FC_Implementation.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "from torch.utils.data import DataLoader\n",
        "import matplotlib.pyplot as plt\n",
        "from tabulate import tabulate\n",
        "\n",
        "# Install required libraries\n",
        "!pip install -U portalocker>=2.0.0\n",
        "!pip install tabulate\n",
        "\n",
        "# Define data transformation\n",
        "transform = transforms.Compose([\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize((0.5,), (0.5,))\n",
        "])\n",
        "\n",
        "# Load FashionMNIST dataset\n",
        "train_dataset = torchvision.datasets.FashionMNIST(root='./data', train=True, download=True, transform=transform)\n",
        "test_dataset = torchvision.datasets.FashionMNIST(root='./data', train=False, download=True, transform=transform)\n",
        "\n",
        "# Create data loaders\n",
        "train_loader = torch.utils.data.DataLoader(train_dataset, batch_size=64, shuffle=True)\n",
        "test_loader = torch.utils.data.DataLoader(test_dataset, batch_size=64, shuffle=False)\n",
        "\n",
        "# Labels mapping\n",
        "labels_map = {\n",
        "    0: \"T-shirt/top\", 1: \"Trouser\", 2: \"Pullover\", 3: \"Dress\", 4: \"Coat\",\n",
        "    5: \"Sandal\", 6: \"Shirt\", 7: \"Sneaker\", 8: \"Bag\", 9: \"Ankle boot\"\n",
        "}\n",
        "\n",
        "# Display sample images\n",
        "def display_sample_images():\n",
        "    figure = plt.figure(figsize=(8, 8))\n",
        "    cols, rows = 3, 3\n",
        "\n",
        "    for i in range(1, cols * rows + 1):\n",
        "        sample_idx = torch.randint(len(train_dataset), size=(1,)).item()\n",
        "        img, label = train_dataset[sample_idx]\n",
        "        figure.add_subplot(rows, cols, i)\n",
        "        plt.title(labels_map[label])\n",
        "        plt.axis(\"off\")\n",
        "        plt.imshow(img.squeeze(), cmap=\"gray\")\n",
        "    plt.show()\n",
        "\n",
        "# Model definition\n",
        "class FashionMLP(nn.Module):\n",
        "    def __init__(self, num_hidden_layers, hidden_layer_sizes):\n",
        "        super(FashionMLP, self).__init__()\n",
        "        layers = []\n",
        "        layers.append(nn.Linear(28 * 28, hidden_layer_sizes[0]))  # Input layer\n",
        "        for i in range(num_hidden_layers - 1):  # Hidden layers\n",
        "            layers.append(nn.Linear(hidden_layer_sizes[i], hidden_layer_sizes[i + 1]))\n",
        "        layers.append(nn.Linear(hidden_layer_sizes[-1], 10))  # Output layer\n",
        "        self.layers = nn.ModuleList(layers)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = x.view(-1, 28 * 28)\n",
        "        for layer in self.layers[:-1]:\n",
        "            x = torch.sigmoid(layer(x))\n",
        "        x = self.layers[-1](x)\n",
        "        return x\n",
        "\n",
        "# Define configurations\n",
        "configurations = [\n",
        "    {\n",
        "        'name': \"Configuration-1\",\n",
        "        'num_hidden_layers': 3,\n",
        "        'hidden_layer_sizes': [32, 64, 128],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-2\",\n",
        "        'num_hidden_layers': 3,\n",
        "        'hidden_layer_sizes': [64, 128, 256],\n",
        "        'batch_size': 512,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.RMSprop\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-3\",\n",
        "        'num_hidden_layers': 3,\n",
        "        'hidden_layer_sizes': [128, 256, 512],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-4\",\n",
        "        'num_hidden_layers': 2,\n",
        "        'hidden_layer_sizes': [64, 128],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-5\",\n",
        "        'num_hidden_layers': 2,\n",
        "        'hidden_layer_sizes': [32, 64],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-6\",\n",
        "        'num_hidden_layers': 2,\n",
        "        'hidden_layer_sizes': [128, 256],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-7\",\n",
        "        'num_hidden_layers': 2,\n",
        "        'hidden_layer_sizes': [256, 512],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.RMSprop\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-8\",\n",
        "        'num_hidden_layers': 1,\n",
        "        'hidden_layer_sizes': [64],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-9\",\n",
        "        'num_hidden_layers': 1,\n",
        "        'hidden_layer_sizes': [128],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-10\",\n",
        "        'num_hidden_layers': 1,\n",
        "        'hidden_layer_sizes': [256],\n",
        "        'batch_size': 64,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-11\",\n",
        "        'num_hidden_layers': 3,\n",
        "        'hidden_layer_sizes': [32, 64, 128],\n",
        "        'batch_size': 128,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.RMSprop\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-12\",\n",
        "        'num_hidden_layers': 4,\n",
        "        'hidden_layer_sizes': [64, 128, 256, 512],\n",
        "        'batch_size': 128,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-13\",\n",
        "        'num_hidden_layers': 2,\n",
        "        'hidden_layer_sizes': [128, 256],\n",
        "        'batch_size': 256,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.RMSprop\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-14\",\n",
        "        'num_hidden_layers': 1,\n",
        "        'hidden_layer_sizes': [512],\n",
        "        'batch_size': 256,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.Adam\n",
        "    },\n",
        "    {\n",
        "        'name': \"Configuration-15\",\n",
        "        'num_hidden_layers': 4,\n",
        "        'hidden_layer_sizes': [64, 128, 256, 512],\n",
        "        'batch_size': 32,\n",
        "        'learning_rate': 0.001,\n",
        "        'optimizer': optim.RMSprop\n",
        "    }\n",
        "]\n",
        "\n",
        "\n",
        "# Running experiments\n",
        "FC_results = []\n",
        "\n",
        "for config in configurations:\n",
        "    name = config['name']\n",
        "    num_hidden_layers = config['num_hidden_layers']\n",
        "    hidden_layer_sizes = config['hidden_layer_sizes']\n",
        "    batch_size = config['batch_size']\n",
        "    learning_rate = config['learning_rate']\n",
        "    optimizer_func = config['optimizer']\n",
        "\n",
        "    # Initialize model, loss function, and optimizer\n",
        "    model = FashionMLP(num_hidden_layers, hidden_layer_sizes)\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    optimizer = optimizer_func(model.parameters(), lr=learning_rate)\n",
        "\n",
        "    # Display configuration\n",
        "    print(f\"Configuration - {name}\")\n",
        "    print(\"Input data:\")\n",
        "    print(f\"- Num Hidden Layers: {num_hidden_layers}\")\n",
        "    print(f\"- Hidden Layer Sizes: {hidden_layer_sizes}\")\n",
        "    print(f\"- Batch Size: {batch_size}\")\n",
        "    print(f\"- Learning Rate: {learning_rate}\")\n",
        "\n",
        "    # Training\n",
        "    num_epochs = 10\n",
        "    epoch_losses = []\n",
        "    for epoch in range(num_epochs):\n",
        "        model.train()\n",
        "        running_loss = 0.0\n",
        "        for i, data in enumerate(train_loader, 0):\n",
        "            inputs, labels = data\n",
        "            optimizer.zero_grad()\n",
        "            outputs = model(inputs)\n",
        "            loss = criterion(outputs, labels)\n",
        "            loss.backward()\n",
        "            optimizer.step()\n",
        "            running_loss += loss.item()\n",
        "\n",
        "            # Print epoch-wise batch loss\n",
        "            if i % 100 == 99:\n",
        "                print(f'Epoch [{epoch + 1}], Batch [{i + 1}/{len(train_loader)}], Loss: {running_loss / 100:.4f}')\n",
        "                running_loss = 0.0\n",
        "\n",
        "        epoch_losses.append(running_loss)\n",
        "\n",
        "    # Evaluating\n",
        "    model.eval()\n",
        "    correct = 0\n",
        "    total = 0\n",
        "    with torch.no_grad():\n",
        "        for data in test_loader:\n",
        "            images, labels = data\n",
        "            outputs = model(images)\n",
        "            _, predicted = torch.max(outputs.data, 1)\n",
        "            total += labels.size(0)\n",
        "            correct += (predicted == labels).sum().item()\n",
        "\n",
        "    accuracy = correct / total\n",
        "    print(f\"Accuracy for {name}: {accuracy * 100:.2f}%\\n\")\n",
        "\n",
        "    FC_results.append([name, num_hidden_layers, hidden_layer_sizes, batch_size, learning_rate, accuracy])\n",
        "\n",
        "    # Print dotted line break\n",
        "    print(\".\" * 50)\n",
        "    print()\n",
        "\n",
        "# Display FC_results\n",
        "print(tabulate(FC_results, headers=[\"Configuration\", \"Num Hidden Layers\", \"Hidden Layer Sizes\", \"Batch Size\", \"Learning Rate\", \"Accuracy\"],\n",
        "               floatfmt=\".2f\"))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GC7T-4h3mIhW",
        "outputId": "be9a5fee-2c3c-404d-e715-4a6ba7f06d12"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: tabulate in /usr/local/lib/python3.10/dist-packages (0.9.0)\n",
            "Configuration - Configuration-1\n",
            "Input data:\n",
            "- Num Hidden Layers: 3\n",
            "- Hidden Layer Sizes: [32, 64, 128]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 2.2035\n",
            "Epoch [1], Batch [200/938], Loss: 1.5984\n",
            "Epoch [1], Batch [300/938], Loss: 1.2569\n",
            "Epoch [1], Batch [400/938], Loss: 1.0815\n",
            "Epoch [1], Batch [500/938], Loss: 0.9633\n",
            "Epoch [1], Batch [600/938], Loss: 0.8694\n",
            "Epoch [1], Batch [700/938], Loss: 0.8051\n",
            "Epoch [1], Batch [800/938], Loss: 0.7442\n",
            "Epoch [1], Batch [900/938], Loss: 0.7179\n",
            "Epoch [2], Batch [100/938], Loss: 0.6849\n",
            "Epoch [2], Batch [200/938], Loss: 0.6778\n",
            "Epoch [2], Batch [300/938], Loss: 0.6354\n",
            "Epoch [2], Batch [400/938], Loss: 0.6444\n",
            "Epoch [2], Batch [500/938], Loss: 0.6512\n",
            "Epoch [2], Batch [600/938], Loss: 0.6123\n",
            "Epoch [2], Batch [700/938], Loss: 0.6203\n",
            "Epoch [2], Batch [800/938], Loss: 0.6024\n",
            "Epoch [2], Batch [900/938], Loss: 0.5857\n",
            "Epoch [3], Batch [100/938], Loss: 0.5701\n",
            "Epoch [3], Batch [200/938], Loss: 0.5539\n",
            "Epoch [3], Batch [300/938], Loss: 0.5303\n",
            "Epoch [3], Batch [400/938], Loss: 0.5269\n",
            "Epoch [3], Batch [500/938], Loss: 0.5113\n",
            "Epoch [3], Batch [600/938], Loss: 0.5126\n",
            "Epoch [3], Batch [700/938], Loss: 0.5165\n",
            "Epoch [3], Batch [800/938], Loss: 0.4772\n",
            "Epoch [3], Batch [900/938], Loss: 0.4935\n",
            "Epoch [4], Batch [100/938], Loss: 0.4729\n",
            "Epoch [4], Batch [200/938], Loss: 0.4488\n",
            "Epoch [4], Batch [300/938], Loss: 0.4497\n",
            "Epoch [4], Batch [400/938], Loss: 0.4597\n",
            "Epoch [4], Batch [500/938], Loss: 0.4245\n",
            "Epoch [4], Batch [600/938], Loss: 0.4295\n",
            "Epoch [4], Batch [700/938], Loss: 0.4157\n",
            "Epoch [4], Batch [800/938], Loss: 0.4298\n",
            "Epoch [4], Batch [900/938], Loss: 0.4124\n",
            "Epoch [5], Batch [100/938], Loss: 0.4170\n",
            "Epoch [5], Batch [200/938], Loss: 0.4056\n",
            "Epoch [5], Batch [300/938], Loss: 0.4098\n",
            "Epoch [5], Batch [400/938], Loss: 0.3910\n",
            "Epoch [5], Batch [500/938], Loss: 0.3825\n",
            "Epoch [5], Batch [600/938], Loss: 0.4065\n",
            "Epoch [5], Batch [700/938], Loss: 0.3851\n",
            "Epoch [5], Batch [800/938], Loss: 0.3863\n",
            "Epoch [5], Batch [900/938], Loss: 0.3879\n",
            "Epoch [6], Batch [100/938], Loss: 0.3495\n",
            "Epoch [6], Batch [200/938], Loss: 0.3924\n",
            "Epoch [6], Batch [300/938], Loss: 0.3572\n",
            "Epoch [6], Batch [400/938], Loss: 0.3802\n",
            "Epoch [6], Batch [500/938], Loss: 0.3794\n",
            "Epoch [6], Batch [600/938], Loss: 0.3666\n",
            "Epoch [6], Batch [700/938], Loss: 0.3748\n",
            "Epoch [6], Batch [800/938], Loss: 0.3929\n",
            "Epoch [6], Batch [900/938], Loss: 0.3804\n",
            "Epoch [7], Batch [100/938], Loss: 0.3545\n",
            "Epoch [7], Batch [200/938], Loss: 0.3741\n",
            "Epoch [7], Batch [300/938], Loss: 0.3524\n",
            "Epoch [7], Batch [400/938], Loss: 0.3643\n",
            "Epoch [7], Batch [500/938], Loss: 0.3499\n",
            "Epoch [7], Batch [600/938], Loss: 0.3810\n",
            "Epoch [7], Batch [700/938], Loss: 0.3605\n",
            "Epoch [7], Batch [800/938], Loss: 0.3421\n",
            "Epoch [7], Batch [900/938], Loss: 0.3350\n",
            "Epoch [8], Batch [100/938], Loss: 0.3321\n",
            "Epoch [8], Batch [200/938], Loss: 0.3417\n",
            "Epoch [8], Batch [300/938], Loss: 0.3629\n",
            "Epoch [8], Batch [400/938], Loss: 0.3381\n",
            "Epoch [8], Batch [500/938], Loss: 0.3551\n",
            "Epoch [8], Batch [600/938], Loss: 0.3392\n",
            "Epoch [8], Batch [700/938], Loss: 0.3514\n",
            "Epoch [8], Batch [800/938], Loss: 0.3662\n",
            "Epoch [8], Batch [900/938], Loss: 0.3383\n",
            "Epoch [9], Batch [100/938], Loss: 0.3396\n",
            "Epoch [9], Batch [200/938], Loss: 0.3211\n",
            "Epoch [9], Batch [300/938], Loss: 0.3249\n",
            "Epoch [9], Batch [400/938], Loss: 0.3353\n",
            "Epoch [9], Batch [500/938], Loss: 0.3395\n",
            "Epoch [9], Batch [600/938], Loss: 0.3318\n",
            "Epoch [9], Batch [700/938], Loss: 0.3540\n",
            "Epoch [9], Batch [800/938], Loss: 0.3334\n",
            "Epoch [9], Batch [900/938], Loss: 0.3420\n",
            "Epoch [10], Batch [100/938], Loss: 0.3484\n",
            "Epoch [10], Batch [200/938], Loss: 0.3205\n",
            "Epoch [10], Batch [300/938], Loss: 0.3215\n",
            "Epoch [10], Batch [400/938], Loss: 0.3257\n",
            "Epoch [10], Batch [500/938], Loss: 0.3240\n",
            "Epoch [10], Batch [600/938], Loss: 0.3342\n",
            "Epoch [10], Batch [700/938], Loss: 0.3282\n",
            "Epoch [10], Batch [800/938], Loss: 0.3188\n",
            "Epoch [10], Batch [900/938], Loss: 0.3310\n",
            "Accuracy for Configuration-1: 85.86%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-2\n",
            "Input data:\n",
            "- Num Hidden Layers: 3\n",
            "- Hidden Layer Sizes: [64, 128, 256]\n",
            "- Batch Size: 512\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.5954\n",
            "Epoch [1], Batch [200/938], Loss: 1.0414\n",
            "Epoch [1], Batch [300/938], Loss: 0.8373\n",
            "Epoch [1], Batch [400/938], Loss: 0.7267\n",
            "Epoch [1], Batch [500/938], Loss: 0.6814\n",
            "Epoch [1], Batch [600/938], Loss: 0.6520\n",
            "Epoch [1], Batch [700/938], Loss: 0.6315\n",
            "Epoch [1], Batch [800/938], Loss: 0.5900\n",
            "Epoch [1], Batch [900/938], Loss: 0.5931\n",
            "Epoch [2], Batch [100/938], Loss: 0.5529\n",
            "Epoch [2], Batch [200/938], Loss: 0.5116\n",
            "Epoch [2], Batch [300/938], Loss: 0.4961\n",
            "Epoch [2], Batch [400/938], Loss: 0.4981\n",
            "Epoch [2], Batch [500/938], Loss: 0.4849\n",
            "Epoch [2], Batch [600/938], Loss: 0.4878\n",
            "Epoch [2], Batch [700/938], Loss: 0.4781\n",
            "Epoch [2], Batch [800/938], Loss: 0.4725\n",
            "Epoch [2], Batch [900/938], Loss: 0.4555\n",
            "Epoch [3], Batch [100/938], Loss: 0.4513\n",
            "Epoch [3], Batch [200/938], Loss: 0.4313\n",
            "Epoch [3], Batch [300/938], Loss: 0.4233\n",
            "Epoch [3], Batch [400/938], Loss: 0.4118\n",
            "Epoch [3], Batch [500/938], Loss: 0.4208\n",
            "Epoch [3], Batch [600/938], Loss: 0.4145\n",
            "Epoch [3], Batch [700/938], Loss: 0.4057\n",
            "Epoch [3], Batch [800/938], Loss: 0.4214\n",
            "Epoch [3], Batch [900/938], Loss: 0.4275\n",
            "Epoch [4], Batch [100/938], Loss: 0.3873\n",
            "Epoch [4], Batch [200/938], Loss: 0.4102\n",
            "Epoch [4], Batch [300/938], Loss: 0.3914\n",
            "Epoch [4], Batch [400/938], Loss: 0.3799\n",
            "Epoch [4], Batch [500/938], Loss: 0.3850\n",
            "Epoch [4], Batch [600/938], Loss: 0.3899\n",
            "Epoch [4], Batch [700/938], Loss: 0.3764\n",
            "Epoch [4], Batch [800/938], Loss: 0.3647\n",
            "Epoch [4], Batch [900/938], Loss: 0.3851\n",
            "Epoch [5], Batch [100/938], Loss: 0.3693\n",
            "Epoch [5], Batch [200/938], Loss: 0.3512\n",
            "Epoch [5], Batch [300/938], Loss: 0.3842\n",
            "Epoch [5], Batch [400/938], Loss: 0.3665\n",
            "Epoch [5], Batch [500/938], Loss: 0.3730\n",
            "Epoch [5], Batch [600/938], Loss: 0.3512\n",
            "Epoch [5], Batch [700/938], Loss: 0.3623\n",
            "Epoch [5], Batch [800/938], Loss: 0.3427\n",
            "Epoch [5], Batch [900/938], Loss: 0.3439\n",
            "Epoch [6], Batch [100/938], Loss: 0.3516\n",
            "Epoch [6], Batch [200/938], Loss: 0.3391\n",
            "Epoch [6], Batch [300/938], Loss: 0.3400\n",
            "Epoch [6], Batch [400/938], Loss: 0.3600\n",
            "Epoch [6], Batch [500/938], Loss: 0.3508\n",
            "Epoch [6], Batch [600/938], Loss: 0.3426\n",
            "Epoch [6], Batch [700/938], Loss: 0.3462\n",
            "Epoch [6], Batch [800/938], Loss: 0.3377\n",
            "Epoch [6], Batch [900/938], Loss: 0.3272\n",
            "Epoch [7], Batch [100/938], Loss: 0.3437\n",
            "Epoch [7], Batch [200/938], Loss: 0.3392\n",
            "Epoch [7], Batch [300/938], Loss: 0.3249\n",
            "Epoch [7], Batch [400/938], Loss: 0.3261\n",
            "Epoch [7], Batch [500/938], Loss: 0.3337\n",
            "Epoch [7], Batch [600/938], Loss: 0.3220\n",
            "Epoch [7], Batch [700/938], Loss: 0.3212\n",
            "Epoch [7], Batch [800/938], Loss: 0.3234\n",
            "Epoch [7], Batch [900/938], Loss: 0.3149\n",
            "Epoch [8], Batch [100/938], Loss: 0.3122\n",
            "Epoch [8], Batch [200/938], Loss: 0.3175\n",
            "Epoch [8], Batch [300/938], Loss: 0.3048\n",
            "Epoch [8], Batch [400/938], Loss: 0.3278\n",
            "Epoch [8], Batch [500/938], Loss: 0.3341\n",
            "Epoch [8], Batch [600/938], Loss: 0.3175\n",
            "Epoch [8], Batch [700/938], Loss: 0.3128\n",
            "Epoch [8], Batch [800/938], Loss: 0.3159\n",
            "Epoch [8], Batch [900/938], Loss: 0.3138\n",
            "Epoch [9], Batch [100/938], Loss: 0.3075\n",
            "Epoch [9], Batch [200/938], Loss: 0.2977\n",
            "Epoch [9], Batch [300/938], Loss: 0.2970\n",
            "Epoch [9], Batch [400/938], Loss: 0.2852\n",
            "Epoch [9], Batch [500/938], Loss: 0.3001\n",
            "Epoch [9], Batch [600/938], Loss: 0.3100\n",
            "Epoch [9], Batch [700/938], Loss: 0.3243\n",
            "Epoch [9], Batch [800/938], Loss: 0.3164\n",
            "Epoch [9], Batch [900/938], Loss: 0.3002\n",
            "Epoch [10], Batch [100/938], Loss: 0.2873\n",
            "Epoch [10], Batch [200/938], Loss: 0.2979\n",
            "Epoch [10], Batch [300/938], Loss: 0.2970\n",
            "Epoch [10], Batch [400/938], Loss: 0.3091\n",
            "Epoch [10], Batch [500/938], Loss: 0.2898\n",
            "Epoch [10], Batch [600/938], Loss: 0.3063\n",
            "Epoch [10], Batch [700/938], Loss: 0.2923\n",
            "Epoch [10], Batch [800/938], Loss: 0.2776\n",
            "Epoch [10], Batch [900/938], Loss: 0.3016\n",
            "Accuracy for Configuration-2: 83.65%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-3\n",
            "Input data:\n",
            "- Num Hidden Layers: 3\n",
            "- Hidden Layer Sizes: [128, 256, 512]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.7136\n",
            "Epoch [1], Batch [200/938], Loss: 0.9039\n",
            "Epoch [1], Batch [300/938], Loss: 0.7334\n",
            "Epoch [1], Batch [400/938], Loss: 0.6828\n",
            "Epoch [1], Batch [500/938], Loss: 0.6215\n",
            "Epoch [1], Batch [600/938], Loss: 0.5889\n",
            "Epoch [1], Batch [700/938], Loss: 0.5456\n",
            "Epoch [1], Batch [800/938], Loss: 0.5371\n",
            "Epoch [1], Batch [900/938], Loss: 0.5309\n",
            "Epoch [2], Batch [100/938], Loss: 0.4609\n",
            "Epoch [2], Batch [200/938], Loss: 0.4503\n",
            "Epoch [2], Batch [300/938], Loss: 0.4705\n",
            "Epoch [2], Batch [400/938], Loss: 0.4358\n",
            "Epoch [2], Batch [500/938], Loss: 0.4314\n",
            "Epoch [2], Batch [600/938], Loss: 0.4249\n",
            "Epoch [2], Batch [700/938], Loss: 0.4379\n",
            "Epoch [2], Batch [800/938], Loss: 0.4379\n",
            "Epoch [2], Batch [900/938], Loss: 0.4225\n",
            "Epoch [3], Batch [100/938], Loss: 0.3734\n",
            "Epoch [3], Batch [200/938], Loss: 0.3836\n",
            "Epoch [3], Batch [300/938], Loss: 0.3868\n",
            "Epoch [3], Batch [400/938], Loss: 0.3868\n",
            "Epoch [3], Batch [500/938], Loss: 0.3824\n",
            "Epoch [3], Batch [600/938], Loss: 0.4166\n",
            "Epoch [3], Batch [700/938], Loss: 0.3751\n",
            "Epoch [3], Batch [800/938], Loss: 0.3810\n",
            "Epoch [3], Batch [900/938], Loss: 0.3619\n",
            "Epoch [4], Batch [100/938], Loss: 0.3493\n",
            "Epoch [4], Batch [200/938], Loss: 0.3655\n",
            "Epoch [4], Batch [300/938], Loss: 0.3493\n",
            "Epoch [4], Batch [400/938], Loss: 0.3589\n",
            "Epoch [4], Batch [500/938], Loss: 0.3476\n",
            "Epoch [4], Batch [600/938], Loss: 0.3389\n",
            "Epoch [4], Batch [700/938], Loss: 0.3526\n",
            "Epoch [4], Batch [800/938], Loss: 0.3345\n",
            "Epoch [4], Batch [900/938], Loss: 0.3455\n",
            "Epoch [5], Batch [100/938], Loss: 0.3153\n",
            "Epoch [5], Batch [200/938], Loss: 0.3271\n",
            "Epoch [5], Batch [300/938], Loss: 0.3345\n",
            "Epoch [5], Batch [400/938], Loss: 0.3424\n",
            "Epoch [5], Batch [500/938], Loss: 0.3198\n",
            "Epoch [5], Batch [600/938], Loss: 0.3271\n",
            "Epoch [5], Batch [700/938], Loss: 0.3254\n",
            "Epoch [5], Batch [800/938], Loss: 0.3251\n",
            "Epoch [5], Batch [900/938], Loss: 0.3075\n",
            "Epoch [6], Batch [100/938], Loss: 0.3082\n",
            "Epoch [6], Batch [200/938], Loss: 0.3008\n",
            "Epoch [6], Batch [300/938], Loss: 0.3206\n",
            "Epoch [6], Batch [400/938], Loss: 0.3239\n",
            "Epoch [6], Batch [500/938], Loss: 0.3188\n",
            "Epoch [6], Batch [600/938], Loss: 0.3078\n",
            "Epoch [6], Batch [700/938], Loss: 0.2988\n",
            "Epoch [6], Batch [800/938], Loss: 0.2932\n",
            "Epoch [6], Batch [900/938], Loss: 0.3194\n",
            "Epoch [7], Batch [100/938], Loss: 0.2982\n",
            "Epoch [7], Batch [200/938], Loss: 0.2807\n",
            "Epoch [7], Batch [300/938], Loss: 0.3039\n",
            "Epoch [7], Batch [400/938], Loss: 0.2974\n",
            "Epoch [7], Batch [500/938], Loss: 0.2897\n",
            "Epoch [7], Batch [600/938], Loss: 0.3054\n",
            "Epoch [7], Batch [700/938], Loss: 0.2874\n",
            "Epoch [7], Batch [800/938], Loss: 0.2817\n",
            "Epoch [7], Batch [900/938], Loss: 0.3053\n",
            "Epoch [8], Batch [100/938], Loss: 0.2750\n",
            "Epoch [8], Batch [200/938], Loss: 0.2671\n",
            "Epoch [8], Batch [300/938], Loss: 0.2783\n",
            "Epoch [8], Batch [400/938], Loss: 0.2720\n",
            "Epoch [8], Batch [500/938], Loss: 0.2925\n",
            "Epoch [8], Batch [600/938], Loss: 0.3023\n",
            "Epoch [8], Batch [700/938], Loss: 0.2764\n",
            "Epoch [8], Batch [800/938], Loss: 0.2935\n",
            "Epoch [8], Batch [900/938], Loss: 0.2764\n",
            "Epoch [9], Batch [100/938], Loss: 0.2664\n",
            "Epoch [9], Batch [200/938], Loss: 0.2758\n",
            "Epoch [9], Batch [300/938], Loss: 0.2727\n",
            "Epoch [9], Batch [400/938], Loss: 0.2831\n",
            "Epoch [9], Batch [500/938], Loss: 0.2636\n",
            "Epoch [9], Batch [600/938], Loss: 0.2818\n",
            "Epoch [9], Batch [700/938], Loss: 0.2660\n",
            "Epoch [9], Batch [800/938], Loss: 0.2637\n",
            "Epoch [9], Batch [900/938], Loss: 0.2809\n",
            "Epoch [10], Batch [100/938], Loss: 0.2500\n",
            "Epoch [10], Batch [200/938], Loss: 0.2673\n",
            "Epoch [10], Batch [300/938], Loss: 0.2695\n",
            "Epoch [10], Batch [400/938], Loss: 0.2615\n",
            "Epoch [10], Batch [500/938], Loss: 0.2483\n",
            "Epoch [10], Batch [600/938], Loss: 0.2635\n",
            "Epoch [10], Batch [700/938], Loss: 0.2516\n",
            "Epoch [10], Batch [800/938], Loss: 0.2630\n",
            "Epoch [10], Batch [900/938], Loss: 0.2660\n",
            "Accuracy for Configuration-3: 87.50%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-4\n",
            "Input data:\n",
            "- Num Hidden Layers: 2\n",
            "- Hidden Layer Sizes: [64, 128]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.8498\n",
            "Epoch [1], Batch [200/938], Loss: 1.1225\n",
            "Epoch [1], Batch [300/938], Loss: 0.8254\n",
            "Epoch [1], Batch [400/938], Loss: 0.7020\n",
            "Epoch [1], Batch [500/938], Loss: 0.6065\n",
            "Epoch [1], Batch [600/938], Loss: 0.5631\n",
            "Epoch [1], Batch [700/938], Loss: 0.5315\n",
            "Epoch [1], Batch [800/938], Loss: 0.5214\n",
            "Epoch [1], Batch [900/938], Loss: 0.4591\n",
            "Epoch [2], Batch [100/938], Loss: 0.4554\n",
            "Epoch [2], Batch [200/938], Loss: 0.4365\n",
            "Epoch [2], Batch [300/938], Loss: 0.4412\n",
            "Epoch [2], Batch [400/938], Loss: 0.4386\n",
            "Epoch [2], Batch [500/938], Loss: 0.4193\n",
            "Epoch [2], Batch [600/938], Loss: 0.4142\n",
            "Epoch [2], Batch [700/938], Loss: 0.4288\n",
            "Epoch [2], Batch [800/938], Loss: 0.4034\n",
            "Epoch [2], Batch [900/938], Loss: 0.3901\n",
            "Epoch [3], Batch [100/938], Loss: 0.3798\n",
            "Epoch [3], Batch [200/938], Loss: 0.3757\n",
            "Epoch [3], Batch [300/938], Loss: 0.3917\n",
            "Epoch [3], Batch [400/938], Loss: 0.3847\n",
            "Epoch [3], Batch [500/938], Loss: 0.3513\n",
            "Epoch [3], Batch [600/938], Loss: 0.3677\n",
            "Epoch [3], Batch [700/938], Loss: 0.3605\n",
            "Epoch [3], Batch [800/938], Loss: 0.3629\n",
            "Epoch [3], Batch [900/938], Loss: 0.3733\n",
            "Epoch [4], Batch [100/938], Loss: 0.3320\n",
            "Epoch [4], Batch [200/938], Loss: 0.3453\n",
            "Epoch [4], Batch [300/938], Loss: 0.3504\n",
            "Epoch [4], Batch [400/938], Loss: 0.3607\n",
            "Epoch [4], Batch [500/938], Loss: 0.3423\n",
            "Epoch [4], Batch [600/938], Loss: 0.3512\n",
            "Epoch [4], Batch [700/938], Loss: 0.3468\n",
            "Epoch [4], Batch [800/938], Loss: 0.3344\n",
            "Epoch [4], Batch [900/938], Loss: 0.3261\n",
            "Epoch [5], Batch [100/938], Loss: 0.3269\n",
            "Epoch [5], Batch [200/938], Loss: 0.3248\n",
            "Epoch [5], Batch [300/938], Loss: 0.3212\n",
            "Epoch [5], Batch [400/938], Loss: 0.3228\n",
            "Epoch [5], Batch [500/938], Loss: 0.3325\n",
            "Epoch [5], Batch [600/938], Loss: 0.2944\n",
            "Epoch [5], Batch [700/938], Loss: 0.3292\n",
            "Epoch [5], Batch [800/938], Loss: 0.3286\n",
            "Epoch [5], Batch [900/938], Loss: 0.3342\n",
            "Epoch [6], Batch [100/938], Loss: 0.3023\n",
            "Epoch [6], Batch [200/938], Loss: 0.2910\n",
            "Epoch [6], Batch [300/938], Loss: 0.3242\n",
            "Epoch [6], Batch [400/938], Loss: 0.3092\n",
            "Epoch [6], Batch [500/938], Loss: 0.3043\n",
            "Epoch [6], Batch [600/938], Loss: 0.3083\n",
            "Epoch [6], Batch [700/938], Loss: 0.3101\n",
            "Epoch [6], Batch [800/938], Loss: 0.3220\n",
            "Epoch [6], Batch [900/938], Loss: 0.3092\n",
            "Epoch [7], Batch [100/938], Loss: 0.2885\n",
            "Epoch [7], Batch [200/938], Loss: 0.2969\n",
            "Epoch [7], Batch [300/938], Loss: 0.2878\n",
            "Epoch [7], Batch [400/938], Loss: 0.2900\n",
            "Epoch [7], Batch [500/938], Loss: 0.3077\n",
            "Epoch [7], Batch [600/938], Loss: 0.3129\n",
            "Epoch [7], Batch [700/938], Loss: 0.3077\n",
            "Epoch [7], Batch [800/938], Loss: 0.3010\n",
            "Epoch [7], Batch [900/938], Loss: 0.2996\n",
            "Epoch [8], Batch [100/938], Loss: 0.2812\n",
            "Epoch [8], Batch [200/938], Loss: 0.2706\n",
            "Epoch [8], Batch [300/938], Loss: 0.2892\n",
            "Epoch [8], Batch [400/938], Loss: 0.2898\n",
            "Epoch [8], Batch [500/938], Loss: 0.2908\n",
            "Epoch [8], Batch [600/938], Loss: 0.2769\n",
            "Epoch [8], Batch [700/938], Loss: 0.2913\n",
            "Epoch [8], Batch [800/938], Loss: 0.2947\n",
            "Epoch [8], Batch [900/938], Loss: 0.2842\n",
            "Epoch [9], Batch [100/938], Loss: 0.2751\n",
            "Epoch [9], Batch [200/938], Loss: 0.2677\n",
            "Epoch [9], Batch [300/938], Loss: 0.2809\n",
            "Epoch [9], Batch [400/938], Loss: 0.2783\n",
            "Epoch [9], Batch [500/938], Loss: 0.2777\n",
            "Epoch [9], Batch [600/938], Loss: 0.2674\n",
            "Epoch [9], Batch [700/938], Loss: 0.2857\n",
            "Epoch [9], Batch [800/938], Loss: 0.2775\n",
            "Epoch [9], Batch [900/938], Loss: 0.2866\n",
            "Epoch [10], Batch [100/938], Loss: 0.2751\n",
            "Epoch [10], Batch [200/938], Loss: 0.2680\n",
            "Epoch [10], Batch [300/938], Loss: 0.2713\n",
            "Epoch [10], Batch [400/938], Loss: 0.2793\n",
            "Epoch [10], Batch [500/938], Loss: 0.2775\n",
            "Epoch [10], Batch [600/938], Loss: 0.2744\n",
            "Epoch [10], Batch [700/938], Loss: 0.2713\n",
            "Epoch [10], Batch [800/938], Loss: 0.2586\n",
            "Epoch [10], Batch [900/938], Loss: 0.2677\n",
            "Accuracy for Configuration-4: 87.63%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-5\n",
            "Input data:\n",
            "- Num Hidden Layers: 2\n",
            "- Hidden Layer Sizes: [32, 64]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 2.0591\n",
            "Epoch [1], Batch [200/938], Loss: 1.4761\n",
            "Epoch [1], Batch [300/938], Loss: 1.1512\n",
            "Epoch [1], Batch [400/938], Loss: 0.9382\n",
            "Epoch [1], Batch [500/938], Loss: 0.7957\n",
            "Epoch [1], Batch [600/938], Loss: 0.7042\n",
            "Epoch [1], Batch [700/938], Loss: 0.6493\n",
            "Epoch [1], Batch [800/938], Loss: 0.6297\n",
            "Epoch [1], Batch [900/938], Loss: 0.5738\n",
            "Epoch [2], Batch [100/938], Loss: 0.5291\n",
            "Epoch [2], Batch [200/938], Loss: 0.5240\n",
            "Epoch [2], Batch [300/938], Loss: 0.4941\n",
            "Epoch [2], Batch [400/938], Loss: 0.5040\n",
            "Epoch [2], Batch [500/938], Loss: 0.4804\n",
            "Epoch [2], Batch [600/938], Loss: 0.4764\n",
            "Epoch [2], Batch [700/938], Loss: 0.4658\n",
            "Epoch [2], Batch [800/938], Loss: 0.4469\n",
            "Epoch [2], Batch [900/938], Loss: 0.4537\n",
            "Epoch [3], Batch [100/938], Loss: 0.4264\n",
            "Epoch [3], Batch [200/938], Loss: 0.4306\n",
            "Epoch [3], Batch [300/938], Loss: 0.4335\n",
            "Epoch [3], Batch [400/938], Loss: 0.4107\n",
            "Epoch [3], Batch [500/938], Loss: 0.4318\n",
            "Epoch [3], Batch [600/938], Loss: 0.4205\n",
            "Epoch [3], Batch [700/938], Loss: 0.4060\n",
            "Epoch [3], Batch [800/938], Loss: 0.4199\n",
            "Epoch [3], Batch [900/938], Loss: 0.3933\n",
            "Epoch [4], Batch [100/938], Loss: 0.3856\n",
            "Epoch [4], Batch [200/938], Loss: 0.3849\n",
            "Epoch [4], Batch [300/938], Loss: 0.3924\n",
            "Epoch [4], Batch [400/938], Loss: 0.3725\n",
            "Epoch [4], Batch [500/938], Loss: 0.3951\n",
            "Epoch [4], Batch [600/938], Loss: 0.3955\n",
            "Epoch [4], Batch [700/938], Loss: 0.3852\n",
            "Epoch [4], Batch [800/938], Loss: 0.3951\n",
            "Epoch [4], Batch [900/938], Loss: 0.4025\n",
            "Epoch [5], Batch [100/938], Loss: 0.3677\n",
            "Epoch [5], Batch [200/938], Loss: 0.3654\n",
            "Epoch [5], Batch [300/938], Loss: 0.3661\n",
            "Epoch [5], Batch [400/938], Loss: 0.3578\n",
            "Epoch [5], Batch [500/938], Loss: 0.3597\n",
            "Epoch [5], Batch [600/938], Loss: 0.3735\n",
            "Epoch [5], Batch [700/938], Loss: 0.3787\n",
            "Epoch [5], Batch [800/938], Loss: 0.3642\n",
            "Epoch [5], Batch [900/938], Loss: 0.3728\n",
            "Epoch [6], Batch [100/938], Loss: 0.3618\n",
            "Epoch [6], Batch [200/938], Loss: 0.3290\n",
            "Epoch [6], Batch [300/938], Loss: 0.3557\n",
            "Epoch [6], Batch [400/938], Loss: 0.3457\n",
            "Epoch [6], Batch [500/938], Loss: 0.3587\n",
            "Epoch [6], Batch [600/938], Loss: 0.3571\n",
            "Epoch [6], Batch [700/938], Loss: 0.3611\n",
            "Epoch [6], Batch [800/938], Loss: 0.3574\n",
            "Epoch [6], Batch [900/938], Loss: 0.3486\n",
            "Epoch [7], Batch [100/938], Loss: 0.3572\n",
            "Epoch [7], Batch [200/938], Loss: 0.3318\n",
            "Epoch [7], Batch [300/938], Loss: 0.3310\n",
            "Epoch [7], Batch [400/938], Loss: 0.3468\n",
            "Epoch [7], Batch [500/938], Loss: 0.3255\n",
            "Epoch [7], Batch [600/938], Loss: 0.3521\n",
            "Epoch [7], Batch [700/938], Loss: 0.3416\n",
            "Epoch [7], Batch [800/938], Loss: 0.3481\n",
            "Epoch [7], Batch [900/938], Loss: 0.3377\n",
            "Epoch [8], Batch [100/938], Loss: 0.3423\n",
            "Epoch [8], Batch [200/938], Loss: 0.3310\n",
            "Epoch [8], Batch [300/938], Loss: 0.3259\n",
            "Epoch [8], Batch [400/938], Loss: 0.3227\n",
            "Epoch [8], Batch [500/938], Loss: 0.3250\n",
            "Epoch [8], Batch [600/938], Loss: 0.3282\n",
            "Epoch [8], Batch [700/938], Loss: 0.3123\n",
            "Epoch [8], Batch [800/938], Loss: 0.3464\n",
            "Epoch [8], Batch [900/938], Loss: 0.3325\n",
            "Epoch [9], Batch [100/938], Loss: 0.3199\n",
            "Epoch [9], Batch [200/938], Loss: 0.3234\n",
            "Epoch [9], Batch [300/938], Loss: 0.3382\n",
            "Epoch [9], Batch [400/938], Loss: 0.3150\n",
            "Epoch [9], Batch [500/938], Loss: 0.3320\n",
            "Epoch [9], Batch [600/938], Loss: 0.3168\n",
            "Epoch [9], Batch [700/938], Loss: 0.3151\n",
            "Epoch [9], Batch [800/938], Loss: 0.3237\n",
            "Epoch [9], Batch [900/938], Loss: 0.3223\n",
            "Epoch [10], Batch [100/938], Loss: 0.3185\n",
            "Epoch [10], Batch [200/938], Loss: 0.3254\n",
            "Epoch [10], Batch [300/938], Loss: 0.2984\n",
            "Epoch [10], Batch [400/938], Loss: 0.3133\n",
            "Epoch [10], Batch [500/938], Loss: 0.3091\n",
            "Epoch [10], Batch [600/938], Loss: 0.3098\n",
            "Epoch [10], Batch [700/938], Loss: 0.3288\n",
            "Epoch [10], Batch [800/938], Loss: 0.3088\n",
            "Epoch [10], Batch [900/938], Loss: 0.3205\n",
            "Accuracy for Configuration-5: 86.13%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-6\n",
            "Input data:\n",
            "- Num Hidden Layers: 2\n",
            "- Hidden Layer Sizes: [128, 256]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.5670\n",
            "Epoch [1], Batch [200/938], Loss: 0.7956\n",
            "Epoch [1], Batch [300/938], Loss: 0.6306\n",
            "Epoch [1], Batch [400/938], Loss: 0.5568\n",
            "Epoch [1], Batch [500/938], Loss: 0.5099\n",
            "Epoch [1], Batch [600/938], Loss: 0.4876\n",
            "Epoch [1], Batch [700/938], Loss: 0.4785\n",
            "Epoch [1], Batch [800/938], Loss: 0.4447\n",
            "Epoch [1], Batch [900/938], Loss: 0.4366\n",
            "Epoch [2], Batch [100/938], Loss: 0.4179\n",
            "Epoch [2], Batch [200/938], Loss: 0.3982\n",
            "Epoch [2], Batch [300/938], Loss: 0.4187\n",
            "Epoch [2], Batch [400/938], Loss: 0.3986\n",
            "Epoch [2], Batch [500/938], Loss: 0.4009\n",
            "Epoch [2], Batch [600/938], Loss: 0.3764\n",
            "Epoch [2], Batch [700/938], Loss: 0.3915\n",
            "Epoch [2], Batch [800/938], Loss: 0.3836\n",
            "Epoch [2], Batch [900/938], Loss: 0.3871\n",
            "Epoch [3], Batch [100/938], Loss: 0.3715\n",
            "Epoch [3], Batch [200/938], Loss: 0.3643\n",
            "Epoch [3], Batch [300/938], Loss: 0.3663\n",
            "Epoch [3], Batch [400/938], Loss: 0.3422\n",
            "Epoch [3], Batch [500/938], Loss: 0.3605\n",
            "Epoch [3], Batch [600/938], Loss: 0.3454\n",
            "Epoch [3], Batch [700/938], Loss: 0.3473\n",
            "Epoch [3], Batch [800/938], Loss: 0.3296\n",
            "Epoch [3], Batch [900/938], Loss: 0.3556\n",
            "Epoch [4], Batch [100/938], Loss: 0.3272\n",
            "Epoch [4], Batch [200/938], Loss: 0.3252\n",
            "Epoch [4], Batch [300/938], Loss: 0.3379\n",
            "Epoch [4], Batch [400/938], Loss: 0.3358\n",
            "Epoch [4], Batch [500/938], Loss: 0.3343\n",
            "Epoch [4], Batch [600/938], Loss: 0.3255\n",
            "Epoch [4], Batch [700/938], Loss: 0.3104\n",
            "Epoch [4], Batch [800/938], Loss: 0.3126\n",
            "Epoch [4], Batch [900/938], Loss: 0.3263\n",
            "Epoch [5], Batch [100/938], Loss: 0.2939\n",
            "Epoch [5], Batch [200/938], Loss: 0.3027\n",
            "Epoch [5], Batch [300/938], Loss: 0.3224\n",
            "Epoch [5], Batch [400/938], Loss: 0.2903\n",
            "Epoch [5], Batch [500/938], Loss: 0.3106\n",
            "Epoch [5], Batch [600/938], Loss: 0.3091\n",
            "Epoch [5], Batch [700/938], Loss: 0.3058\n",
            "Epoch [5], Batch [800/938], Loss: 0.3187\n",
            "Epoch [5], Batch [900/938], Loss: 0.3021\n",
            "Epoch [6], Batch [100/938], Loss: 0.2769\n",
            "Epoch [6], Batch [200/938], Loss: 0.2964\n",
            "Epoch [6], Batch [300/938], Loss: 0.3051\n",
            "Epoch [6], Batch [400/938], Loss: 0.2913\n",
            "Epoch [6], Batch [500/938], Loss: 0.2857\n",
            "Epoch [6], Batch [600/938], Loss: 0.2966\n",
            "Epoch [6], Batch [700/938], Loss: 0.2967\n",
            "Epoch [6], Batch [800/938], Loss: 0.2828\n",
            "Epoch [6], Batch [900/938], Loss: 0.2881\n",
            "Epoch [7], Batch [100/938], Loss: 0.2650\n",
            "Epoch [7], Batch [200/938], Loss: 0.2609\n",
            "Epoch [7], Batch [300/938], Loss: 0.2750\n",
            "Epoch [7], Batch [400/938], Loss: 0.2862\n",
            "Epoch [7], Batch [500/938], Loss: 0.2685\n",
            "Epoch [7], Batch [600/938], Loss: 0.2937\n",
            "Epoch [7], Batch [700/938], Loss: 0.2889\n",
            "Epoch [7], Batch [800/938], Loss: 0.2684\n",
            "Epoch [7], Batch [900/938], Loss: 0.2842\n",
            "Epoch [8], Batch [100/938], Loss: 0.2709\n",
            "Epoch [8], Batch [200/938], Loss: 0.2635\n",
            "Epoch [8], Batch [300/938], Loss: 0.2645\n",
            "Epoch [8], Batch [400/938], Loss: 0.2392\n",
            "Epoch [8], Batch [500/938], Loss: 0.2708\n",
            "Epoch [8], Batch [600/938], Loss: 0.2796\n",
            "Epoch [8], Batch [700/938], Loss: 0.2667\n",
            "Epoch [8], Batch [800/938], Loss: 0.2745\n",
            "Epoch [8], Batch [900/938], Loss: 0.2641\n",
            "Epoch [9], Batch [100/938], Loss: 0.2582\n",
            "Epoch [9], Batch [200/938], Loss: 0.2494\n",
            "Epoch [9], Batch [300/938], Loss: 0.2512\n",
            "Epoch [9], Batch [400/938], Loss: 0.2477\n",
            "Epoch [9], Batch [500/938], Loss: 0.2681\n",
            "Epoch [9], Batch [600/938], Loss: 0.2526\n",
            "Epoch [9], Batch [700/938], Loss: 0.2480\n",
            "Epoch [9], Batch [800/938], Loss: 0.2541\n",
            "Epoch [9], Batch [900/938], Loss: 0.2717\n",
            "Epoch [10], Batch [100/938], Loss: 0.2441\n",
            "Epoch [10], Batch [200/938], Loss: 0.2472\n",
            "Epoch [10], Batch [300/938], Loss: 0.2428\n",
            "Epoch [10], Batch [400/938], Loss: 0.2482\n",
            "Epoch [10], Batch [500/938], Loss: 0.2555\n",
            "Epoch [10], Batch [600/938], Loss: 0.2512\n",
            "Epoch [10], Batch [700/938], Loss: 0.2530\n",
            "Epoch [10], Batch [800/938], Loss: 0.2471\n",
            "Epoch [10], Batch [900/938], Loss: 0.2259\n",
            "Accuracy for Configuration-6: 87.84%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-7\n",
            "Input data:\n",
            "- Num Hidden Layers: 2\n",
            "- Hidden Layer Sizes: [256, 512]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.0660\n",
            "Epoch [1], Batch [200/938], Loss: 0.5908\n",
            "Epoch [1], Batch [300/938], Loss: 0.5509\n",
            "Epoch [1], Batch [400/938], Loss: 0.4987\n",
            "Epoch [1], Batch [500/938], Loss: 0.4684\n",
            "Epoch [1], Batch [600/938], Loss: 0.4511\n",
            "Epoch [1], Batch [700/938], Loss: 0.4617\n",
            "Epoch [1], Batch [800/938], Loss: 0.4457\n",
            "Epoch [1], Batch [900/938], Loss: 0.4258\n",
            "Epoch [2], Batch [100/938], Loss: 0.3877\n",
            "Epoch [2], Batch [200/938], Loss: 0.4152\n",
            "Epoch [2], Batch [300/938], Loss: 0.3790\n",
            "Epoch [2], Batch [400/938], Loss: 0.3818\n",
            "Epoch [2], Batch [500/938], Loss: 0.3764\n",
            "Epoch [2], Batch [600/938], Loss: 0.3873\n",
            "Epoch [2], Batch [700/938], Loss: 0.3776\n",
            "Epoch [2], Batch [800/938], Loss: 0.3868\n",
            "Epoch [2], Batch [900/938], Loss: 0.3787\n",
            "Epoch [3], Batch [100/938], Loss: 0.3362\n",
            "Epoch [3], Batch [200/938], Loss: 0.3492\n",
            "Epoch [3], Batch [300/938], Loss: 0.3440\n",
            "Epoch [3], Batch [400/938], Loss: 0.3504\n",
            "Epoch [3], Batch [500/938], Loss: 0.3528\n",
            "Epoch [3], Batch [600/938], Loss: 0.3416\n",
            "Epoch [3], Batch [700/938], Loss: 0.3474\n",
            "Epoch [3], Batch [800/938], Loss: 0.3301\n",
            "Epoch [3], Batch [900/938], Loss: 0.3452\n",
            "Epoch [4], Batch [100/938], Loss: 0.3334\n",
            "Epoch [4], Batch [200/938], Loss: 0.3199\n",
            "Epoch [4], Batch [300/938], Loss: 0.3056\n",
            "Epoch [4], Batch [400/938], Loss: 0.3243\n",
            "Epoch [4], Batch [500/938], Loss: 0.3149\n",
            "Epoch [4], Batch [600/938], Loss: 0.3142\n",
            "Epoch [4], Batch [700/938], Loss: 0.3083\n",
            "Epoch [4], Batch [800/938], Loss: 0.3247\n",
            "Epoch [4], Batch [900/938], Loss: 0.3076\n",
            "Epoch [5], Batch [100/938], Loss: 0.2967\n",
            "Epoch [5], Batch [200/938], Loss: 0.3034\n",
            "Epoch [5], Batch [300/938], Loss: 0.3156\n",
            "Epoch [5], Batch [400/938], Loss: 0.2891\n",
            "Epoch [5], Batch [500/938], Loss: 0.2967\n",
            "Epoch [5], Batch [600/938], Loss: 0.2866\n",
            "Epoch [5], Batch [700/938], Loss: 0.2822\n",
            "Epoch [5], Batch [800/938], Loss: 0.2961\n",
            "Epoch [5], Batch [900/938], Loss: 0.2955\n",
            "Epoch [6], Batch [100/938], Loss: 0.2829\n",
            "Epoch [6], Batch [200/938], Loss: 0.2670\n",
            "Epoch [6], Batch [300/938], Loss: 0.2877\n",
            "Epoch [6], Batch [400/938], Loss: 0.2735\n",
            "Epoch [6], Batch [500/938], Loss: 0.2680\n",
            "Epoch [6], Batch [600/938], Loss: 0.2777\n",
            "Epoch [6], Batch [700/938], Loss: 0.2852\n",
            "Epoch [6], Batch [800/938], Loss: 0.2839\n",
            "Epoch [6], Batch [900/938], Loss: 0.2807\n",
            "Epoch [7], Batch [100/938], Loss: 0.2568\n",
            "Epoch [7], Batch [200/938], Loss: 0.2740\n",
            "Epoch [7], Batch [300/938], Loss: 0.2630\n",
            "Epoch [7], Batch [400/938], Loss: 0.2841\n",
            "Epoch [7], Batch [500/938], Loss: 0.2532\n",
            "Epoch [7], Batch [600/938], Loss: 0.2620\n",
            "Epoch [7], Batch [700/938], Loss: 0.2579\n",
            "Epoch [7], Batch [800/938], Loss: 0.2614\n",
            "Epoch [7], Batch [900/938], Loss: 0.2622\n",
            "Epoch [8], Batch [100/938], Loss: 0.2426\n",
            "Epoch [8], Batch [200/938], Loss: 0.2534\n",
            "Epoch [8], Batch [300/938], Loss: 0.2634\n",
            "Epoch [8], Batch [400/938], Loss: 0.2486\n",
            "Epoch [8], Batch [500/938], Loss: 0.2591\n",
            "Epoch [8], Batch [600/938], Loss: 0.2512\n",
            "Epoch [8], Batch [700/938], Loss: 0.2475\n",
            "Epoch [8], Batch [800/938], Loss: 0.2437\n",
            "Epoch [8], Batch [900/938], Loss: 0.2481\n",
            "Epoch [9], Batch [100/938], Loss: 0.2385\n",
            "Epoch [9], Batch [200/938], Loss: 0.2247\n",
            "Epoch [9], Batch [300/938], Loss: 0.2454\n",
            "Epoch [9], Batch [400/938], Loss: 0.2230\n",
            "Epoch [9], Batch [500/938], Loss: 0.2406\n",
            "Epoch [9], Batch [600/938], Loss: 0.2375\n",
            "Epoch [9], Batch [700/938], Loss: 0.2349\n",
            "Epoch [9], Batch [800/938], Loss: 0.2396\n",
            "Epoch [9], Batch [900/938], Loss: 0.2390\n",
            "Epoch [10], Batch [100/938], Loss: 0.2084\n",
            "Epoch [10], Batch [200/938], Loss: 0.2209\n",
            "Epoch [10], Batch [300/938], Loss: 0.2257\n",
            "Epoch [10], Batch [400/938], Loss: 0.2237\n",
            "Epoch [10], Batch [500/938], Loss: 0.2365\n",
            "Epoch [10], Batch [600/938], Loss: 0.2427\n",
            "Epoch [10], Batch [700/938], Loss: 0.2122\n",
            "Epoch [10], Batch [800/938], Loss: 0.2333\n",
            "Epoch [10], Batch [900/938], Loss: 0.2318\n",
            "Accuracy for Configuration-7: 88.33%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-8\n",
            "Input data:\n",
            "- Num Hidden Layers: 1\n",
            "- Hidden Layer Sizes: [64]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.3881\n",
            "Epoch [1], Batch [200/938], Loss: 0.8128\n",
            "Epoch [1], Batch [300/938], Loss: 0.6485\n",
            "Epoch [1], Batch [400/938], Loss: 0.5760\n",
            "Epoch [1], Batch [500/938], Loss: 0.5448\n",
            "Epoch [1], Batch [600/938], Loss: 0.5123\n",
            "Epoch [1], Batch [700/938], Loss: 0.5056\n",
            "Epoch [1], Batch [800/938], Loss: 0.4690\n",
            "Epoch [1], Batch [900/938], Loss: 0.4593\n",
            "Epoch [2], Batch [100/938], Loss: 0.4534\n",
            "Epoch [2], Batch [200/938], Loss: 0.4156\n",
            "Epoch [2], Batch [300/938], Loss: 0.4242\n",
            "Epoch [2], Batch [400/938], Loss: 0.4317\n",
            "Epoch [2], Batch [500/938], Loss: 0.3998\n",
            "Epoch [2], Batch [600/938], Loss: 0.4143\n",
            "Epoch [2], Batch [700/938], Loss: 0.4115\n",
            "Epoch [2], Batch [800/938], Loss: 0.4026\n",
            "Epoch [2], Batch [900/938], Loss: 0.4115\n",
            "Epoch [3], Batch [100/938], Loss: 0.3876\n",
            "Epoch [3], Batch [200/938], Loss: 0.3948\n",
            "Epoch [3], Batch [300/938], Loss: 0.4036\n",
            "Epoch [3], Batch [400/938], Loss: 0.3638\n",
            "Epoch [3], Batch [500/938], Loss: 0.3639\n",
            "Epoch [3], Batch [600/938], Loss: 0.3776\n",
            "Epoch [3], Batch [700/938], Loss: 0.3800\n",
            "Epoch [3], Batch [800/938], Loss: 0.3764\n",
            "Epoch [3], Batch [900/938], Loss: 0.3622\n",
            "Epoch [4], Batch [100/938], Loss: 0.3552\n",
            "Epoch [4], Batch [200/938], Loss: 0.3579\n",
            "Epoch [4], Batch [300/938], Loss: 0.3456\n",
            "Epoch [4], Batch [400/938], Loss: 0.3592\n",
            "Epoch [4], Batch [500/938], Loss: 0.3576\n",
            "Epoch [4], Batch [600/938], Loss: 0.3541\n",
            "Epoch [4], Batch [700/938], Loss: 0.3470\n",
            "Epoch [4], Batch [800/938], Loss: 0.3566\n",
            "Epoch [4], Batch [900/938], Loss: 0.3492\n",
            "Epoch [5], Batch [100/938], Loss: 0.3454\n",
            "Epoch [5], Batch [200/938], Loss: 0.3423\n",
            "Epoch [5], Batch [300/938], Loss: 0.3387\n",
            "Epoch [5], Batch [400/938], Loss: 0.3311\n",
            "Epoch [5], Batch [500/938], Loss: 0.3420\n",
            "Epoch [5], Batch [600/938], Loss: 0.3434\n",
            "Epoch [5], Batch [700/938], Loss: 0.3351\n",
            "Epoch [5], Batch [800/938], Loss: 0.3332\n",
            "Epoch [5], Batch [900/938], Loss: 0.3186\n",
            "Epoch [6], Batch [100/938], Loss: 0.3131\n",
            "Epoch [6], Batch [200/938], Loss: 0.3266\n",
            "Epoch [6], Batch [300/938], Loss: 0.3238\n",
            "Epoch [6], Batch [400/938], Loss: 0.3164\n",
            "Epoch [6], Batch [500/938], Loss: 0.3215\n",
            "Epoch [6], Batch [600/938], Loss: 0.3316\n",
            "Epoch [6], Batch [700/938], Loss: 0.3043\n",
            "Epoch [6], Batch [800/938], Loss: 0.3285\n",
            "Epoch [6], Batch [900/938], Loss: 0.3224\n",
            "Epoch [7], Batch [100/938], Loss: 0.3123\n",
            "Epoch [7], Batch [200/938], Loss: 0.3068\n",
            "Epoch [7], Batch [300/938], Loss: 0.3219\n",
            "Epoch [7], Batch [400/938], Loss: 0.3133\n",
            "Epoch [7], Batch [500/938], Loss: 0.3001\n",
            "Epoch [7], Batch [600/938], Loss: 0.3103\n",
            "Epoch [7], Batch [700/938], Loss: 0.3096\n",
            "Epoch [7], Batch [800/938], Loss: 0.3052\n",
            "Epoch [7], Batch [900/938], Loss: 0.3111\n",
            "Epoch [8], Batch [100/938], Loss: 0.3096\n",
            "Epoch [8], Batch [200/938], Loss: 0.3142\n",
            "Epoch [8], Batch [300/938], Loss: 0.2831\n",
            "Epoch [8], Batch [400/938], Loss: 0.2938\n",
            "Epoch [8], Batch [500/938], Loss: 0.3039\n",
            "Epoch [8], Batch [600/938], Loss: 0.2993\n",
            "Epoch [8], Batch [700/938], Loss: 0.2981\n",
            "Epoch [8], Batch [800/938], Loss: 0.2902\n",
            "Epoch [8], Batch [900/938], Loss: 0.3045\n",
            "Epoch [9], Batch [100/938], Loss: 0.2899\n",
            "Epoch [9], Batch [200/938], Loss: 0.2920\n",
            "Epoch [9], Batch [300/938], Loss: 0.2833\n",
            "Epoch [9], Batch [400/938], Loss: 0.2761\n",
            "Epoch [9], Batch [500/938], Loss: 0.2896\n",
            "Epoch [9], Batch [600/938], Loss: 0.2948\n",
            "Epoch [9], Batch [700/938], Loss: 0.2925\n",
            "Epoch [9], Batch [800/938], Loss: 0.2945\n",
            "Epoch [9], Batch [900/938], Loss: 0.2968\n",
            "Epoch [10], Batch [100/938], Loss: 0.2883\n",
            "Epoch [10], Batch [200/938], Loss: 0.2673\n",
            "Epoch [10], Batch [300/938], Loss: 0.2904\n",
            "Epoch [10], Batch [400/938], Loss: 0.2799\n",
            "Epoch [10], Batch [500/938], Loss: 0.2764\n",
            "Epoch [10], Batch [600/938], Loss: 0.2860\n",
            "Epoch [10], Batch [700/938], Loss: 0.2779\n",
            "Epoch [10], Batch [800/938], Loss: 0.2825\n",
            "Epoch [10], Batch [900/938], Loss: 0.2749\n",
            "Accuracy for Configuration-8: 87.41%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-9\n",
            "Input data:\n",
            "- Num Hidden Layers: 1\n",
            "- Hidden Layer Sizes: [128]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.1574\n",
            "Epoch [1], Batch [200/938], Loss: 0.6670\n",
            "Epoch [1], Batch [300/938], Loss: 0.5880\n",
            "Epoch [1], Batch [400/938], Loss: 0.5295\n",
            "Epoch [1], Batch [500/938], Loss: 0.4997\n",
            "Epoch [1], Batch [600/938], Loss: 0.4752\n",
            "Epoch [1], Batch [700/938], Loss: 0.4716\n",
            "Epoch [1], Batch [800/938], Loss: 0.4516\n",
            "Epoch [1], Batch [900/938], Loss: 0.4360\n",
            "Epoch [2], Batch [100/938], Loss: 0.4287\n",
            "Epoch [2], Batch [200/938], Loss: 0.4142\n",
            "Epoch [2], Batch [300/938], Loss: 0.4089\n",
            "Epoch [2], Batch [400/938], Loss: 0.4019\n",
            "Epoch [2], Batch [500/938], Loss: 0.4005\n",
            "Epoch [2], Batch [600/938], Loss: 0.3866\n",
            "Epoch [2], Batch [700/938], Loss: 0.3915\n",
            "Epoch [2], Batch [800/938], Loss: 0.3793\n",
            "Epoch [2], Batch [900/938], Loss: 0.3856\n",
            "Epoch [3], Batch [100/938], Loss: 0.3768\n",
            "Epoch [3], Batch [200/938], Loss: 0.3617\n",
            "Epoch [3], Batch [300/938], Loss: 0.3577\n",
            "Epoch [3], Batch [400/938], Loss: 0.3625\n",
            "Epoch [3], Batch [500/938], Loss: 0.3524\n",
            "Epoch [3], Batch [600/938], Loss: 0.3691\n",
            "Epoch [3], Batch [700/938], Loss: 0.3477\n",
            "Epoch [3], Batch [800/938], Loss: 0.3716\n",
            "Epoch [3], Batch [900/938], Loss: 0.3470\n",
            "Epoch [4], Batch [100/938], Loss: 0.3491\n",
            "Epoch [4], Batch [200/938], Loss: 0.3316\n",
            "Epoch [4], Batch [300/938], Loss: 0.3411\n",
            "Epoch [4], Batch [400/938], Loss: 0.3432\n",
            "Epoch [4], Batch [500/938], Loss: 0.3362\n",
            "Epoch [4], Batch [600/938], Loss: 0.3362\n",
            "Epoch [4], Batch [700/938], Loss: 0.3311\n",
            "Epoch [4], Batch [800/938], Loss: 0.3326\n",
            "Epoch [4], Batch [900/938], Loss: 0.3158\n",
            "Epoch [5], Batch [100/938], Loss: 0.3267\n",
            "Epoch [5], Batch [200/938], Loss: 0.3127\n",
            "Epoch [5], Batch [300/938], Loss: 0.3015\n",
            "Epoch [5], Batch [400/938], Loss: 0.3141\n",
            "Epoch [5], Batch [500/938], Loss: 0.3203\n",
            "Epoch [5], Batch [600/938], Loss: 0.3164\n",
            "Epoch [5], Batch [700/938], Loss: 0.3102\n",
            "Epoch [5], Batch [800/938], Loss: 0.3227\n",
            "Epoch [5], Batch [900/938], Loss: 0.3089\n",
            "Epoch [6], Batch [100/938], Loss: 0.3000\n",
            "Epoch [6], Batch [200/938], Loss: 0.3113\n",
            "Epoch [6], Batch [300/938], Loss: 0.2959\n",
            "Epoch [6], Batch [400/938], Loss: 0.2875\n",
            "Epoch [6], Batch [500/938], Loss: 0.2933\n",
            "Epoch [6], Batch [600/938], Loss: 0.2856\n",
            "Epoch [6], Batch [700/938], Loss: 0.3063\n",
            "Epoch [6], Batch [800/938], Loss: 0.3095\n",
            "Epoch [6], Batch [900/938], Loss: 0.3163\n",
            "Epoch [7], Batch [100/938], Loss: 0.2800\n",
            "Epoch [7], Batch [200/938], Loss: 0.2743\n",
            "Epoch [7], Batch [300/938], Loss: 0.3025\n",
            "Epoch [7], Batch [400/938], Loss: 0.2988\n",
            "Epoch [7], Batch [500/938], Loss: 0.2907\n",
            "Epoch [7], Batch [600/938], Loss: 0.2662\n",
            "Epoch [7], Batch [700/938], Loss: 0.2701\n",
            "Epoch [7], Batch [800/938], Loss: 0.2875\n",
            "Epoch [7], Batch [900/938], Loss: 0.2781\n",
            "Epoch [8], Batch [100/938], Loss: 0.2762\n",
            "Epoch [8], Batch [200/938], Loss: 0.2736\n",
            "Epoch [8], Batch [300/938], Loss: 0.2730\n",
            "Epoch [8], Batch [400/938], Loss: 0.2838\n",
            "Epoch [8], Batch [500/938], Loss: 0.2599\n",
            "Epoch [8], Batch [600/938], Loss: 0.2751\n",
            "Epoch [8], Batch [700/938], Loss: 0.2732\n",
            "Epoch [8], Batch [800/938], Loss: 0.2622\n",
            "Epoch [8], Batch [900/938], Loss: 0.2850\n",
            "Epoch [9], Batch [100/938], Loss: 0.2494\n",
            "Epoch [9], Batch [200/938], Loss: 0.2475\n",
            "Epoch [9], Batch [300/938], Loss: 0.2711\n",
            "Epoch [9], Batch [400/938], Loss: 0.2702\n",
            "Epoch [9], Batch [500/938], Loss: 0.2773\n",
            "Epoch [9], Batch [600/938], Loss: 0.2539\n",
            "Epoch [9], Batch [700/938], Loss: 0.2719\n",
            "Epoch [9], Batch [800/938], Loss: 0.2598\n",
            "Epoch [9], Batch [900/938], Loss: 0.2518\n",
            "Epoch [10], Batch [100/938], Loss: 0.2493\n",
            "Epoch [10], Batch [200/938], Loss: 0.2511\n",
            "Epoch [10], Batch [300/938], Loss: 0.2454\n",
            "Epoch [10], Batch [400/938], Loss: 0.2566\n",
            "Epoch [10], Batch [500/938], Loss: 0.2576\n",
            "Epoch [10], Batch [600/938], Loss: 0.2572\n",
            "Epoch [10], Batch [700/938], Loss: 0.2390\n",
            "Epoch [10], Batch [800/938], Loss: 0.2481\n",
            "Epoch [10], Batch [900/938], Loss: 0.2625\n",
            "Accuracy for Configuration-9: 87.92%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-10\n",
            "Input data:\n",
            "- Num Hidden Layers: 1\n",
            "- Hidden Layer Sizes: [256]\n",
            "- Batch Size: 64\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.0064\n",
            "Epoch [1], Batch [200/938], Loss: 0.5938\n",
            "Epoch [1], Batch [300/938], Loss: 0.5557\n",
            "Epoch [1], Batch [400/938], Loss: 0.4903\n",
            "Epoch [1], Batch [500/938], Loss: 0.4707\n",
            "Epoch [1], Batch [600/938], Loss: 0.4675\n",
            "Epoch [1], Batch [700/938], Loss: 0.4521\n",
            "Epoch [1], Batch [800/938], Loss: 0.4365\n",
            "Epoch [1], Batch [900/938], Loss: 0.4406\n",
            "Epoch [2], Batch [100/938], Loss: 0.3963\n",
            "Epoch [2], Batch [200/938], Loss: 0.4133\n",
            "Epoch [2], Batch [300/938], Loss: 0.3966\n",
            "Epoch [2], Batch [400/938], Loss: 0.3960\n",
            "Epoch [2], Batch [500/938], Loss: 0.3878\n",
            "Epoch [2], Batch [600/938], Loss: 0.3757\n",
            "Epoch [2], Batch [700/938], Loss: 0.3764\n",
            "Epoch [2], Batch [800/938], Loss: 0.3811\n",
            "Epoch [2], Batch [900/938], Loss: 0.3776\n",
            "Epoch [3], Batch [100/938], Loss: 0.3620\n",
            "Epoch [3], Batch [200/938], Loss: 0.3682\n",
            "Epoch [3], Batch [300/938], Loss: 0.3439\n",
            "Epoch [3], Batch [400/938], Loss: 0.3462\n",
            "Epoch [3], Batch [500/938], Loss: 0.3272\n",
            "Epoch [3], Batch [600/938], Loss: 0.3569\n",
            "Epoch [3], Batch [700/938], Loss: 0.3428\n",
            "Epoch [3], Batch [800/938], Loss: 0.3624\n",
            "Epoch [3], Batch [900/938], Loss: 0.3347\n",
            "Epoch [4], Batch [100/938], Loss: 0.3192\n",
            "Epoch [4], Batch [200/938], Loss: 0.3268\n",
            "Epoch [4], Batch [300/938], Loss: 0.3274\n",
            "Epoch [4], Batch [400/938], Loss: 0.3286\n",
            "Epoch [4], Batch [500/938], Loss: 0.3199\n",
            "Epoch [4], Batch [600/938], Loss: 0.3220\n",
            "Epoch [4], Batch [700/938], Loss: 0.3164\n",
            "Epoch [4], Batch [800/938], Loss: 0.3062\n",
            "Epoch [4], Batch [900/938], Loss: 0.3297\n",
            "Epoch [5], Batch [100/938], Loss: 0.3215\n",
            "Epoch [5], Batch [200/938], Loss: 0.3130\n",
            "Epoch [5], Batch [300/938], Loss: 0.3063\n",
            "Epoch [5], Batch [400/938], Loss: 0.2923\n",
            "Epoch [5], Batch [500/938], Loss: 0.3087\n",
            "Epoch [5], Batch [600/938], Loss: 0.2866\n",
            "Epoch [5], Batch [700/938], Loss: 0.2965\n",
            "Epoch [5], Batch [800/938], Loss: 0.2977\n",
            "Epoch [5], Batch [900/938], Loss: 0.2948\n",
            "Epoch [6], Batch [100/938], Loss: 0.2851\n",
            "Epoch [6], Batch [200/938], Loss: 0.2932\n",
            "Epoch [6], Batch [300/938], Loss: 0.2776\n",
            "Epoch [6], Batch [400/938], Loss: 0.2920\n",
            "Epoch [6], Batch [500/938], Loss: 0.2734\n",
            "Epoch [6], Batch [600/938], Loss: 0.2964\n",
            "Epoch [6], Batch [700/938], Loss: 0.2795\n",
            "Epoch [6], Batch [800/938], Loss: 0.2621\n",
            "Epoch [6], Batch [900/938], Loss: 0.2836\n",
            "Epoch [7], Batch [100/938], Loss: 0.2573\n",
            "Epoch [7], Batch [200/938], Loss: 0.2668\n",
            "Epoch [7], Batch [300/938], Loss: 0.2691\n",
            "Epoch [7], Batch [400/938], Loss: 0.2706\n",
            "Epoch [7], Batch [500/938], Loss: 0.2712\n",
            "Epoch [7], Batch [600/938], Loss: 0.2788\n",
            "Epoch [7], Batch [700/938], Loss: 0.2655\n",
            "Epoch [7], Batch [800/938], Loss: 0.2584\n",
            "Epoch [7], Batch [900/938], Loss: 0.2639\n",
            "Epoch [8], Batch [100/938], Loss: 0.2542\n",
            "Epoch [8], Batch [200/938], Loss: 0.2489\n",
            "Epoch [8], Batch [300/938], Loss: 0.2553\n",
            "Epoch [8], Batch [400/938], Loss: 0.2539\n",
            "Epoch [8], Batch [500/938], Loss: 0.2476\n",
            "Epoch [8], Batch [600/938], Loss: 0.2554\n",
            "Epoch [8], Batch [700/938], Loss: 0.2686\n",
            "Epoch [8], Batch [800/938], Loss: 0.2504\n",
            "Epoch [8], Batch [900/938], Loss: 0.2640\n",
            "Epoch [9], Batch [100/938], Loss: 0.2526\n",
            "Epoch [9], Batch [200/938], Loss: 0.2334\n",
            "Epoch [9], Batch [300/938], Loss: 0.2479\n",
            "Epoch [9], Batch [400/938], Loss: 0.2306\n",
            "Epoch [9], Batch [500/938], Loss: 0.2382\n",
            "Epoch [9], Batch [600/938], Loss: 0.2397\n",
            "Epoch [9], Batch [700/938], Loss: 0.2435\n",
            "Epoch [9], Batch [800/938], Loss: 0.2377\n",
            "Epoch [9], Batch [900/938], Loss: 0.2570\n",
            "Epoch [10], Batch [100/938], Loss: 0.2379\n",
            "Epoch [10], Batch [200/938], Loss: 0.2131\n",
            "Epoch [10], Batch [300/938], Loss: 0.2227\n",
            "Epoch [10], Batch [400/938], Loss: 0.2481\n",
            "Epoch [10], Batch [500/938], Loss: 0.2391\n",
            "Epoch [10], Batch [600/938], Loss: 0.2221\n",
            "Epoch [10], Batch [700/938], Loss: 0.2320\n",
            "Epoch [10], Batch [800/938], Loss: 0.2407\n",
            "Epoch [10], Batch [900/938], Loss: 0.2372\n",
            "Accuracy for Configuration-10: 87.59%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-11\n",
            "Input data:\n",
            "- Num Hidden Layers: 3\n",
            "- Hidden Layer Sizes: [32, 64, 128]\n",
            "- Batch Size: 128\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.8350\n",
            "Epoch [1], Batch [200/938], Loss: 1.4774\n",
            "Epoch [1], Batch [300/938], Loss: 1.2719\n",
            "Epoch [1], Batch [400/938], Loss: 1.1355\n",
            "Epoch [1], Batch [500/938], Loss: 1.0321\n",
            "Epoch [1], Batch [600/938], Loss: 0.9338\n",
            "Epoch [1], Batch [700/938], Loss: 0.8622\n",
            "Epoch [1], Batch [800/938], Loss: 0.8283\n",
            "Epoch [1], Batch [900/938], Loss: 0.8038\n",
            "Epoch [2], Batch [100/938], Loss: 0.7436\n",
            "Epoch [2], Batch [200/938], Loss: 0.7239\n",
            "Epoch [2], Batch [300/938], Loss: 0.7135\n",
            "Epoch [2], Batch [400/938], Loss: 0.6921\n",
            "Epoch [2], Batch [500/938], Loss: 0.7020\n",
            "Epoch [2], Batch [600/938], Loss: 0.6731\n",
            "Epoch [2], Batch [700/938], Loss: 0.6689\n",
            "Epoch [2], Batch [800/938], Loss: 0.6449\n",
            "Epoch [2], Batch [900/938], Loss: 0.6406\n",
            "Epoch [3], Batch [100/938], Loss: 0.5877\n",
            "Epoch [3], Batch [200/938], Loss: 0.5800\n",
            "Epoch [3], Batch [300/938], Loss: 0.5615\n",
            "Epoch [3], Batch [400/938], Loss: 0.5661\n",
            "Epoch [3], Batch [500/938], Loss: 0.5258\n",
            "Epoch [3], Batch [600/938], Loss: 0.5155\n",
            "Epoch [3], Batch [700/938], Loss: 0.5330\n",
            "Epoch [3], Batch [800/938], Loss: 0.5139\n",
            "Epoch [3], Batch [900/938], Loss: 0.5043\n",
            "Epoch [4], Batch [100/938], Loss: 0.4793\n",
            "Epoch [4], Batch [200/938], Loss: 0.4923\n",
            "Epoch [4], Batch [300/938], Loss: 0.4816\n",
            "Epoch [4], Batch [400/938], Loss: 0.4568\n",
            "Epoch [4], Batch [500/938], Loss: 0.4591\n",
            "Epoch [4], Batch [600/938], Loss: 0.4370\n",
            "Epoch [4], Batch [700/938], Loss: 0.4270\n",
            "Epoch [4], Batch [800/938], Loss: 0.4520\n",
            "Epoch [4], Batch [900/938], Loss: 0.4350\n",
            "Epoch [5], Batch [100/938], Loss: 0.4175\n",
            "Epoch [5], Batch [200/938], Loss: 0.4228\n",
            "Epoch [5], Batch [300/938], Loss: 0.4089\n",
            "Epoch [5], Batch [400/938], Loss: 0.4220\n",
            "Epoch [5], Batch [500/938], Loss: 0.4196\n",
            "Epoch [5], Batch [600/938], Loss: 0.4062\n",
            "Epoch [5], Batch [700/938], Loss: 0.4326\n",
            "Epoch [5], Batch [800/938], Loss: 0.4180\n",
            "Epoch [5], Batch [900/938], Loss: 0.4038\n",
            "Epoch [6], Batch [100/938], Loss: 0.3986\n",
            "Epoch [6], Batch [200/938], Loss: 0.3762\n",
            "Epoch [6], Batch [300/938], Loss: 0.4118\n",
            "Epoch [6], Batch [400/938], Loss: 0.4007\n",
            "Epoch [6], Batch [500/938], Loss: 0.4028\n",
            "Epoch [6], Batch [600/938], Loss: 0.3945\n",
            "Epoch [6], Batch [700/938], Loss: 0.4035\n",
            "Epoch [6], Batch [800/938], Loss: 0.3791\n",
            "Epoch [6], Batch [900/938], Loss: 0.3839\n",
            "Epoch [7], Batch [100/938], Loss: 0.3720\n",
            "Epoch [7], Batch [200/938], Loss: 0.3813\n",
            "Epoch [7], Batch [300/938], Loss: 0.3846\n",
            "Epoch [7], Batch [400/938], Loss: 0.3597\n",
            "Epoch [7], Batch [500/938], Loss: 0.3720\n",
            "Epoch [7], Batch [600/938], Loss: 0.3934\n",
            "Epoch [7], Batch [700/938], Loss: 0.3745\n",
            "Epoch [7], Batch [800/938], Loss: 0.3804\n",
            "Epoch [7], Batch [900/938], Loss: 0.3681\n",
            "Epoch [8], Batch [100/938], Loss: 0.3671\n",
            "Epoch [8], Batch [200/938], Loss: 0.3539\n",
            "Epoch [8], Batch [300/938], Loss: 0.3610\n",
            "Epoch [8], Batch [400/938], Loss: 0.3779\n",
            "Epoch [8], Batch [500/938], Loss: 0.3717\n",
            "Epoch [8], Batch [600/938], Loss: 0.3615\n",
            "Epoch [8], Batch [700/938], Loss: 0.3656\n",
            "Epoch [8], Batch [800/938], Loss: 0.3582\n",
            "Epoch [8], Batch [900/938], Loss: 0.3640\n",
            "Epoch [9], Batch [100/938], Loss: 0.3513\n",
            "Epoch [9], Batch [200/938], Loss: 0.3451\n",
            "Epoch [9], Batch [300/938], Loss: 0.3495\n",
            "Epoch [9], Batch [400/938], Loss: 0.3371\n",
            "Epoch [9], Batch [500/938], Loss: 0.3548\n",
            "Epoch [9], Batch [600/938], Loss: 0.3576\n",
            "Epoch [9], Batch [700/938], Loss: 0.3736\n",
            "Epoch [9], Batch [800/938], Loss: 0.3494\n",
            "Epoch [9], Batch [900/938], Loss: 0.3609\n",
            "Epoch [10], Batch [100/938], Loss: 0.3407\n",
            "Epoch [10], Batch [200/938], Loss: 0.3518\n",
            "Epoch [10], Batch [300/938], Loss: 0.3840\n",
            "Epoch [10], Batch [400/938], Loss: 0.3267\n",
            "Epoch [10], Batch [500/938], Loss: 0.3544\n",
            "Epoch [10], Batch [600/938], Loss: 0.3434\n",
            "Epoch [10], Batch [700/938], Loss: 0.3475\n",
            "Epoch [10], Batch [800/938], Loss: 0.3388\n",
            "Epoch [10], Batch [900/938], Loss: 0.3378\n",
            "Accuracy for Configuration-11: 85.86%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-12\n",
            "Input data:\n",
            "- Num Hidden Layers: 4\n",
            "- Hidden Layer Sizes: [64, 128, 256, 512]\n",
            "- Batch Size: 128\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 2.0985\n",
            "Epoch [1], Batch [200/938], Loss: 1.4503\n",
            "Epoch [1], Batch [300/938], Loss: 1.1066\n",
            "Epoch [1], Batch [400/938], Loss: 0.8991\n",
            "Epoch [1], Batch [500/938], Loss: 0.8033\n",
            "Epoch [1], Batch [600/938], Loss: 0.7465\n",
            "Epoch [1], Batch [700/938], Loss: 0.7180\n",
            "Epoch [1], Batch [800/938], Loss: 0.6476\n",
            "Epoch [1], Batch [900/938], Loss: 0.6451\n",
            "Epoch [2], Batch [100/938], Loss: 0.6078\n",
            "Epoch [2], Batch [200/938], Loss: 0.5889\n",
            "Epoch [2], Batch [300/938], Loss: 0.5620\n",
            "Epoch [2], Batch [400/938], Loss: 0.5469\n",
            "Epoch [2], Batch [500/938], Loss: 0.5509\n",
            "Epoch [2], Batch [600/938], Loss: 0.5322\n",
            "Epoch [2], Batch [700/938], Loss: 0.5314\n",
            "Epoch [2], Batch [800/938], Loss: 0.5230\n",
            "Epoch [2], Batch [900/938], Loss: 0.5374\n",
            "Epoch [3], Batch [100/938], Loss: 0.4798\n",
            "Epoch [3], Batch [200/938], Loss: 0.4819\n",
            "Epoch [3], Batch [300/938], Loss: 0.4787\n",
            "Epoch [3], Batch [400/938], Loss: 0.4903\n",
            "Epoch [3], Batch [500/938], Loss: 0.4682\n",
            "Epoch [3], Batch [600/938], Loss: 0.4493\n",
            "Epoch [3], Batch [700/938], Loss: 0.4696\n",
            "Epoch [3], Batch [800/938], Loss: 0.4527\n",
            "Epoch [3], Batch [900/938], Loss: 0.4341\n",
            "Epoch [4], Batch [100/938], Loss: 0.4199\n",
            "Epoch [4], Batch [200/938], Loss: 0.4237\n",
            "Epoch [4], Batch [300/938], Loss: 0.4201\n",
            "Epoch [4], Batch [400/938], Loss: 0.4244\n",
            "Epoch [4], Batch [500/938], Loss: 0.4242\n",
            "Epoch [4], Batch [600/938], Loss: 0.4124\n",
            "Epoch [4], Batch [700/938], Loss: 0.4222\n",
            "Epoch [4], Batch [800/938], Loss: 0.4142\n",
            "Epoch [4], Batch [900/938], Loss: 0.4287\n",
            "Epoch [5], Batch [100/938], Loss: 0.4081\n",
            "Epoch [5], Batch [200/938], Loss: 0.3972\n",
            "Epoch [5], Batch [300/938], Loss: 0.3943\n",
            "Epoch [5], Batch [400/938], Loss: 0.3853\n",
            "Epoch [5], Batch [500/938], Loss: 0.3884\n",
            "Epoch [5], Batch [600/938], Loss: 0.3784\n",
            "Epoch [5], Batch [700/938], Loss: 0.3862\n",
            "Epoch [5], Batch [800/938], Loss: 0.3868\n",
            "Epoch [5], Batch [900/938], Loss: 0.4063\n",
            "Epoch [6], Batch [100/938], Loss: 0.3833\n",
            "Epoch [6], Batch [200/938], Loss: 0.3809\n",
            "Epoch [6], Batch [300/938], Loss: 0.3755\n",
            "Epoch [6], Batch [400/938], Loss: 0.3681\n",
            "Epoch [6], Batch [500/938], Loss: 0.3644\n",
            "Epoch [6], Batch [600/938], Loss: 0.3746\n",
            "Epoch [6], Batch [700/938], Loss: 0.3862\n",
            "Epoch [6], Batch [800/938], Loss: 0.3676\n",
            "Epoch [6], Batch [900/938], Loss: 0.3449\n",
            "Epoch [7], Batch [100/938], Loss: 0.3495\n",
            "Epoch [7], Batch [200/938], Loss: 0.3367\n",
            "Epoch [7], Batch [300/938], Loss: 0.3633\n",
            "Epoch [7], Batch [400/938], Loss: 0.3705\n",
            "Epoch [7], Batch [500/938], Loss: 0.3579\n",
            "Epoch [7], Batch [600/938], Loss: 0.3758\n",
            "Epoch [7], Batch [700/938], Loss: 0.3511\n",
            "Epoch [7], Batch [800/938], Loss: 0.3727\n",
            "Epoch [7], Batch [900/938], Loss: 0.3560\n",
            "Epoch [8], Batch [100/938], Loss: 0.3456\n",
            "Epoch [8], Batch [200/938], Loss: 0.3265\n",
            "Epoch [8], Batch [300/938], Loss: 0.3383\n",
            "Epoch [8], Batch [400/938], Loss: 0.3350\n",
            "Epoch [8], Batch [500/938], Loss: 0.3374\n",
            "Epoch [8], Batch [600/938], Loss: 0.3500\n",
            "Epoch [8], Batch [700/938], Loss: 0.3432\n",
            "Epoch [8], Batch [800/938], Loss: 0.3472\n",
            "Epoch [8], Batch [900/938], Loss: 0.3580\n",
            "Epoch [9], Batch [100/938], Loss: 0.3502\n",
            "Epoch [9], Batch [200/938], Loss: 0.3195\n",
            "Epoch [9], Batch [300/938], Loss: 0.3237\n",
            "Epoch [9], Batch [400/938], Loss: 0.3357\n",
            "Epoch [9], Batch [500/938], Loss: 0.3361\n",
            "Epoch [9], Batch [600/938], Loss: 0.3214\n",
            "Epoch [9], Batch [700/938], Loss: 0.3254\n",
            "Epoch [9], Batch [800/938], Loss: 0.3388\n",
            "Epoch [9], Batch [900/938], Loss: 0.3329\n",
            "Epoch [10], Batch [100/938], Loss: 0.3235\n",
            "Epoch [10], Batch [200/938], Loss: 0.3196\n",
            "Epoch [10], Batch [300/938], Loss: 0.3271\n",
            "Epoch [10], Batch [400/938], Loss: 0.3284\n",
            "Epoch [10], Batch [500/938], Loss: 0.3012\n",
            "Epoch [10], Batch [600/938], Loss: 0.3495\n",
            "Epoch [10], Batch [700/938], Loss: 0.3300\n",
            "Epoch [10], Batch [800/938], Loss: 0.3290\n",
            "Epoch [10], Batch [900/938], Loss: 0.3170\n",
            "Accuracy for Configuration-12: 85.75%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-13\n",
            "Input data:\n",
            "- Num Hidden Layers: 2\n",
            "- Hidden Layer Sizes: [128, 256]\n",
            "- Batch Size: 256\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.1174\n",
            "Epoch [1], Batch [200/938], Loss: 0.7010\n",
            "Epoch [1], Batch [300/938], Loss: 0.6004\n",
            "Epoch [1], Batch [400/938], Loss: 0.5574\n",
            "Epoch [1], Batch [500/938], Loss: 0.5194\n",
            "Epoch [1], Batch [600/938], Loss: 0.4831\n",
            "Epoch [1], Batch [700/938], Loss: 0.4626\n",
            "Epoch [1], Batch [800/938], Loss: 0.4608\n",
            "Epoch [1], Batch [900/938], Loss: 0.4622\n",
            "Epoch [2], Batch [100/938], Loss: 0.4318\n",
            "Epoch [2], Batch [200/938], Loss: 0.4240\n",
            "Epoch [2], Batch [300/938], Loss: 0.4180\n",
            "Epoch [2], Batch [400/938], Loss: 0.4196\n",
            "Epoch [2], Batch [500/938], Loss: 0.4115\n",
            "Epoch [2], Batch [600/938], Loss: 0.3773\n",
            "Epoch [2], Batch [700/938], Loss: 0.3868\n",
            "Epoch [2], Batch [800/938], Loss: 0.3976\n",
            "Epoch [2], Batch [900/938], Loss: 0.3813\n",
            "Epoch [3], Batch [100/938], Loss: 0.3531\n",
            "Epoch [3], Batch [200/938], Loss: 0.3571\n",
            "Epoch [3], Batch [300/938], Loss: 0.3639\n",
            "Epoch [3], Batch [400/938], Loss: 0.3759\n",
            "Epoch [3], Batch [500/938], Loss: 0.3536\n",
            "Epoch [3], Batch [600/938], Loss: 0.3452\n",
            "Epoch [3], Batch [700/938], Loss: 0.3812\n",
            "Epoch [3], Batch [800/938], Loss: 0.3593\n",
            "Epoch [3], Batch [900/938], Loss: 0.3550\n",
            "Epoch [4], Batch [100/938], Loss: 0.3466\n",
            "Epoch [4], Batch [200/938], Loss: 0.3315\n",
            "Epoch [4], Batch [300/938], Loss: 0.3338\n",
            "Epoch [4], Batch [400/938], Loss: 0.3272\n",
            "Epoch [4], Batch [500/938], Loss: 0.3626\n",
            "Epoch [4], Batch [600/938], Loss: 0.3445\n",
            "Epoch [4], Batch [700/938], Loss: 0.3218\n",
            "Epoch [4], Batch [800/938], Loss: 0.3296\n",
            "Epoch [4], Batch [900/938], Loss: 0.3193\n",
            "Epoch [5], Batch [100/938], Loss: 0.3187\n",
            "Epoch [5], Batch [200/938], Loss: 0.3078\n",
            "Epoch [5], Batch [300/938], Loss: 0.3057\n",
            "Epoch [5], Batch [400/938], Loss: 0.3086\n",
            "Epoch [5], Batch [500/938], Loss: 0.3155\n",
            "Epoch [5], Batch [600/938], Loss: 0.3066\n",
            "Epoch [5], Batch [700/938], Loss: 0.3251\n",
            "Epoch [5], Batch [800/938], Loss: 0.3171\n",
            "Epoch [5], Batch [900/938], Loss: 0.3227\n",
            "Epoch [6], Batch [100/938], Loss: 0.3049\n",
            "Epoch [6], Batch [200/938], Loss: 0.3015\n",
            "Epoch [6], Batch [300/938], Loss: 0.3109\n",
            "Epoch [6], Batch [400/938], Loss: 0.2954\n",
            "Epoch [6], Batch [500/938], Loss: 0.2816\n",
            "Epoch [6], Batch [600/938], Loss: 0.3007\n",
            "Epoch [6], Batch [700/938], Loss: 0.3046\n",
            "Epoch [6], Batch [800/938], Loss: 0.2999\n",
            "Epoch [6], Batch [900/938], Loss: 0.2894\n",
            "Epoch [7], Batch [100/938], Loss: 0.2801\n",
            "Epoch [7], Batch [200/938], Loss: 0.2864\n",
            "Epoch [7], Batch [300/938], Loss: 0.2806\n",
            "Epoch [7], Batch [400/938], Loss: 0.2896\n",
            "Epoch [7], Batch [500/938], Loss: 0.2881\n",
            "Epoch [7], Batch [600/938], Loss: 0.2941\n",
            "Epoch [7], Batch [700/938], Loss: 0.2839\n",
            "Epoch [7], Batch [800/938], Loss: 0.2694\n",
            "Epoch [7], Batch [900/938], Loss: 0.2887\n",
            "Epoch [8], Batch [100/938], Loss: 0.2616\n",
            "Epoch [8], Batch [200/938], Loss: 0.2808\n",
            "Epoch [8], Batch [300/938], Loss: 0.2641\n",
            "Epoch [8], Batch [400/938], Loss: 0.2737\n",
            "Epoch [8], Batch [500/938], Loss: 0.2706\n",
            "Epoch [8], Batch [600/938], Loss: 0.2738\n",
            "Epoch [8], Batch [700/938], Loss: 0.2727\n",
            "Epoch [8], Batch [800/938], Loss: 0.2655\n",
            "Epoch [8], Batch [900/938], Loss: 0.2850\n",
            "Epoch [9], Batch [100/938], Loss: 0.2699\n",
            "Epoch [9], Batch [200/938], Loss: 0.2555\n",
            "Epoch [9], Batch [300/938], Loss: 0.2524\n",
            "Epoch [9], Batch [400/938], Loss: 0.2588\n",
            "Epoch [9], Batch [500/938], Loss: 0.2826\n",
            "Epoch [9], Batch [600/938], Loss: 0.2677\n",
            "Epoch [9], Batch [700/938], Loss: 0.2682\n",
            "Epoch [9], Batch [800/938], Loss: 0.2513\n",
            "Epoch [9], Batch [900/938], Loss: 0.2569\n",
            "Epoch [10], Batch [100/938], Loss: 0.2477\n",
            "Epoch [10], Batch [200/938], Loss: 0.2529\n",
            "Epoch [10], Batch [300/938], Loss: 0.2504\n",
            "Epoch [10], Batch [400/938], Loss: 0.2514\n",
            "Epoch [10], Batch [500/938], Loss: 0.2577\n",
            "Epoch [10], Batch [600/938], Loss: 0.2548\n",
            "Epoch [10], Batch [700/938], Loss: 0.2597\n",
            "Epoch [10], Batch [800/938], Loss: 0.2641\n",
            "Epoch [10], Batch [900/938], Loss: 0.2543\n",
            "Accuracy for Configuration-13: 87.98%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-14\n",
            "Input data:\n",
            "- Num Hidden Layers: 1\n",
            "- Hidden Layer Sizes: [512]\n",
            "- Batch Size: 256\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 0.9049\n",
            "Epoch [1], Batch [200/938], Loss: 0.5633\n",
            "Epoch [1], Batch [300/938], Loss: 0.5360\n",
            "Epoch [1], Batch [400/938], Loss: 0.4821\n",
            "Epoch [1], Batch [500/938], Loss: 0.4677\n",
            "Epoch [1], Batch [600/938], Loss: 0.4433\n",
            "Epoch [1], Batch [700/938], Loss: 0.4329\n",
            "Epoch [1], Batch [800/938], Loss: 0.4341\n",
            "Epoch [1], Batch [900/938], Loss: 0.4404\n",
            "Epoch [2], Batch [100/938], Loss: 0.4044\n",
            "Epoch [2], Batch [200/938], Loss: 0.3959\n",
            "Epoch [2], Batch [300/938], Loss: 0.3911\n",
            "Epoch [2], Batch [400/938], Loss: 0.3940\n",
            "Epoch [2], Batch [500/938], Loss: 0.3810\n",
            "Epoch [2], Batch [600/938], Loss: 0.3732\n",
            "Epoch [2], Batch [700/938], Loss: 0.3651\n",
            "Epoch [2], Batch [800/938], Loss: 0.3689\n",
            "Epoch [2], Batch [900/938], Loss: 0.3760\n",
            "Epoch [3], Batch [100/938], Loss: 0.3614\n",
            "Epoch [3], Batch [200/938], Loss: 0.3295\n",
            "Epoch [3], Batch [300/938], Loss: 0.3367\n",
            "Epoch [3], Batch [400/938], Loss: 0.3515\n",
            "Epoch [3], Batch [500/938], Loss: 0.3372\n",
            "Epoch [3], Batch [600/938], Loss: 0.3464\n",
            "Epoch [3], Batch [700/938], Loss: 0.3505\n",
            "Epoch [3], Batch [800/938], Loss: 0.3344\n",
            "Epoch [3], Batch [900/938], Loss: 0.3425\n",
            "Epoch [4], Batch [100/938], Loss: 0.3186\n",
            "Epoch [4], Batch [200/938], Loss: 0.3269\n",
            "Epoch [4], Batch [300/938], Loss: 0.3400\n",
            "Epoch [4], Batch [400/938], Loss: 0.3098\n",
            "Epoch [4], Batch [500/938], Loss: 0.3076\n",
            "Epoch [4], Batch [600/938], Loss: 0.3170\n",
            "Epoch [4], Batch [700/938], Loss: 0.3064\n",
            "Epoch [4], Batch [800/938], Loss: 0.2937\n",
            "Epoch [4], Batch [900/938], Loss: 0.3015\n",
            "Epoch [5], Batch [100/938], Loss: 0.2859\n",
            "Epoch [5], Batch [200/938], Loss: 0.3064\n",
            "Epoch [5], Batch [300/938], Loss: 0.2926\n",
            "Epoch [5], Batch [400/938], Loss: 0.3013\n",
            "Epoch [5], Batch [500/938], Loss: 0.2871\n",
            "Epoch [5], Batch [600/938], Loss: 0.2876\n",
            "Epoch [5], Batch [700/938], Loss: 0.2887\n",
            "Epoch [5], Batch [800/938], Loss: 0.2951\n",
            "Epoch [5], Batch [900/938], Loss: 0.2777\n",
            "Epoch [6], Batch [100/938], Loss: 0.2689\n",
            "Epoch [6], Batch [200/938], Loss: 0.2694\n",
            "Epoch [6], Batch [300/938], Loss: 0.2861\n",
            "Epoch [6], Batch [400/938], Loss: 0.2758\n",
            "Epoch [6], Batch [500/938], Loss: 0.2651\n",
            "Epoch [6], Batch [600/938], Loss: 0.2800\n",
            "Epoch [6], Batch [700/938], Loss: 0.2710\n",
            "Epoch [6], Batch [800/938], Loss: 0.2732\n",
            "Epoch [6], Batch [900/938], Loss: 0.2681\n",
            "Epoch [7], Batch [100/938], Loss: 0.2482\n",
            "Epoch [7], Batch [200/938], Loss: 0.2512\n",
            "Epoch [7], Batch [300/938], Loss: 0.2685\n",
            "Epoch [7], Batch [400/938], Loss: 0.2493\n",
            "Epoch [7], Batch [500/938], Loss: 0.2582\n",
            "Epoch [7], Batch [600/938], Loss: 0.2472\n",
            "Epoch [7], Batch [700/938], Loss: 0.2647\n",
            "Epoch [7], Batch [800/938], Loss: 0.2587\n",
            "Epoch [7], Batch [900/938], Loss: 0.2630\n",
            "Epoch [8], Batch [100/938], Loss: 0.2436\n",
            "Epoch [8], Batch [200/938], Loss: 0.2557\n",
            "Epoch [8], Batch [300/938], Loss: 0.2406\n",
            "Epoch [8], Batch [400/938], Loss: 0.2420\n",
            "Epoch [8], Batch [500/938], Loss: 0.2393\n",
            "Epoch [8], Batch [600/938], Loss: 0.2270\n",
            "Epoch [8], Batch [700/938], Loss: 0.2439\n",
            "Epoch [8], Batch [800/938], Loss: 0.2420\n",
            "Epoch [8], Batch [900/938], Loss: 0.2492\n",
            "Epoch [9], Batch [100/938], Loss: 0.2287\n",
            "Epoch [9], Batch [200/938], Loss: 0.2327\n",
            "Epoch [9], Batch [300/938], Loss: 0.2237\n",
            "Epoch [9], Batch [400/938], Loss: 0.2337\n",
            "Epoch [9], Batch [500/938], Loss: 0.2351\n",
            "Epoch [9], Batch [600/938], Loss: 0.2377\n",
            "Epoch [9], Batch [700/938], Loss: 0.2214\n",
            "Epoch [9], Batch [800/938], Loss: 0.2227\n",
            "Epoch [9], Batch [900/938], Loss: 0.2322\n",
            "Epoch [10], Batch [100/938], Loss: 0.2222\n",
            "Epoch [10], Batch [200/938], Loss: 0.2146\n",
            "Epoch [10], Batch [300/938], Loss: 0.2092\n",
            "Epoch [10], Batch [400/938], Loss: 0.2101\n",
            "Epoch [10], Batch [500/938], Loss: 0.2204\n",
            "Epoch [10], Batch [600/938], Loss: 0.2180\n",
            "Epoch [10], Batch [700/938], Loss: 0.2207\n",
            "Epoch [10], Batch [800/938], Loss: 0.2155\n",
            "Epoch [10], Batch [900/938], Loss: 0.2259\n",
            "Accuracy for Configuration-14: 88.76%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration - Configuration-15\n",
            "Input data:\n",
            "- Num Hidden Layers: 4\n",
            "- Hidden Layer Sizes: [64, 128, 256, 512]\n",
            "- Batch Size: 32\n",
            "- Learning Rate: 0.001\n",
            "Epoch [1], Batch [100/938], Loss: 1.7930\n",
            "Epoch [1], Batch [200/938], Loss: 1.2091\n",
            "Epoch [1], Batch [300/938], Loss: 0.9837\n",
            "Epoch [1], Batch [400/938], Loss: 0.8913\n",
            "Epoch [1], Batch [500/938], Loss: 0.8403\n",
            "Epoch [1], Batch [600/938], Loss: 0.7832\n",
            "Epoch [1], Batch [700/938], Loss: 0.7585\n",
            "Epoch [1], Batch [800/938], Loss: 0.7373\n",
            "Epoch [1], Batch [900/938], Loss: 0.7106\n",
            "Epoch [2], Batch [100/938], Loss: 0.6760\n",
            "Epoch [2], Batch [200/938], Loss: 0.6636\n",
            "Epoch [2], Batch [300/938], Loss: 0.6523\n",
            "Epoch [2], Batch [400/938], Loss: 0.6318\n",
            "Epoch [2], Batch [500/938], Loss: 0.5913\n",
            "Epoch [2], Batch [600/938], Loss: 0.5952\n",
            "Epoch [2], Batch [700/938], Loss: 0.5934\n",
            "Epoch [2], Batch [800/938], Loss: 0.5423\n",
            "Epoch [2], Batch [900/938], Loss: 0.5716\n",
            "Epoch [3], Batch [100/938], Loss: 0.5265\n",
            "Epoch [3], Batch [200/938], Loss: 0.5365\n",
            "Epoch [3], Batch [300/938], Loss: 0.4981\n",
            "Epoch [3], Batch [400/938], Loss: 0.5082\n",
            "Epoch [3], Batch [500/938], Loss: 0.5014\n",
            "Epoch [3], Batch [600/938], Loss: 0.4851\n",
            "Epoch [3], Batch [700/938], Loss: 0.4727\n",
            "Epoch [3], Batch [800/938], Loss: 0.5020\n",
            "Epoch [3], Batch [900/938], Loss: 0.4678\n",
            "Epoch [4], Batch [100/938], Loss: 0.4546\n",
            "Epoch [4], Batch [200/938], Loss: 0.4521\n",
            "Epoch [4], Batch [300/938], Loss: 0.4709\n",
            "Epoch [4], Batch [400/938], Loss: 0.4540\n",
            "Epoch [4], Batch [500/938], Loss: 0.4519\n",
            "Epoch [4], Batch [600/938], Loss: 0.4408\n",
            "Epoch [4], Batch [700/938], Loss: 0.4276\n",
            "Epoch [4], Batch [800/938], Loss: 0.4379\n",
            "Epoch [4], Batch [900/938], Loss: 0.4503\n",
            "Epoch [5], Batch [100/938], Loss: 0.4118\n",
            "Epoch [5], Batch [200/938], Loss: 0.4238\n",
            "Epoch [5], Batch [300/938], Loss: 0.4201\n",
            "Epoch [5], Batch [400/938], Loss: 0.4370\n",
            "Epoch [5], Batch [500/938], Loss: 0.4159\n",
            "Epoch [5], Batch [600/938], Loss: 0.4079\n",
            "Epoch [5], Batch [700/938], Loss: 0.4368\n",
            "Epoch [5], Batch [800/938], Loss: 0.3858\n",
            "Epoch [5], Batch [900/938], Loss: 0.4068\n",
            "Epoch [6], Batch [100/938], Loss: 0.3867\n",
            "Epoch [6], Batch [200/938], Loss: 0.3888\n",
            "Epoch [6], Batch [300/938], Loss: 0.3994\n",
            "Epoch [6], Batch [400/938], Loss: 0.3907\n",
            "Epoch [6], Batch [500/938], Loss: 0.3976\n",
            "Epoch [6], Batch [600/938], Loss: 0.3961\n",
            "Epoch [6], Batch [700/938], Loss: 0.3929\n",
            "Epoch [6], Batch [800/938], Loss: 0.4058\n",
            "Epoch [6], Batch [900/938], Loss: 0.3732\n",
            "Epoch [7], Batch [100/938], Loss: 0.3770\n",
            "Epoch [7], Batch [200/938], Loss: 0.3781\n",
            "Epoch [7], Batch [300/938], Loss: 0.3878\n",
            "Epoch [7], Batch [400/938], Loss: 0.3759\n",
            "Epoch [7], Batch [500/938], Loss: 0.3683\n",
            "Epoch [7], Batch [600/938], Loss: 0.3661\n",
            "Epoch [7], Batch [700/938], Loss: 0.3700\n",
            "Epoch [7], Batch [800/938], Loss: 0.3688\n",
            "Epoch [7], Batch [900/938], Loss: 0.3813\n",
            "Epoch [8], Batch [100/938], Loss: 0.3676\n",
            "Epoch [8], Batch [200/938], Loss: 0.3545\n",
            "Epoch [8], Batch [300/938], Loss: 0.3466\n",
            "Epoch [8], Batch [400/938], Loss: 0.3459\n",
            "Epoch [8], Batch [500/938], Loss: 0.3693\n",
            "Epoch [8], Batch [600/938], Loss: 0.3473\n",
            "Epoch [8], Batch [700/938], Loss: 0.3551\n",
            "Epoch [8], Batch [800/938], Loss: 0.3738\n",
            "Epoch [8], Batch [900/938], Loss: 0.3643\n",
            "Epoch [9], Batch [100/938], Loss: 0.3499\n",
            "Epoch [9], Batch [200/938], Loss: 0.3561\n",
            "Epoch [9], Batch [300/938], Loss: 0.3385\n",
            "Epoch [9], Batch [400/938], Loss: 0.3502\n",
            "Epoch [9], Batch [500/938], Loss: 0.3524\n",
            "Epoch [9], Batch [600/938], Loss: 0.3421\n",
            "Epoch [9], Batch [700/938], Loss: 0.3563\n",
            "Epoch [9], Batch [800/938], Loss: 0.3404\n",
            "Epoch [9], Batch [900/938], Loss: 0.3443\n",
            "Epoch [10], Batch [100/938], Loss: 0.3210\n",
            "Epoch [10], Batch [200/938], Loss: 0.3362\n",
            "Epoch [10], Batch [300/938], Loss: 0.3491\n",
            "Epoch [10], Batch [400/938], Loss: 0.3256\n",
            "Epoch [10], Batch [500/938], Loss: 0.3420\n",
            "Epoch [10], Batch [600/938], Loss: 0.3201\n",
            "Epoch [10], Batch [700/938], Loss: 0.3205\n",
            "Epoch [10], Batch [800/938], Loss: 0.3411\n",
            "Epoch [10], Batch [900/938], Loss: 0.3595\n",
            "Accuracy for Configuration-15: 85.80%\n",
            "\n",
            "..................................................\n",
            "\n",
            "Configuration       Num Hidden Layers  Hidden Layer Sizes      Batch Size    Learning Rate    Accuracy\n",
            "----------------  -------------------  --------------------  ------------  ---------------  ----------\n",
            "Configuration-1                     3  [32, 64, 128]                   64             0.00        0.86\n",
            "Configuration-2                     3  [64, 128, 256]                 512             0.00        0.84\n",
            "Configuration-3                     3  [128, 256, 512]                 64             0.00        0.88\n",
            "Configuration-4                     2  [64, 128]                       64             0.00        0.88\n",
            "Configuration-5                     2  [32, 64]                        64             0.00        0.86\n",
            "Configuration-6                     2  [128, 256]                      64             0.00        0.88\n",
            "Configuration-7                     2  [256, 512]                      64             0.00        0.88\n",
            "Configuration-8                     1  [64]                            64             0.00        0.87\n",
            "Configuration-9                     1  [128]                           64             0.00        0.88\n",
            "Configuration-10                    1  [256]                           64             0.00        0.88\n",
            "Configuration-11                    3  [32, 64, 128]                  128             0.00        0.86\n",
            "Configuration-12                    4  [64, 128, 256, 512]            128             0.00        0.86\n",
            "Configuration-13                    2  [128, 256]                     256             0.00        0.88\n",
            "Configuration-14                    1  [512]                          256             0.00        0.89\n",
            "Configuration-15                    4  [64, 128, 256, 512]             32             0.00        0.86\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Extract configurations and accuracies for plotting\n",
        "configs = [result[0] for result in FC_results]\n",
        "accuracies = [result[5] * 100 for result in FC_results]\n",
        "\n",
        "# Plotting accuracy vs configurations\n",
        "plt.figure(figsize=(10, 6))\n",
        "bars = plt.bar(range(1, len(configs) + 1), accuracies, color='skyblue')\n",
        "plt.xlabel('Configuration')\n",
        "plt.ylabel('Accuracy (%)')  # Updated ylabel to show percentage\n",
        "plt.title('Accuracy vs Configurations')\n",
        "plt.xticks(range(1, len(configs) + 1), range(1, len(configs) + 1), rotation=45, ha='right')\n",
        "plt.ylim(0, 100)  # Setting y-axis limit from 0 to 100 for percentage\n",
        "plt.grid(axis='y', linestyle='--', alpha=0.7)\n",
        "\n",
        "# Adding text inside the bars\n",
        "for bar, accuracy in zip(bars, accuracies):\n",
        "    plt.text(bar.get_x() + bar.get_width() / 2, bar.get_height() - 5, f'{accuracy:.2f}%',\n",
        "             ha='center', va='bottom', color='black', fontsize=7)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "id": "inCFtkJnoCJn",
        "outputId": "1ffb3b10-7e19-4b7c-81af-768ccc5dafc5"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sorted_results = sorted(FC_results, key=lambda x: x[5], reverse=True)\n",
        "top_5_results = sorted_results[:5]\n",
        "\n",
        "# Extract configurations and accuracies for plotting\n",
        "top_5_configs = [result[0] for result in top_5_results]\n",
        "top_5_accuracies = [result[5] * 100 for result in top_5_results]\n",
        "\n",
        "# Plotting top 5 accuracies\n",
        "plt.figure(figsize=(10, 6))\n",
        "bars = plt.bar(top_5_configs, top_5_accuracies, color='skyblue')\n",
        "plt.xlabel('Configuration')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.title('Top 5 Configurations with Highest Accuracy')\n",
        "plt.xticks(rotation=45, ha='right')\n",
        "plt.ylim(0, 100)  # Setting y-axis limit from 0 to 100 for percentage\n",
        "plt.grid(axis='y', linestyle='--', alpha=0.7)\n",
        "\n",
        "# Adding text inside the bars\n",
        "for bar, accuracy in zip(bars, top_5_accuracies):\n",
        "    plt.text(bar.get_x() + bar.get_width() / 2, bar.get_height() - 5, f'{accuracy:.2f}%',\n",
        "             ha='center', va='bottom', color='black', fontsize=10)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "id": "5hSWWWNApMbe",
        "outputId": "3c3ef2f4-c83b-41cf-fdbb-9133546640f8"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Extract the configuration with the highest accuracy\n",
        "best_config = max(FC_results, key=lambda x: x[5])\n",
        "\n",
        "# Extract configuration parameters\n",
        "config_name = best_config[0]\n",
        "num_layers = best_config[1]\n",
        "layer_sizes = sum(best_config[2])\n",
        "batch_size = best_config[3]\n",
        "learning_rate = best_config[4]\n",
        "accuracy = best_config[5] * 100\n",
        "\n",
        "# Create a bar plot for the configuration parameters\n",
        "labels = ['Num Hidden Layers', 'Hidden Layer Sizes', 'Batch Size', 'Learning Rate']\n",
        "values = [num_layers, layer_sizes, batch_size, learning_rate]\n",
        "\n",
        "# Plot each parameter separately\n",
        "fig, axs = plt.subplots(2, 2, figsize=(12, 12))\n",
        "fig.suptitle(f\"{config_name} (Accuracy: {accuracy:.2f}%)\", fontsize=16)\n",
        "\n",
        "for i, label in enumerate(labels):\n",
        "    ax = axs[i//2][i%2]\n",
        "    ax.bar(label, values[i], color='skyblue')\n",
        "    ax.set_title(label)\n",
        "    ax.set_ylabel('Values')\n",
        "\n",
        "    # Annotate the bars with their corresponding values inside the bars\n",
        "    ax.text(0.5, values[i], str(values[i]), color='black', va='center', ha='center')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "c9TKxwLDprHQ",
        "outputId": "608faf77-7cd0-4c01-954e-5d0cea23ea8b"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x1200 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "name": "Welcome to Colaboratory",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}