/s, test_acc=99.140, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 150/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.02it/s, train_acc=99.900, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.66it/s, test_acc=99.219, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Train time: 2250.130 seconds\n"
          ]
        }
      ],
      "source": [
        "hparams = [\"MNIST\", \"SCNNB-a\", 150]\n",
        "SCNNB_a_MNIST_model, SCNNB_a_MNIST_train_loss, SCNNB_a_MNIST_train_accuracy, SCNNB_a_MNIST_test_loss, SCNNB_a_MNIST_test_accuracy = train(hparams ,MNIST_train_dataloader, MNIST_test_dataloader)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sgrCe5aIYWkv",
        "outputId": "b885112e-7a29-4ac6-e633-445dd627651f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Maximum Test Accuracy for SCNNB-a model is:  99.27809240506328\n"
          ]
        }
      ],
      "source": [
        "SCNNB_a_max_acc = max(SCNNB_a_MNIST_test_accuracy)\n",
        "SCNNB_a_MEAN = sum(SCNNB_a_MNIST_test_accuracy)/len(SCNNB_a_MNIST_test_accuracy)\n",
        "print(\"Maximum Test Accuracy for SCNNB-a model is: \", str(SCNNB_a_max_acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "tDJQ4aon8-Uq"
      },
      "outputs": [],
      "source": [
        "torch.save(SCNNB_a_MNIST_model.state_dict(), 'SCNNB_a_MNIST')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h-qfM7bC8Rgq",
        "outputId": "0f844b60-51a8-4348-b4ee-1c5fde13361d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 1/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.91it/s, train_acc=56.086, train_loss=1.92]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.28it/s, test_acc=80.963, test_loss=1.66]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 2/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.97it/s, train_acc=81.146, train_loss=1.65]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.24it/s, test_acc=84.405, test_loss=1.62]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 3/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.90it/s, train_acc=84.353, train_loss=1.62]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.11it/s, test_acc=86.689, test_loss=1.6]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 4/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.63it/s, train_acc=86.163, train_loss=1.6]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.02it/s, test_acc=91.564, test_loss=1.55]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 5/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.05it/s, train_acc=93.537, train_loss=1.53]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.86it/s, test_acc=95.431, test_loss=1.51]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 6/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.80it/s, train_acc=94.907, train_loss=1.52]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.34it/s, test_acc=96.608, test_loss=1.5]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 7/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.19it/s, train_acc=95.465, train_loss=1.51]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.01it/s, test_acc=96.934, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 8/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.59it/s, train_acc=96.063, train_loss=1.5]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.95it/s, test_acc=97.458, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 9/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.61it/s, train_acc=96.534, train_loss=1.5]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.97it/s, test_acc=97.686, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 10/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.05it/s, train_acc=96.765, train_loss=1.5]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.19it/s, test_acc=97.508, test_loss=1.49]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 11/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.73it/s, train_acc=97.060, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.31it/s, test_acc=97.735, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 12/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.93it/s, train_acc=97.204, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.08it/s, test_acc=97.824, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 13/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.08it/s, train_acc=97.390, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.58it/s, test_acc=97.933, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 14/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.85it/s, train_acc=97.531, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 38.00it/s, test_acc=98.032, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 15/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.15it/s, train_acc=97.724, train_loss=1.49]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.46it/s, test_acc=98.269, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 16/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=97.797, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.49it/s, test_acc=98.161, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 17/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.99it/s, train_acc=97.961, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.01it/s, test_acc=97.983, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 18/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.31it/s, train_acc=97.996, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.06it/s, test_acc=98.230, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 19/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.09it/s, train_acc=98.058, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.15it/s, test_acc=98.467, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 20/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.29it/s, train_acc=98.098, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.08it/s, test_acc=98.358, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 21/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.68it/s, train_acc=98.273, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.70it/s, test_acc=98.339, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 22/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.89it/s, train_acc=98.295, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.59it/s, test_acc=98.467, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 23/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.40it/s, train_acc=98.370, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.96it/s, test_acc=98.418, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 24/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.45it/s, train_acc=98.314, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.00it/s, test_acc=98.398, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 25/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.85it/s, train_acc=98.450, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.99it/s, test_acc=98.546, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 26/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.50it/s, train_acc=98.541, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.16it/s, test_acc=98.527, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 27/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.10it/s, train_acc=98.432, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.61it/s, test_acc=98.546, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 28/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.54it/s, train_acc=98.597, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.63it/s, test_acc=98.606, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 29/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.84it/s, train_acc=98.616, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.26it/s, test_acc=98.625, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 30/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.18it/s, train_acc=98.590, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.39it/s, test_acc=98.596, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 31/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.55it/s, train_acc=98.630, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.09it/s, test_acc=98.714, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 32/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.87it/s, train_acc=98.653, train_loss=1.48]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.75it/s, test_acc=98.744, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 33/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.83it/s, train_acc=98.757, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.91it/s, test_acc=98.744, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 34/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.08it/s, train_acc=98.757, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.67it/s, test_acc=98.655, test_loss=1.48]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 35/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.12it/s, train_acc=98.790, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.63it/s, test_acc=98.714, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 36/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.35it/s, train_acc=98.774, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.71it/s, test_acc=98.754, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 37/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.94it/s, train_acc=98.901, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.53it/s, test_acc=98.853, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 38/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.36it/s, train_acc=98.844, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.04it/s, test_acc=98.813, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 39/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.68it/s, train_acc=98.813, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.45it/s, test_acc=98.813, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 40/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.47it/s, train_acc=98.876, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.12it/s, test_acc=98.724, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 41/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.95it/s, train_acc=98.865, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.09it/s, test_acc=98.645, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 42/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.60it/s, train_acc=98.991, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.90it/s, test_acc=98.784, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 43/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.34it/s, train_acc=98.893, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.49it/s, test_acc=98.754, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 44/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.22it/s, train_acc=98.988, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.40it/s, test_acc=98.902, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 45/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.45it/s, train_acc=99.007, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.09it/s, test_acc=98.863, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 46/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.63it/s, train_acc=99.037, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.46it/s, test_acc=98.912, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 47/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.20it/s, train_acc=99.027, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.89it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 48/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.99it/s, train_acc=99.041, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.85it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 49/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.81it/s, train_acc=99.052, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.34it/s, test_acc=98.813, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 50/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.28it/s, train_acc=99.084, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.50it/s, test_acc=98.863, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 51/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.74it/s, train_acc=99.127, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.36it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 52/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.25it/s, train_acc=99.069, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.81it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 53/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.10it/s, train_acc=99.169, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.11it/s, test_acc=98.813, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 54/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.90it/s, train_acc=99.182, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 30.82it/s, test_acc=98.823, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 55/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.16it/s, train_acc=99.172, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.88it/s, test_acc=98.912, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 56/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.50it/s, train_acc=99.147, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.41it/s, test_acc=98.873, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 57/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.63it/s, train_acc=99.168, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.06it/s, test_acc=98.942, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 58/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.77it/s, train_acc=99.176, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.63it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 59/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.12it/s, train_acc=99.253, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.06it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 60/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.71it/s, train_acc=99.200, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.53it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 61/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.22it/s, train_acc=99.274, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.19it/s, test_acc=98.922, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 62/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.64it/s, train_acc=99.259, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.71it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 63/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.48it/s, train_acc=99.260, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.43it/s, test_acc=98.883, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 64/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.48it/s, train_acc=99.295, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.89it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 65/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=99.307, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.76it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 66/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.94it/s, train_acc=99.269, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.25it/s, test_acc=98.853, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 67/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.50it/s, train_acc=99.291, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.51it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 68/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.66it/s, train_acc=99.348, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.29it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 69/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.21it/s, train_acc=99.293, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.29it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 70/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.94it/s, train_acc=99.338, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.88it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 71/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.73it/s, train_acc=99.330, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.47it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 72/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.43it/s, train_acc=99.323, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.71it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 73/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=99.387, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.31it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 74/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.10it/s, train_acc=99.380, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.79it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 75/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.52it/s, train_acc=99.409, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.86it/s, test_acc=98.873, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 76/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.74it/s, train_acc=99.334, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.49it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 77/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.02it/s, train_acc=99.444, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.89it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 78/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.67it/s, train_acc=99.375, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.01it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 79/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.20it/s, train_acc=99.436, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.58it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 80/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.74it/s, train_acc=99.459, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.43it/s, test_acc=98.873, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 81/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.84it/s, train_acc=99.422, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.25it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 82/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.97it/s, train_acc=99.432, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.42it/s, test_acc=98.853, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 83/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.02it/s, train_acc=99.390, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.86it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 84/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.02it/s, train_acc=99.434, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.02it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 85/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.20it/s, train_acc=99.438, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.90it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 86/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.79it/s, train_acc=99.435, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.93it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 87/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.99it/s, train_acc=99.471, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.87it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 88/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.67it/s, train_acc=99.475, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.43it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 89/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.82it/s, train_acc=99.513, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.22it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 90/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.81it/s, train_acc=99.460, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.65it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 91/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.05it/s, train_acc=99.492, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.18it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 92/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.36it/s, train_acc=99.522, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.48it/s, test_acc=98.922, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 93/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.62it/s, train_acc=99.523, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.68it/s, test_acc=98.962, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 94/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.55it/s, train_acc=99.519, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.69it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 95/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.08it/s, train_acc=99.517, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.53it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 96/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.98it/s, train_acc=99.536, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 37.26it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 97/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=99.562, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.08it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 98/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.47it/s, train_acc=99.542, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.04it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 99/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.86it/s, train_acc=99.554, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.39it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 100/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.80it/s, train_acc=99.524, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.42it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 101/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.09it/s, train_acc=99.560, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.42it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 102/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.60it/s, train_acc=99.547, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.89it/s, test_acc=98.962, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 103/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.20it/s, train_acc=99.537, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.17it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 104/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.80it/s, train_acc=99.565, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.02it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 105/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.27it/s, train_acc=99.590, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.30it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 106/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.82it/s, train_acc=99.585, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 31.47it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 107/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.94it/s, train_acc=99.550, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 32.07it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 108/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.40it/s, train_acc=99.589, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.35it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 109/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.91it/s, train_acc=99.599, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.25it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 110/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.62it/s, train_acc=99.593, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.29it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 111/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.72it/s, train_acc=99.611, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.65it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 112/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.35it/s, train_acc=99.614, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.60it/s, test_acc=98.932, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 113/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.14it/s, train_acc=99.585, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.57it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 114/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.47it/s, train_acc=99.594, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.13it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 115/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.69it/s, train_acc=99.617, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.42it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 116/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.70it/s, train_acc=99.605, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 27.60it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 117/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.11it/s, train_acc=99.642, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.68it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 118/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.70it/s, train_acc=99.600, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.81it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 119/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.83it/s, train_acc=99.627, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.57it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 120/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.06it/s, train_acc=99.630, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.08it/s, test_acc=99.031, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 121/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.80it/s, train_acc=99.637, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.65it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 122/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.78it/s, train_acc=99.656, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.12it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 123/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.97it/s, train_acc=99.610, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.72it/s, test_acc=98.952, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 124/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.99it/s, train_acc=99.632, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.62it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 125/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.15it/s, train_acc=99.658, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.91it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 126/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.53it/s, train_acc=99.647, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.83it/s, test_acc=98.902, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 127/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.43it/s, train_acc=99.675, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.63it/s, test_acc=99.041, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 128/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.37it/s, train_acc=99.651, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.95it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 129/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.46it/s, train_acc=99.642, train_loss=1.47]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.26it/s, test_acc=99.001, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 130/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.14it/s, train_acc=99.701, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.30it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 131/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.51it/s, train_acc=99.665, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.23it/s, test_acc=98.991, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 132/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.98it/s, train_acc=99.695, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.20it/s, test_acc=98.962, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 133/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.28it/s, train_acc=99.689, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.72it/s, test_acc=99.011, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 134/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.07it/s, train_acc=99.685, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 33.84it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 135/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.53it/s, train_acc=99.685, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.28it/s, test_acc=98.972, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 136/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.66it/s, train_acc=99.679, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.72it/s, test_acc=99.150, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 137/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.17it/s, train_acc=99.694, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.97it/s, test_acc=99.120, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 138/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.78it/s, train_acc=99.717, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.40it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 139/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.18it/s, train_acc=99.691, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.77it/s, test_acc=98.981, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 140/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.48it/s, train_acc=99.677, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.59it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 141/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.12it/s, train_acc=99.705, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.33it/s, test_acc=99.051, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 142/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.55it/s, train_acc=99.704, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.81it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 143/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.65it/s, train_acc=99.683, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.84it/s, test_acc=99.100, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 144/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.11it/s, train_acc=99.689, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 29.07it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 145/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.95it/s, train_acc=99.713, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.72it/s, test_acc=99.070, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 146/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 38.05it/s, train_acc=99.701, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.37it/s, test_acc=99.090, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 147/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.70it/s, train_acc=99.742, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.62it/s, test_acc=99.021, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 148/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 36.90it/s, train_acc=99.702, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 28.64it/s, test_acc=99.110, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 149/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:13<00:00, 35.27it/s, train_acc=99.750, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.80it/s, test_acc=99.061, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "EPOCH 150/150 :\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  train: 100%|██████████| 469/469 [00:12<00:00, 37.03it/s, train_acc=99.712, train_loss=1.46]\n",
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.35it/s, test_acc=99.011, test_loss=1.47]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Train time: 2219.791 seconds\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        }
      ],
      "source": [
        "hparams = [\"MNIST\", \"SCNNB-b\", 150]\n",
        "SCNNB_b_MNIST_model, SCNNB_b_MNIST_train_loss, SCNNB_b_MNIST_train_accuracy, SCNNB_b_MNIST_test_loss, SCNNB_b_MNIST_test_accuracy = train(hparams ,MNIST_train_dataloader, MNIST_test_dataloader)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "poyEgfzHY3k1",
        "outputId": "a5bd9737-4ac2-4834-d76b-9fc1a8110831"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Maximum Test Accuracy for SCNNB-b model is:  99.14953164556962\n"
          ]
        }
      ],
      "source": [
        "SCNNB_b_max_acc = max(SCNNB_b_MNIST_test_accuracy)\n",
        "SCNNB_b_MEAN = sum(SCNNB_b_MNIST_test_accuracy)/len(SCNNB_b_MNIST_test_accuracy)\n",
        "print(\"Maximum Test Accuracy for SCNNB-b model is: \", str(SCNNB_b_max_acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "OAKtHLdjjUCh"
      },
      "outputs": [],
      "source": [
        "torch.save(SCNNB_b_MNIST_model.state_dict(), 'SCNNB_b_MNIST')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "Wl9oq6WA0ogH"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import classification_report\n",
        "\n",
        "def Testing(model, model_name, data_loader):\n",
        "\n",
        "    model.load_state_dict(torch.load(model_name))\n",
        "    # Send model to GPU\n",
        "    device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
        "    \n",
        "    model.to(device)\n",
        "\n",
        "    # put model in eval mode\n",
        "    model.eval() \n",
        "\n",
        "    loss_fn = nn.CrossEntropyLoss()\n",
        "    def accuracy_fn(true, pred):\n",
        "        pred = F.softmax(pred, dim = 1)\n",
        "        true = torch.zeros(pred.shape[0], pred.shape[1]).scatter_(1, true.unsqueeze(1), 1.)\n",
        "        acc = (true.argmax(-1) == pred.argmax(-1)).float().detach().numpy()\n",
        "        acc = float((100 * acc.sum()) / len(acc))\n",
        "        return round(acc, 4)\n",
        "\n",
        "    test_loss, test_acc, counter = 0, 0, 0\n",
        "    with tqdm(data_loader, desc =\"   test\") as test_tqdm:\n",
        "        for X, y in test_tqdm:\n",
        "            # Send data to GPU\n",
        "            X, y = X.to(device), y.to(device)\n",
        "            \n",
        "            # 1. Forward pass\n",
        "            test_pred = model(X)\n",
        "\n",
        "            _, preds = torch.max(test_pred.data, 1)\n",
        "            \n",
        "            # 2. Calculate loss and accuracy\n",
        "            test_loss += loss_fn(test_pred, y)\n",
        "            test_acc += accuracy_fn(true=y.cpu(),\n",
        "                                    pred=test_pred.cpu())\n",
        "            counter += 1\n",
        "\n",
        "            if counter==1 :\n",
        "                y_true = (y.detach().cpu().numpy())\n",
        "                y_pred = (preds.detach().cpu().numpy())\n",
        "            else:\n",
        "                y_true = np.concatenate((y_true, np.array(y.detach().cpu().numpy())), axis=0)\n",
        "                y_pred = np.concatenate((y_pred, np.array(preds.detach().cpu().numpy())), axis=0)\n",
        "\n",
        "\n",
        "            test_tqdm.set_postfix(test_acc=\"{:.3f}\".format(test_acc/counter), test_loss=test_loss.item()/counter, refresh=True)\n",
        "\n",
        "\n",
        "        # Adjust metrics and print out\n",
        "        test_loss /= len(data_loader)\n",
        "        test_acc  /= len(data_loader)\n",
        "\n",
        "    class_names = [\"Zero\",\"One\", \"Two\", \"Three\", \"Four\", \"Five\", \"Six\", \"Seven\", \"Eight\", \"Nine\"]\n",
        "    cnf_matrix = confusion_matrix(y_true, y_pred)   ## Sklearn confusion matrix object\n",
        "    accuracy = np.trace(cnf_matrix) / float(np.sum(cnf_matrix))\n",
        "    misclass = 1 - accuracy\n",
        "\n",
        "    fig, ax = plt.subplots()\n",
        "    tick_marks = np.arange(len(class_names))\n",
        "    plt.xticks(tick_marks, class_names, rotation=45)\n",
        "    plt.yticks(tick_marks, class_names)\n",
        "\n",
        "    sns.heatmap(pd.DataFrame(cnf_matrix), annot=True, cmap = \"Blues\" ,fmt='g', xticklabels=class_names , yticklabels=class_names)\n",
        "    ax.xaxis.set_label_position(\"bottom\")\n",
        "    plt.title('Confusion matrix for '+ model_name, y=1.1)\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label\\naccuracy={:0.2f}; misclass={:0.2f}'.format(accuracy, misclass))\n",
        "    plt.tight_layout()\n",
        "    print(classification_report(y_true, y_pred, target_names=class_names, digits=4))\n",
        "    return test_loss.item(), test_acc\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 837
        },
        "id": "WJr49-cy0tXf",
        "outputId": "58b7d2ae-eca7-492b-c994-dac0961a06b8"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "   test: 100%|██████████| 79/79 [00:02<00:00, 34.79it/s, test_acc=99.199, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        Zero     0.9939    0.9959    0.9949       980\n",
            "         One     0.9956    0.9956    0.9956      1135\n",
            "         Two     0.9865    0.9893    0.9879      1032\n",
            "       Three     0.9980    0.9941    0.9960      1010\n",
            "        Four     0.9949    0.9969    0.9959       982\n",
            "        Five     0.9779    0.9899    0.9838       892\n",
            "         Six     0.9906    0.9864    0.9885       958\n",
            "       Seven     0.9913    0.9922    0.9917      1028\n",
            "       Eight     0.9939    0.9969    0.9954       974\n",
            "        Nine     0.9950    0.9812    0.9880      1009\n",
            "\n",
            "    accuracy                         0.9919     10000\n",
            "   macro avg     0.9917    0.9919    0.9918     10000\n",
            "weighted avg     0.9919    0.9919    0.9919     10000\n",
            "\n",
            "\n",
            "Test Accuracy for SCNNB model on MNIST is: 99.19898734177214\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "SCNNB_test_loss, SCNNB_test_acc = Testing(SCNNB_MNIST_model, \"SCNNB_MNIST\", MNIST_test_dataloader )\n",
        "print(\"\\nTest Accuracy for SCNNB model on MNIST is: \"+str(SCNNB_test_acc))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 837
        },
        "id": "DCP9s8ga0uaP",
        "outputId": "80a5bc4d-cedb-47f8-9896-5811719327b2"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "   test: 100%|██████████| 79/79 [00:02<00:00, 35.74it/s, test_acc=99.189, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        Zero     0.9919    0.9959    0.9939       980\n",
            "         One     0.9947    0.9982    0.9965      1135\n",
            "         Two     0.9874    0.9874    0.9874      1032\n",
            "       Three     0.9980    0.9911    0.9945      1010\n",
            "        Four     0.9929    0.9959    0.9944       982\n",
            "        Five     0.9810    0.9865    0.9838       892\n",
            "         Six     0.9947    0.9823    0.9884       958\n",
            "       Seven     0.9884    0.9951    0.9918      1028\n",
            "       Eight     0.9949    0.9959    0.9954       974\n",
            "        Nine     0.9930    0.9881    0.9906      1009\n",
            "\n",
            "    accuracy                         0.9918     10000\n",
            "   macro avg     0.9917    0.9917    0.9917     10000\n",
            "weighted avg     0.9918    0.9918    0.9918     10000\n",
            "\n",
            "\n",
            "Test Accuracy for SCNNB-a model on MNIST is: 99.18908860759491\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "SCNNB_a_test_loss, SCNNB_a_test_acc = Testing(SCNNB_a_MNIST_model, \"SCNNB_a_MNIST\", MNIST_test_dataloader )\n",
        "print(\"\\nTest Accuracy for SCNNB-a model on MNIST is: \"+str(SCNNB_a_test_acc))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 837
        },
        "id": "cVQGbZEn0vnX",
        "outputId": "f95c6ec6-02a2-4137-8bd4-432be851a8b8"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "   test: 100%|██████████| 79/79 [00:02<00:00, 36.25it/s, test_acc=99.080, test_loss=1.47]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        Zero     0.9909    0.9949    0.9929       980\n",
            "         One     0.9956    0.9956    0.9956      1135\n",
            "         Two     0.9874    0.9893    0.9884      1032\n",
            "       Three     0.9901    0.9921    0.9911      1010\n",
            "        Four     0.9898    0.9929    0.9914       982\n",
            "        Five     0.9886    0.9765    0.9825       892\n",
            "         Six     0.9906    0.9875    0.9890       958\n",
            "       Seven     0.9903    0.9932    0.9917      1028\n",
            "       Eight     0.9908    0.9949    0.9928       974\n",
            "        Nine     0.9920    0.9881    0.9901      1009\n",
            "\n",
            "    accuracy                         0.9907     10000\n",
            "   macro avg     0.9906    0.9905    0.9905     10000\n",
            "weighted avg     0.9907    0.9907    0.9907     10000\n",
            "\n",
            "\n",
            "Test Accuracy for SCNNB-b model on MNIST is: 99.08031012658228\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "SCNNB_b_test_loss, SCNNB_b_test_acc = Testing(SCNNB_b_MNIST_model, \"SCNNB_b_MNIST\", MNIST_test_dataloader )\n",
        "print(\"\\nTest Accuracy for SCNNB-b model on MNIST is: \"+str(SCNNB_b_test_acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "id": "1GYqdKiDjXVj"
      },
      "outputs": [],
      "source": [
        "def acc_test_plots(SCNNB_test, SCNNB_a_test, SCNNB_b_test):\n",
        "  \n",
        "    num_epochs = len(SCNNB_test)\n",
        "    plt.figure(figsize=(13,5))\n",
        "    plt.plot(SCNNB_a_test , 'b', label='SCNNB-a', linestyle = '-')\n",
        "    plt.plot(SCNNB_b_test , 'g', label='SCNNB-b', linestyle = '-.')\n",
        "    plt.plot(SCNNB_test, 'r', label='SCNNB', linestyle = ':')\n",
        "    plt.ylabel('Test Accuracy', fontsize=10, labelpad=8)\n",
        "    plt.xlabel('Epochs', fontsize=10, labelpad=8)\n",
        "    plt.title('Test Accuracy for MNIST dataset', fontsize=25, pad=15)\n",
        "    plt.legend()\n",
        "    sns.despine()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 518
        },
        "id": "yA35a8uajYEH",
        "outputId": "111e867b-3f8e-49ac-9cb6-1d1f810b3d08"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1300x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "acc_test_plots(SCNNB_MNIST_test_accuracy, SCNNB_a_MNIST_test_accuracy, SCNNB_b_MNIST_test_accuracy[1:])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {
        "id": "bLbRmVazbCwA"
      },
      "outputs": [],
      "source": [
        "def acc_train_plots(SCNNB_train, SCNNB_a_train, SCNNB_b_train):\n",
        "  \n",
        "    num_epochs = len(SCNNB_train)\n",
        "    plt.figure(figsize=(13,5))\n",
        "    plt.plot(SCNNB_a_train , 'b', label='SCNNB-a', linestyle = '-')\n",
        "    plt.plot(SCNNB_b_train , 'g', label='SCNNB-b', linestyle = '-.')\n",
        "    plt.plot(SCNNB_train, 'r', label='SCNNB', linestyle = ':')\n",
        "    plt.ylabel('Train Accuracy', fontsize=10, labelpad=8)\n",
        "    plt.xlabel('Epochs', fontsize=10, labelpad=8)\n",
        "    plt.title('Train Accuracy for MNIST dataset', fontsize=25, pad=15)\n",
        "    plt.legend()\n",
        "    sns.despine()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 518
        },
        "id": "khNfXH-tbCwE",
        "outputId": "91e52a59-7d54-49ed-9347-e045bd4cb2cc"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1300x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "acc_train_plots(SCNNB_MNIST_train_accuracy, SCNNB_a_MNIST_train_accuracy, SCNNB_b_MNIST_train_accuracy[2:])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "id": "vq6uE-7TbZSW"
      },
      "outputs": [],
      "source": [
        "def loss_train_plots(SCNNB_train, SCNNB_a_train, SCNNB_b_train):\n",
        "  \n",
        "    num_epochs = len(SCNNB_train)\n",
        "    plt.figure(figsize=(13,5))\n",
        "    plt.plot(SCNNB_a_train , 'b', label='SCNNB-a', linestyle = '-')\n",
        "    plt.plot(SCNNB_b_train , 'g', label='SCNNB-b', linestyle = '-.')\n",
        "    plt.plot(SCNNB_train, 'r', label='SCNNB', linestyle = ':')\n",
        "    plt.ylabel('Train Accuracy', fontsize=10, labelpad=8)\n",
        "    plt.xlabel('Epochs', fontsize=10, labelpad=8)\n",
        "    plt.title('Train Loss for MNIST dataset', fontsize=25, pad=15)\n",
        "    plt.legend()\n",
        "    sns.despine()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 49,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 518
        },
        "id": "1B8IyqXWbZSX",
        "outputId": "120d560e-db8e-41bb-a368-260f37f6d32c"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1300x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "loss_train_plots(SCNNB_MNIST_train_loss, SCNNB_a_MNIST_train_loss, SCNNB_b_MNIST_train_loss[1:])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WZ728vl6pOJK"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}