ML3_Mukaddas_Khusniddinova.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOj0tmTnBZXmW3jSc0Pey4Q",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Khusniddinova/ml/blob/main/ML3_Mukaddas_Khusniddinova.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Student: Mukaddas Khusniddinova\n",
        "\n",
        "ID: 2110146\n",
        "\n",
        "Group: JSE-02\n",
        "\n",
        "\n",
        "\n",
        "**task2**"
      ],
      "metadata": {
        "id": "Klk9K44EQHgS"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "qIBmSCyLOH_D"
      },
      "outputs": [],
      "source": [
        "import os.path\n",
        "def check_fashion_mnist_dataset_exists(path_data='data/'):\n",
        "    flag_train_data = os.path.isfile(path_data + 'fashion-mnist/train_data.pt')\n",
        "    flag_train_label = os.path.isfile(path_data + 'fashion-mnist/train_label.pt')\n",
        "    flag_test_data = os.path.isfile(path_data + 'fashion-mnist/test_data.pt')\n",
        "    flag_test_label = os.path.isfile(path_data + 'fashion-mnist/test_label.pt')\n",
        "    if flag_train_data==False or flag_train_label==False or flag_test_data==False or flag_test_label==False:\n",
        "        print('FASHION-MNIST dataset missing - downloading...')\n",
        "        import torchvision\n",
        "        import torchvision.transforms as transforms\n",
        "        trainset = torchvision.datasets.FashionMNIST(root=path_data + 'fashion-mnist/temp', train=True,\n",
        "                                                download=True, transform=transforms.ToTensor())\n",
        "        testset = torchvision.datasets.FashionMNIST(root=path_data + 'fashion-mnist/temp', train=False,\n",
        "                                               download=True, transform=transforms.ToTensor())\n",
        "        train_data=torch.Tensor(60000,28,28)\n",
        "        train_label=torch.LongTensor(60000)\n",
        "        for idx , example in enumerate(trainset):\n",
        "            train_data[idx]=example[0].squeeze()\n",
        "            train_label[idx]=example[1]\n",
        "        torch.save(train_data,path_data + 'fashion-mnist/train_data.pt')\n",
        "        torch.save(train_label,path_data + 'fashion-mnist/train_label.pt')\n",
        "        test_data=torch.Tensor(10000,28,28)\n",
        "        test_label=torch.LongTensor(10000)\n",
        "        for idx , example in enumerate(testset):\n",
        "            test_data[idx]=example[0].squeeze()\n",
        "            test_label[idx]=example[1]\n",
        "        torch.save(test_data,path_data + 'fashion-mnist/test_data.pt')\n",
        "        torch.save(test_label,path_data + 'fashion-mnist/test_label.pt')\n",
        "    return path_data"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def show(X):\n",
        "    if X.dim() == 3 and X.size(0) == 3:\n",
        "        plt.imshow( np.transpose(  X.numpy() , (1, 2, 0))  )\n",
        "        plt.show()\n",
        "    elif X.dim() == 2:\n",
        "        plt.imshow(   X.numpy() , cmap='gray'  )\n",
        "        plt.show()\n",
        "    else:\n",
        "        print('WRONG TENSOR SIZE')"
      ],
      "metadata": {
        "id": "sQOwecuyOZRY"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torchvision\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "# Check and download the dataset\n",
        "data_path = check_fashion_mnist_dataset_exists()\n",
        "\n",
        "# Load Fashion MNIST dataset\n",
        "train_dataset = torchvision.datasets.FashionMNIST(root=data_path + 'fashion-mnist/', train=True, download=True)\n",
        "\n",
        "import torchvision.transforms as transforms\n",
        "\n",
        "# Transformation to convert PIL images to tensors\n",
        "transform = transforms.ToTensor()\n",
        "\n",
        "# Display some sample images\n",
        "num_samples_to_display = 11\n",
        "\n",
        "plt.figure(figsize=(10, 5))\n",
        "for i in range(num_samples_to_display):\n",
        "    image, label = train_dataset[i]\n",
        "    image_tensor = transform(image)  # Convert PIL image to tensor\n",
        "    plt.subplot(1, num_samples_to_display, i + 1)\n",
        "\n",
        "    plt.imshow(image)\n",
        "    plt.title(f\"Label: {label}\")\n",
        "    plt.axis('off')\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 844
        },
        "id": "Zf9erRguOeN6",
        "outputId": "7e576388-2278-454b-88da-44a924c92b50"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "FASHION-MNIST dataset missing - downloading...\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 26421880/26421880 [00:03<00:00, 8144202.52it/s] \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/temp/FashionMNIST/raw/train-images-idx3-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 29515/29515 [00:00<00:00, 135930.04it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/temp/FashionMNIST/raw/train-labels-idx1-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 4422102/4422102 [00:01<00:00, 2567222.75it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/temp/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 5148/5148 [00:00<00:00, 15830115.10it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/temp/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to data/fashion-mnist/temp/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to data/fashion-mnist/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 26421880/26421880 [00:03<00:00, 8354053.39it/s] \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/FashionMNIST/raw/train-images-idx3-ubyte.gz to data/fashion-mnist/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to data/fashion-mnist/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 29515/29515 [00:00<00:00, 136798.84it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/FashionMNIST/raw/train-labels-idx1-ubyte.gz to data/fashion-mnist/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to data/fashion-mnist/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 4422102/4422102 [00:01<00:00, 2629805.21it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to data/fashion-mnist/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to data/fashion-mnist/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 5148/5148 [00:00<00:00, 13444755.29it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting data/fashion-mnist/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to data/fashion-mnist/FashionMNIST/raw\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 11 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "class TwoLayerfnn(nn.Module):\n",
        "    def __init__(self, input_size, hidden_size, output_size):\n",
        "        super(TwoLayerfnn, self).__init__()\n",
        "        self.fc1 = nn.Linear(input_size, hidden_size)\n",
        "        self.fc2 = nn.Linear(hidden_size, output_size)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = x.view(x.size(0), -1)  # Flatten the input\n",
        "        x = F.relu(self.fc1(x))\n",
        "        x = self.fc2(x)\n",
        "        return F.log_softmax(x, dim=1)# Applying log softmax to get probabilities"
      ],
      "metadata": {
        "id": "-ErQhCJaOuMV"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "input_size = 28 * 28  # input images are 28x28\n",
        "hidden_size = 128  # Number of neurons in the hidden layer\n",
        "output_size = 10  # Fashion MNIST has 10 classes\n",
        "\n",
        "# Create an instance of the TwoLayerFNN\n",
        "model = TwoLayerfnn(input_size, hidden_size, output_size)\n",
        "\n",
        "# Count the number of parameters\n",
        "num_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n",
        "print(f\"There are {num_params} parameters in this TwoLayerFNN.\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4Bs2srxXO1i3",
        "outputId": "92ba96b1-abee-4aab-ced7-028664d9fefa"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "There are 101770 parameters in this TwoLayerFNN.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nmtumoYxO4fM",
        "outputId": "7d1944f2-1fa4-48d0-f417-a1eb0123c6bf"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "TwoLayerfnn(\n",
            "  (fc1): Linear(in_features=784, out_features=128, bias=True)\n",
            "  (fc2): Linear(in_features=128, out_features=10, bias=True)\n",
            ")\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "\n",
        "# Get a random image and its label from the training set\n",
        "random_idx = random.randint(0, len(train_dataset) - 1)\n",
        "image, label = train_dataset[random_idx]\n",
        "image_tensor = transform(image).unsqueeze(0)  # Convert PIL image to tensor and add a batch dimension\n",
        "\n",
        "# Forward pass through the model\n",
        "output_probs = model(image_tensor)\n",
        "\n",
        "# Convert output probabilities to numpy array\n",
        "output_probs_np = output_probs.exp().detach().numpy()\n",
        "\n",
        "# Display the output probabilities\n",
        "print(f\"Output Probabilities: {output_probs_np}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pTniRVneO7rh",
        "outputId": "2505ff57-295d-4a1e-8f0d-4e75dc06e3ee"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Output Probabilities: [[0.11228405 0.10636965 0.1289504  0.10121153 0.07655849 0.0622616\n",
            "  0.09159265 0.10573228 0.09742921 0.11761021]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch.optim as optim\n",
        "from torch.utils.data import DataLoader\n",
        "\n",
        "transform = transforms.Compose([\n",
        "    transforms.ToTensor(),  # Convert PIL image to tensor\n",
        "\n",
        "])\n",
        "\n",
        "# Update train_dataset with the new transformation\n",
        "train_dataset = torchvision.datasets.FashionMNIST(root='./data', train=True, download=True, transform=transform)\n",
        "\n",
        "# Define DataLoader for the training set\n",
        "batch_size = 64\n",
        "train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
        "\n",
        "# Define criterion (loss function) and optimizer\n",
        "criterion = nn.CrossEntropyLoss()  # Cross-entropy loss for classification\n",
        "optimizer = optim.SGD(model.parameters(), lr=0.01)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "096XGwu2O_TW",
        "outputId": "96eca377-991a-4c73-a913-2d1b08153525"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to ./data/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 26421880/26421880 [00:13<00:00, 1888387.18it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/FashionMNIST/raw/train-images-idx3-ubyte.gz to ./data/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 29515/29515 [00:00<00:00, 138180.59it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/FashionMNIST/raw/train-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to ./data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 4422102/4422102 [00:01<00:00, 2554813.94it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to ./data/FashionMNIST/raw\n",
            "\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 5148/5148 [00:00<00:00, 16186114.69it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "num_epochs = 7\n",
        "for epoch in range(num_epochs):\n",
        "    running_loss = 0.0\n",
        "    for i, data in enumerate(train_loader, 0):\n",
        "        inputs, labels = data\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        outputs = model(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        running_loss += loss.item()\n",
        "        if i % 300 == 299:  # Print every 100 mini-batches\n",
        "            print(f\"Epoch [{epoch + 1}/{num_epochs}], \"\n",
        "                  f\"Step [{i + 1}/{len(train_loader)}], \"\n",
        "                  f\"Loss: {running_loss / 100:.4f}\")\n",
        "            running_loss = 0.0\n",
        "\n",
        "print('Finished Training')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bE0W-7SxPFN5",
        "outputId": "80920eba-bfde-49e7-d772-404f6ac2db11"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/7], Step [300/938], Loss: 5.1140\n",
            "Epoch [1/7], Step [600/938], Loss: 3.0109\n",
            "Epoch [1/7], Step [900/938], Loss: 2.4415\n",
            "Epoch [2/7], Step [300/938], Loss: 2.1356\n",
            "Epoch [2/7], Step [600/938], Loss: 2.0033\n",
            "Epoch [2/7], Step [900/938], Loss: 1.8836\n",
            "Epoch [3/7], Step [300/938], Loss: 1.7805\n",
            "Epoch [3/7], Step [600/938], Loss: 1.7298\n",
            "Epoch [3/7], Step [900/938], Loss: 1.6606\n",
            "Epoch [4/7], Step [300/938], Loss: 1.6185\n",
            "Epoch [4/7], Step [600/938], Loss: 1.5783\n",
            "Epoch [4/7], Step [900/938], Loss: 1.5678\n",
            "Epoch [5/7], Step [300/938], Loss: 1.5132\n",
            "Epoch [5/7], Step [600/938], Loss: 1.5135\n",
            "Epoch [5/7], Step [900/938], Loss: 1.4774\n",
            "Epoch [6/7], Step [300/938], Loss: 1.4447\n",
            "Epoch [6/7], Step [600/938], Loss: 1.4526\n",
            "Epoch [6/7], Step [900/938], Loss: 1.4417\n",
            "Epoch [7/7], Step [300/938], Loss: 1.4331\n",
            "Epoch [7/7], Step [600/938], Loss: 1.3783\n",
            "Epoch [7/7], Step [900/938], Loss: 1.3869\n",
            "Finished Training\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_transform = transforms.Compose([\n",
        "    transforms.ToTensor(),\n",
        "])\n",
        "test_dataset = torchvision.datasets.FashionMNIST(root='./data', train=False, download=True, transform=test_transform)\n",
        "random_idx = random.randint(0, len(test_dataset) - 1)\n",
        "test_img, test_label = test_dataset[random_idx]"
      ],
      "metadata": {
        "id": "jRexVcDgPPQN"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from torchvision import transforms as T\n",
        "test_image_tensor =test_img\n",
        "\n",
        "\n",
        "model.eval()\n",
        "with torch.no_grad():\n",
        "    output_probs = model(test_image_tensor)\n",
        "\n",
        "# Convert output probabilities to numpy array\n",
        "output_probs_np = output_probs.exp().detach().numpy()\n",
        "\n",
        "test_image_pil = T.ToPILImage()(test_img)\n",
        "# Display the test image and its predicted probabilities\n",
        "plt.imshow(test_image_pil, cmap='gray')\n",
        "plt.title(f\"True Label: {test_label}, Predicted Probabilities: {output_probs_np.squeeze()}\")\n",
        "plt.axis('off')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "9iuJo_00PqKn",
        "outputId": "8aaadb0f-c5ae-4697-a9b1-cf34afe51e22"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
        "\n",
        "# List to store predictions and ground truth labels\n",
        "predicted_labels = []\n",
        "true_labels = []\n",
        "\n",
        "\n",
        "\n",
        "# Iterate through the test set and collect predictions\n",
        "for i in range(len(test_dataset)):\n",
        "    test_image, test_label = test_dataset[i]\n",
        "    test_image_tensor = test_img.unsqueeze(0)\n",
        "    # Forward pass through the trained model\n",
        "    with torch.no_grad():\n",
        "        output_probs = model(image_tensor)\n",
        "\n",
        "    # Get predicted label\n",
        "    predicted_label = output_probs.argmax(dim=1).item()\n",
        "\n",
        "    # Append predictions and true labels\n",
        "    predicted_labels.append(predicted_label)\n",
        "    true_labels.append(test_label)"
      ],
      "metadata": {
        "id": "F_HvSFNmPwNK"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate accuracy\n",
        "accuracy = accuracy_score(true_labels, predicted_labels)\n",
        "\n",
        "# Calculate precision, recall, and F1 score\n",
        "precision = precision_score(true_labels, predicted_labels, average='weighted')\n",
        "recall = recall_score(true_labels, predicted_labels, average='weighted')\n",
        "f1 = f1_score(true_labels, predicted_labels, average='weighted')\n",
        "\n",
        "# Print the metrics\n",
        "print(f\"Accuracy: {accuracy:.4f}\")\n",
        "print(f\"Precision: {precision:.4f}\")\n",
        "print(f\"Recall: {recall:.4f}\")\n",
        "print(f\"F1 Score: {f1:.4f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4xwtqGUjP9li",
        "outputId": "77e650ad-98d8-4abd-9750-2481d79cac06"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.1000\n",
            "Precision: 0.0100\n",
            "Recall: 0.1000\n",
            "F1 Score: 0.0182\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**task1**"
      ],
      "metadata": {
        "id": "PbdJeqwPQBMK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Read data in Python. Then explore it; display datatypes, and check for missing values. Do we need to use scaling?\n",
        "\n",
        "import pandas as pd\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.tree import export_text\n",
        "from sklearn import tree\n",
        "\n",
        "train_data = pd.read_csv('satellite_training.csv')\n",
        "X_train = train_data.iloc[:, :-1]\n",
        "y_train = train_data['Class']\n",
        "\n",
        "satellite_df = pd.DataFrame(X_train, columns=[f'F{i}' for i in range(X_train.shape[1])])\n",
        "satellite_df['Class'] = y_train\n",
        "\n",
        "satellite_df.head()\n",
        "\n",
        "print(satellite_df.dtypes)\n",
        "\n",
        "print(\"\\nMissing values in the dataset:\")\n",
        "print(satellite_df.isnull().sum())\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p48vZjdaQDa7",
        "outputId": "533c095a-80e6-49bc-cfb5-6c02bcf83328"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "F0       int64\n",
            "F1       int64\n",
            "F2       int64\n",
            "F3       int64\n",
            "F4       int64\n",
            "F5       int64\n",
            "F6       int64\n",
            "F7       int64\n",
            "F8       int64\n",
            "F9       int64\n",
            "F10      int64\n",
            "F11      int64\n",
            "F12      int64\n",
            "F13      int64\n",
            "F14      int64\n",
            "F15      int64\n",
            "F16      int64\n",
            "F17      int64\n",
            "F18      int64\n",
            "F19      int64\n",
            "F20      int64\n",
            "F21      int64\n",
            "F22      int64\n",
            "F23      int64\n",
            "F24      int64\n",
            "F25      int64\n",
            "F26      int64\n",
            "F27      int64\n",
            "F28      int64\n",
            "F29      int64\n",
            "F30      int64\n",
            "F31      int64\n",
            "F32      int64\n",
            "F33      int64\n",
            "F34      int64\n",
            "F35      int64\n",
            "Class    int64\n",
            "dtype: object\n",
            "\n",
            "Missing values in the dataset:\n",
            "F0       0\n",
            "F1       0\n",
            "F2       0\n",
            "F3       0\n",
            "F4       0\n",
            "F5       0\n",
            "F6       0\n",
            "F7       0\n",
            "F8       0\n",
            "F9       0\n",
            "F10      0\n",
            "F11      0\n",
            "F12      0\n",
            "F13      0\n",
            "F14      0\n",
            "F15      0\n",
            "F16      0\n",
            "F17      0\n",
            "F18      0\n",
            "F19      0\n",
            "F20      0\n",
            "F21      0\n",
            "F22      0\n",
            "F23      0\n",
            "F24      0\n",
            "F25      0\n",
            "F26      0\n",
            "F27      0\n",
            "F28      0\n",
            "F29      0\n",
            "F30      0\n",
            "F31      0\n",
            "F32      0\n",
            "F33      0\n",
            "F34      0\n",
            "F35      0\n",
            "Class    0\n",
            "dtype: int64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        " We have no missing values\n",
        " No need to use scaling as the values don't differ too much"
      ],
      "metadata": {
        "id": "hiN42AQyRfQ5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#building a basic decision tree classifier and fit it to the training data.\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report\n",
        "\n",
        "test_data = pd.read_csv('satellite_test.csv')\n",
        "\n",
        "X_test = test_data.drop('Class', axis=1)\n",
        "y_test = test_data['Class']\n",
        "\n",
        "decision_tree = DecisionTreeClassifier()\n",
        "decision_tree = decision_tree.fit(X_train, y_train)\n",
        "\n",
        "y_pred = decision_tree.predict(X_test)\n",
        "\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'\\nAccuracy: {accuracy:.4f}')\n",
        "\n",
        "precision = precision_score(y_test, y_pred, average='weighted')\n",
        "print(f'Precision: {precision:.4f}')\n",
        "\n",
        "recall = recall_score(y_test, y_pred, average='weighted')\n",
        "print(f'Recall: {recall:.4f}')\n",
        "\n",
        "f1 = f1_score(y_test, y_pred, average='weighted')\n",
        "print(f'F1 Score: {f1:.4f}\\n\\n')\n",
        "\n",
        "tree_text = export_text(decision_tree, feature_names=list(X_train.columns))\n",
        "print(tree_text)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PQzX_0ezRckp",
        "outputId": "c948c648-ac54-4109-f2db-16ad658ec386"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Accuracy: 0.8470\n",
            "Precision: 0.8487\n",
            "Recall: 0.8470\n",
            "F1 Score: 0.8478\n",
            "\n",
            "\n",
            "|--- F16 <= 79.50\n",
            "|   |--- F19 <= 73.50\n",
            "|   |   |--- F17 <= 61.50\n",
            "|   |   |   |--- F14 <= 109.00\n",
            "|   |   |   |   |--- F17 <= 59.50\n",
            "|   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |--- F17 >  59.50\n",
            "|   |   |   |   |   |--- F0 <= 51.00\n",
            "|   |   |   |   |   |   |--- F14 <= 76.00\n",
            "|   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |--- F14 >  76.00\n",
            "|   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |--- F0 >  51.00\n",
            "|   |   |   |   |   |   |--- F19 <= 57.50\n",
            "|   |   |   |   |   |   |   |--- F23 <= 55.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |--- F23 >  55.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |--- F19 >  57.50\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |--- F14 >  109.00\n",
            "|   |   |   |   |--- class: 2\n",
            "|   |   |--- F17 >  61.50\n",
            "|   |   |   |--- F15 <= 65.50\n",
            "|   |   |   |   |--- F33 <= 62.50\n",
            "|   |   |   |   |   |--- F2 <= 73.50\n",
            "|   |   |   |   |   |   |--- F31 <= 68.50\n",
            "|   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |--- F31 >  68.50\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |--- F2 >  73.50\n",
            "|   |   |   |   |   |   |--- F17 <= 71.50\n",
            "|   |   |   |   |   |   |   |--- F32 <= 54.50\n",
            "|   |   |   |   |   |   |   |   |--- F31 <= 53.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F31 >  53.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F32 >  54.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F17 >  71.50\n",
            "|   |   |   |   |   |   |   |--- F0 <= 69.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F0 >  69.00\n",
            "|   |   |   |   |   |   |   |   |--- F17 <= 76.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F17 >  76.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |--- F33 >  62.50\n",
            "|   |   |   |   |   |--- F19 <= 68.50\n",
            "|   |   |   |   |   |   |--- F7 <= 50.00\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F7 >  50.00\n",
            "|   |   |   |   |   |   |   |--- F16 <= 57.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |--- F16 >  57.50\n",
            "|   |   |   |   |   |   |   |   |--- F23 <= 48.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F23 >  48.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F0 <= 52.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F0 >  52.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F24 <= 83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F24 >  83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |--- F19 >  68.50\n",
            "|   |   |   |   |   |   |--- F11 <= 91.50\n",
            "|   |   |   |   |   |   |   |--- F5 <= 74.00\n",
            "|   |   |   |   |   |   |   |   |--- F15 <= 62.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F15 >  62.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F16 <= 56.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |   |--- F16 >  56.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F26 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F26 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F5 >  74.00\n",
            "|   |   |   |   |   |   |   |   |--- F1 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F31 <= 58.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F31 >  58.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F1 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F11 >  91.50\n",
            "|   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |--- F15 >  65.50\n",
            "|   |   |   |   |--- F16 <= 59.50\n",
            "|   |   |   |   |   |--- F20 <= 57.50\n",
            "|   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |--- F20 >  57.50\n",
            "|   |   |   |   |   |   |--- F4 <= 55.50\n",
            "|   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |--- F4 >  55.50\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |--- F16 >  59.50\n",
            "|   |   |   |   |   |--- F17 <= 82.00\n",
            "|   |   |   |   |   |   |--- F17 <= 72.50\n",
            "|   |   |   |   |   |   |   |--- F26 <= 79.50\n",
            "|   |   |   |   |   |   |   |   |--- F21 <= 63.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F21 >  63.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F19 <= 69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F19 >  69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |--- F26 >  79.50\n",
            "|   |   |   |   |   |   |   |   |--- F6 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F33 <= 80.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F30 <= 78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F30 >  78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F33 >  80.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 <= 81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 >  81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F6 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F2 <= 102.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F2 >  102.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F10 <= 90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F10 >  90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |--- F17 >  72.50\n",
            "|   |   |   |   |   |   |   |--- F31 <= 78.50\n",
            "|   |   |   |   |   |   |   |   |--- F35 <= 49.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F35 >  49.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 <= 55.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 <= 69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 >  69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 >  55.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F23 <= 73.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F23 >  73.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |--- F31 >  78.50\n",
            "|   |   |   |   |   |   |   |   |--- F22 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F20 <= 63.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F20 >  63.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F22 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |--- F17 >  82.00\n",
            "|   |   |   |   |   |   |--- F22 <= 86.50\n",
            "|   |   |   |   |   |   |   |--- F11 <= 47.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F11 >  47.00\n",
            "|   |   |   |   |   |   |   |   |--- F30 <= 71.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F30 >  71.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F0 <= 78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F25 <= 85.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F25 >  85.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F0 >  78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 <= 84.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 >  84.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F22 >  86.50\n",
            "|   |   |   |   |   |   |   |--- F2 <= 85.50\n",
            "|   |   |   |   |   |   |   |   |--- F13 <= 88.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F13 >  88.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F21 <= 80.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F21 >  80.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F2 >  85.50\n",
            "|   |   |   |   |   |   |   |   |--- F14 <= 99.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 <= 83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F26 <= 89.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F26 >  89.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 >  83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F13 <= 81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F13 >  81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6\n",
            "|   |   |   |   |   |   |   |   |--- F14 >  99.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F12 <= 79.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F12 >  79.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |--- F19 >  73.50\n",
            "|   |   |--- F17 <= 65.00\n",
            "|   |   |   |--- F19 <= 90.50\n",
            "|   |   |   |   |--- F27 <= 100.50\n",
            "|   |   |   |   |   |--- F7 <= 97.00\n",
            "|   |   |   |   |   |   |--- F28 <= 49.50\n",
            "|   |   |   |   |   |   |   |--- F27 <= 88.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |--- F27 >  88.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F28 >  49.50\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |--- F7 >  97.00\n",
            "|   |   |   |   |   |   |--- F9 <= 52.00\n",
            "|   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |--- F9 >  52.00\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |--- F27 >  100.50\n",
            "|   |   |   |   |   |--- class: 2\n",
            "|   |   |   |--- F19 >  90.50\n",
            "|   |   |   |   |--- F17 <= 59.50\n",
            "|   |   |   |   |   |--- F17 <= 55.50\n",
            "|   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |--- F17 >  55.50\n",
            "|   |   |   |   |   |   |--- F26 <= 99.00\n",
            "|   |   |   |   |   |   |   |--- F35 <= 108.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |--- F35 >  108.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |--- F26 >  99.00\n",
            "|   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |--- F17 >  59.50\n",
            "|   |   |   |   |   |--- F9 <= 51.50\n",
            "|   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |--- F9 >  51.50\n",
            "|   |   |   |   |   |   |--- F29 <= 50.00\n",
            "|   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |--- F29 >  50.00\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |--- F17 >  65.00\n",
            "|   |   |   |--- F20 <= 73.00\n",
            "|   |   |   |   |--- F9 <= 66.50\n",
            "|   |   |   |   |   |--- F9 <= 51.50\n",
            "|   |   |   |   |   |   |--- F31 <= 86.50\n",
            "|   |   |   |   |   |   |   |--- F26 <= 84.50\n",
            "|   |   |   |   |   |   |   |   |--- F14 <= 102.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F14 >  102.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |--- F26 >  84.50\n",
            "|   |   |   |   |   |   |   |   |--- F7 <= 121.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |   |--- F7 >  121.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F31 >  86.50\n",
            "|   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |--- F9 >  51.50\n",
            "|   |   |   |   |   |   |--- F20 <= 54.50\n",
            "|   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |--- F20 >  54.50\n",
            "|   |   |   |   |   |   |   |--- F1 <= 76.00\n",
            "|   |   |   |   |   |   |   |   |--- F6 <= 72.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F6 >  72.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |--- F1 >  76.00\n",
            "|   |   |   |   |   |   |   |   |--- F2 <= 94.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 <= 73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F12 <= 69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F12 >  69.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 >  73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F2 >  94.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 <= 73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F15 <= 92.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F15 >  92.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 >  73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |--- F9 >  66.50\n",
            "|   |   |   |   |   |--- F24 <= 75.50\n",
            "|   |   |   |   |   |   |--- F25 <= 65.00\n",
            "|   |   |   |   |   |   |   |--- F35 <= 100.00\n",
            "|   |   |   |   |   |   |   |   |--- F12 <= 54.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 <= 91.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 >  91.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F12 >  54.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F3 <= 97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F3 >  97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |--- F35 >  100.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |--- F25 >  65.00\n",
            "|   |   |   |   |   |   |   |--- F11 <= 72.50\n",
            "|   |   |   |   |   |   |   |   |--- F4 <= 64.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 <= 76.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 >  76.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F4 >  64.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F17 <= 78.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F17 >  78.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F35 <= 70.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F35 >  70.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |--- F11 >  72.50\n",
            "|   |   |   |   |   |   |   |   |--- F1 <= 65.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 <= 90.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 <= 70.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 >  70.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 >  90.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F1 >  65.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F29 <= 65.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F7 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F7 >  85.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F29 >  65.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F32 <= 71.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F32 >  71.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5\n",
            "|   |   |   |   |   |--- F24 >  75.50\n",
            "|   |   |   |   |   |   |--- F27 <= 88.50\n",
            "|   |   |   |   |   |   |   |--- F17 <= 80.00\n",
            "|   |   |   |   |   |   |   |   |--- F24 <= 81.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F24 >  81.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F1 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F1 >  85.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F35 <= 95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F35 >  95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |--- F17 >  80.00\n",
            "|   |   |   |   |   |   |   |   |--- F31 <= 80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F22 <= 99.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F22 >  99.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |   |--- F31 >  80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F17 <= 90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F17 >  90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F27 <= 84.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F27 >  84.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F27 >  88.50\n",
            "|   |   |   |   |   |   |   |--- F9 <= 76.00\n",
            "|   |   |   |   |   |   |   |   |--- F8 <= 65.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |   |--- F8 >  65.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F9 >  76.00\n",
            "|   |   |   |   |   |   |   |   |--- F24 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F25 <= 95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 <= 63.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 >  63.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |--- F25 >  95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |--- F24 >  85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F7 <= 93.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |--- F7 >  93.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |--- F20 >  73.00\n",
            "|   |   |   |   |--- F19 <= 89.00\n",
            "|   |   |   |   |   |--- F17 <= 96.50\n",
            "|   |   |   |   |   |   |--- F3 <= 70.50\n",
            "|   |   |   |   |   |   |   |--- F34 <= 115.50\n",
            "|   |   |   |   |   |   |   |   |--- F7 <= 80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F23 <= 71.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F14 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F14 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F23 >  71.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F7 >  80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F34 >  115.50\n",
            "|   |   |   |   |   |   |   |   |--- F19 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |--- F19 >  85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F3 >  70.50\n",
            "|   |   |   |   |   |   |   |--- F10 <= 103.00\n",
            "|   |   |   |   |   |   |   |   |--- F25 <= 84.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F24 <= 72.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 >  87.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |--- F24 >  72.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F25 >  84.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F34 <= 111.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F9 <= 61.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F9 >  61.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6\n",
            "|   |   |   |   |   |   |   |   |   |--- F34 >  111.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F18 <= 101.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F18 >  101.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F10 >  103.00\n",
            "|   |   |   |   |   |   |   |   |--- F12 <= 52.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 2\n",
            "|   |   |   |   |   |   |   |   |--- F12 >  52.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 <= 77.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 <= 77.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F20 >  77.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 >  77.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 <= 97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F5 >  97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |--- F17 >  96.50\n",
            "|   |   |   |   |   |   |--- F28 <= 77.00\n",
            "|   |   |   |   |   |   |   |--- F29 <= 101.50\n",
            "|   |   |   |   |   |   |   |   |--- F35 <= 84.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F19 <= 77.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F9 <= 90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F9 >  90.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F19 >  77.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F35 >  84.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 <= 95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 >  95.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F29 >  101.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |--- F28 >  77.00\n",
            "|   |   |   |   |   |   |   |--- F0 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |--- F7 <= 73.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F7 >  73.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F0 >  85.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |--- F19 >  89.00\n",
            "|   |   |   |   |   |--- F17 <= 101.00\n",
            "|   |   |   |   |   |   |--- F15 <= 90.50\n",
            "|   |   |   |   |   |   |   |--- F22 <= 116.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |--- F22 >  116.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F15 >  90.50\n",
            "|   |   |   |   |   |   |   |--- F29 <= 100.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |   |--- F29 >  100.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |--- F17 >  101.00\n",
            "|   |   |   |   |   |   |--- F12 <= 82.00\n",
            "|   |   |   |   |   |   |   |--- F7 <= 82.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F7 >  82.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |--- F12 >  82.00\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|--- F16 >  79.50\n",
            "|   |--- F17 <= 96.50\n",
            "|   |   |--- F10 <= 100.50\n",
            "|   |   |   |--- F30 <= 107.00\n",
            "|   |   |   |   |--- F29 <= 87.50\n",
            "|   |   |   |   |   |--- F14 <= 97.00\n",
            "|   |   |   |   |   |   |--- F17 <= 90.00\n",
            "|   |   |   |   |   |   |   |--- F9 <= 86.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F9 >  86.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F17 >  90.00\n",
            "|   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F14 >  97.00\n",
            "|   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |--- F29 >  87.50\n",
            "|   |   |   |   |   |--- F34 <= 107.00\n",
            "|   |   |   |   |   |   |--- F18 <= 101.50\n",
            "|   |   |   |   |   |   |   |--- F22 <= 85.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F22 >  85.50\n",
            "|   |   |   |   |   |   |   |   |--- F1 <= 84.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F1 >  84.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F2 <= 108.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F34 <= 84.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F34 >  84.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F2 >  108.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F18 >  101.50\n",
            "|   |   |   |   |   |   |   |--- F8 <= 83.00\n",
            "|   |   |   |   |   |   |   |   |--- F24 <= 80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 <= 81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F28 >  81.00\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F24 >  80.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F8 >  83.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F34 >  107.00\n",
            "|   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |--- F30 >  107.00\n",
            "|   |   |   |   |--- F10 <= 89.50\n",
            "|   |   |   |   |   |--- F31 <= 84.50\n",
            "|   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |--- F31 >  84.50\n",
            "|   |   |   |   |   |   |--- F5 <= 77.00\n",
            "|   |   |   |   |   |   |   |--- class: 5\n",
            "|   |   |   |   |   |   |--- F5 >  77.00\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |--- F10 >  89.50\n",
            "|   |   |   |   |   |--- class: 3\n",
            "|   |   |--- F10 >  100.50\n",
            "|   |   |   |--- F26 <= 94.50\n",
            "|   |   |   |   |--- F25 <= 80.00\n",
            "|   |   |   |   |   |--- F24 <= 67.50\n",
            "|   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F24 >  67.50\n",
            "|   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |--- F25 >  80.00\n",
            "|   |   |   |   |   |--- class: 7\n",
            "|   |   |   |--- F26 >  94.50\n",
            "|   |   |   |   |--- F7 <= 77.00\n",
            "|   |   |   |   |   |--- F22 <= 101.50\n",
            "|   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F22 >  101.50\n",
            "|   |   |   |   |   |   |--- F11 <= 77.50\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F11 >  77.50\n",
            "|   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |--- F7 >  77.00\n",
            "|   |   |   |   |   |--- F35 <= 75.50\n",
            "|   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |--- F35 >  75.50\n",
            "|   |   |   |   |   |   |--- F20 <= 79.00\n",
            "|   |   |   |   |   |   |   |--- F8 <= 82.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |--- F8 >  82.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F20 >  79.00\n",
            "|   |   |   |   |   |   |   |--- F4 <= 90.50\n",
            "|   |   |   |   |   |   |   |   |--- F14 <= 93.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F14 >  93.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 <= 78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F18 <= 101.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F18 >  101.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 >  78.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F4 >  90.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |--- F17 >  96.50\n",
            "|   |   |--- F32 <= 85.50\n",
            "|   |   |   |--- F15 <= 98.50\n",
            "|   |   |   |   |--- F2 <= 96.50\n",
            "|   |   |   |   |   |--- F22 <= 104.50\n",
            "|   |   |   |   |   |   |--- F15 <= 80.50\n",
            "|   |   |   |   |   |   |   |--- F11 <= 81.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |--- F11 >  81.50\n",
            "|   |   |   |   |   |   |   |   |--- F25 <= 92.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |--- F25 >  92.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F15 >  80.50\n",
            "|   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F22 >  104.50\n",
            "|   |   |   |   |   |   |--- F26 <= 100.50\n",
            "|   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |--- F26 >  100.50\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |--- F2 >  96.50\n",
            "|   |   |   |   |   |--- F11 <= 89.50\n",
            "|   |   |   |   |   |   |--- F1 <= 103.50\n",
            "|   |   |   |   |   |   |   |--- F11 <= 84.00\n",
            "|   |   |   |   |   |   |   |   |--- F26 <= 101.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 <= 96.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F1 <= 98.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F1 >  98.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F9 >  96.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 <= 99.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 >  99.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F26 >  101.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F20 <= 83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F25 <= 97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F25 >  97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |--- F20 >  83.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 <= 98.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F29 >  98.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |--- F11 >  84.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F1 >  103.50\n",
            "|   |   |   |   |   |   |   |--- F0 <= 87.50\n",
            "|   |   |   |   |   |   |   |   |--- F4 <= 79.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |--- F4 >  79.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F1 <= 106.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F10 <= 104.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F10 >  104.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3\n",
            "|   |   |   |   |   |   |   |   |   |--- F1 >  106.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F0 >  87.50\n",
            "|   |   |   |   |   |   |   |   |--- F29 <= 83.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |   |   |   |   |--- F29 >  83.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F16 <= 84.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F17 <= 104.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F17 >  104.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |--- F16 >  84.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F23 <= 73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F23 >  73.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5\n",
            "|   |   |   |   |   |--- F11 >  89.50\n",
            "|   |   |   |   |   |   |--- F28 <= 70.50\n",
            "|   |   |   |   |   |   |   |--- F26 <= 93.00\n",
            "|   |   |   |   |   |   |   |   |--- F22 <= 103.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |--- F22 >  103.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F26 >  93.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |--- F28 >  70.50\n",
            "|   |   |   |   |   |   |   |--- F3 <= 101.50\n",
            "|   |   |   |   |   |   |   |   |--- F19 <= 103.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 <= 103.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F3 <= 97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F3 >  97.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |--- F11 >  103.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 <= 91.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 >  91.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |   |--- F19 >  103.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |   |   |--- F3 >  101.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |--- F15 >  98.50\n",
            "|   |   |   |   |--- F28 <= 85.00\n",
            "|   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |--- F28 >  85.00\n",
            "|   |   |   |   |   |--- F0 <= 79.00\n",
            "|   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |   |--- F0 >  79.00\n",
            "|   |   |   |   |   |   |--- class: 3\n",
            "|   |   |--- F32 >  85.50\n",
            "|   |   |   |--- F16 <= 86.50\n",
            "|   |   |   |   |--- F30 <= 92.50\n",
            "|   |   |   |   |   |--- class: 7\n",
            "|   |   |   |   |--- F30 >  92.50\n",
            "|   |   |   |   |   |--- F1 <= 102.50\n",
            "|   |   |   |   |   |   |--- F2 <= 107.00\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F2 >  107.00\n",
            "|   |   |   |   |   |   |   |--- F35 <= 84.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F35 >  84.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |--- F1 >  102.50\n",
            "|   |   |   |   |   |   |--- F10 <= 105.50\n",
            "|   |   |   |   |   |   |   |--- F34 <= 111.50\n",
            "|   |   |   |   |   |   |   |   |--- F25 <= 102.50\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |--- F25 >  102.50\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 <= 91.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F27 >  91.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F34 >  111.50\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F10 >  105.50\n",
            "|   |   |   |   |   |   |   |--- F6 <= 103.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |--- F6 >  103.00\n",
            "|   |   |   |   |   |   |   |   |--- F34 <= 103.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F34 >  103.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 <= 118.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 <= 99.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F21 >  99.50\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 >  118.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F0 <= 86.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |   |   |--- F0 >  86.00\n",
            "|   |   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |--- F16 >  86.50\n",
            "|   |   |   |   |--- F1 <= 87.50\n",
            "|   |   |   |   |   |--- F32 <= 90.50\n",
            "|   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |--- F32 >  90.50\n",
            "|   |   |   |   |   |   |--- class: 1\n",
            "|   |   |   |   |--- F1 >  87.50\n",
            "|   |   |   |   |   |--- F15 <= 80.00\n",
            "|   |   |   |   |   |   |--- F14 <= 106.50\n",
            "|   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |--- F14 >  106.50\n",
            "|   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |--- F15 >  80.00\n",
            "|   |   |   |   |   |   |--- F23 <= 79.50\n",
            "|   |   |   |   |   |   |   |--- F7 <= 90.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F7 >  90.00\n",
            "|   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |--- F23 >  79.50\n",
            "|   |   |   |   |   |   |   |--- F12 <= 83.50\n",
            "|   |   |   |   |   |   |   |   |--- F35 <= 85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |--- F35 >  85.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |--- F12 >  83.50\n",
            "|   |   |   |   |   |   |   |   |--- F29 <= 119.00\n",
            "|   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "|   |   |   |   |   |   |   |   |--- F29 >  119.00\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 <= 119.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 4\n",
            "|   |   |   |   |   |   |   |   |   |--- F30 >  119.50\n",
            "|   |   |   |   |   |   |   |   |   |   |--- class: 3\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Question 3**\n",
        "\n",
        "Define a set of parameters to tune, you can use the scikit-learn documentation of the DecisionTreeClassifier class. Explain briefly what do they stand for.\n",
        "\n",
        "Paramenters:\n",
        "\n",
        "*criterion* ('gini', 'entropy' or “log_loss”; default=”gini”): The function to measure the quality of a split. 'gini' uses the Gini impurity, and 'entropy' uses information gain, “log_loss” and “entropy” both for the Shannon information gain.\n",
        "\n",
        "*splitter* (default value: 'best' or 'random'): The strategy used to choose the split at each node. 'best' selects the best split based on the specified criterion, while 'random' selects a random split.\n",
        "\n",
        "*max_depth* (default value: None): The maximum depth of the tree. If None, nodes are expanded until they contain less than min_samples_split samples or until all leaves contain less than min_samples_leaf samples.\n",
        "\n",
        "*min_samples_split* (default value: 2): The minimum number of samples required to split an internal node. If an internal node has fewer samples than min_samples_split, it won't be split.\n",
        "\n",
        "*max_features* (default value: None): The number of features to consider when looking for the best split. None - it considers all features; int, float or “auto”, “sqrt”, “log2”.\n",
        "\n",
        "<i>min_samples_leaf</i> (default: 1): The minimum number of samples required to be at a leaf node, a smaller min_samples_leaf leads to a more complex tree.\n",
        "\n",
        "<i>random_state</i> (default: None): Controls the randomness of the estimator. Setting random_state ensures reproducibility; int, RandomState instance.\n",
        "\n",
        "We can create a dictionary containing these parameters and their respective values, then use this dictionary in combination with cross-validation techniques to search for the best combination of hyperparameters. This process is also known as hyperparameter tuning."
      ],
      "metadata": {
        "id": "b9oyZ4K_RquN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "param_grid = {\n",
        "    'criterion': ['gini', 'entropy'],\n",
        "    'splitter': ['best', 'random'],\n",
        "    'max_depth': [None, 10, 20],\n",
        "    'min_samples_split': [2, 5, 10],\n",
        "    'max_features': [None, 'sqrt', 'log2'],\n",
        "    'min_samples_leaf': [1, 2, 4],\n",
        "    'random_state': [42],\n",
        "}"
      ],
      "metadata": {
        "id": "DAEelnO3SDqo"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Question 4\n",
        "# Run run a grid search algorithm to find the best combination of the hyper-parameters.\n",
        "\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.metrics import make_scorer, accuracy_score\n",
        "\n",
        "decision_tree = DecisionTreeClassifier()\n",
        "\n",
        "grid_search = GridSearchCV(decision_tree, param_grid, cv=5, scoring=make_scorer(accuracy_score), n_jobs=-1)\n",
        "\n",
        "grid_search.fit(X_train, y_train)\n",
        "\n",
        "print(\"Best Hyperparameters:\", grid_search.best_params_)\n",
        "print(\"Best Accuracy Score:\", grid_search.best_score_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qzyLFFc9SLi_",
        "outputId": "78cf07cd-0cbe-4e40-b89d-3007def81f50"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best Hyperparameters: {'criterion': 'entropy', 'max_depth': 10, 'max_features': 'sqrt', 'min_samples_leaf': 4, 'min_samples_split': 10, 'random_state': 42, 'splitter': 'best'}\n",
            "Best Accuracy Score: 0.8202931228861331\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Question 5\n",
        "# Choose the best three combinations and run each 10 times. Average over the accuracies over the runs. Did the best score change?\n",
        "\n",
        "import numpy as np\n",
        "\n",
        "best_params = grid_search.best_params_\n",
        "best_score = grid_search.best_score_\n",
        "\n",
        "top_three_params = grid_search.cv_results_['params'][:3]\n",
        "\n",
        "for params in top_three_params:\n",
        "    accuracies = []\n",
        "    for _ in range(10):\n",
        "        decision_tree = DecisionTreeClassifier(**params)\n",
        "        decision_tree.fit(X_train, y_train)\n",
        "\n",
        "        y_pred = decision_tree.predict(X_test)\n",
        "\n",
        "        accuracy = accuracy_score(y_test, y_pred)\n",
        "        accuracies.append(accuracy)\n",
        "\n",
        "    avg_accuracy = np.mean(accuracies)\n",
        "    print(f\"\\nHyperparameters: {params}\")\n",
        "    print(f\"Average Accuracy over 10 runs: {avg_accuracy:.4f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "myMNO2HzSPc7",
        "outputId": "cd3635b5-6c42-4bc0-d614-20ee9dfa0f5e"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Hyperparameters: {'criterion': 'gini', 'max_depth': None, 'max_features': None, 'min_samples_leaf': 1, 'min_samples_split': 2, 'random_state': 42, 'splitter': 'best'}\n",
            "Average Accuracy over 10 runs: 0.8490\n",
            "\n",
            "Hyperparameters: {'criterion': 'gini', 'max_depth': None, 'max_features': None, 'min_samples_leaf': 1, 'min_samples_split': 2, 'random_state': 42, 'splitter': 'random'}\n",
            "Average Accuracy over 10 runs: 0.8215\n",
            "\n",
            "Hyperparameters: {'criterion': 'gini', 'max_depth': None, 'max_features': None, 'min_samples_leaf': 1, 'min_samples_split': 5, 'random_state': 42, 'splitter': 'best'}\n",
            "Average Accuracy over 10 runs: 0.8545\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Best score changed, the accuracy with new paramenters became higher and reached 85 percents (as we can see in the previous case it was 82%)\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "YyVPyhy3SWGl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**task4**"
      ],
      "metadata": {
        "id": "jZWBw9bkSkxf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Question 1\n",
        "\n",
        "Testing AND function for different input combinations:\n",
        "\n",
        "1. $ x_1 = 0, x_2 = 0: $<br>$\n",
        "v = 1*0 + 1*0 = 0$<br>$\n",
        "v < 2, f(v) = 0 $<br><br>\n",
        "\n",
        "2. $ x_1 = 0, x_2 = 1: $<br>$\n",
        "v = 1*0 + 1*1 = 1$<br>$\n",
        "v < 2, f(v) = 0 $<br><br>\n",
        "\n",
        "3. $ x_1 = 1, x_2 = 0: $<br>$\n",
        "v = 1*1 + 1*0 = 1$<br>$\n",
        "v < 2, f(v) = 0 $<br><br>\n",
        "\n",
        "4. $ x_1 = 1, x_2 = 1: $<br>$\n",
        "v = 1*1 + 1*1 = 2$<br>$\n",
        "v = 2, f(v) = 0 $<br><br>\n",
        "\n",
        "We should have $f(v) = 1$ if $v >= 2$, in order to make the expression suitable for AND, otherwise, with this condition we won't get correct neural AND function."
      ],
      "metadata": {
        "id": "U0rdTqEFSnz-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Question 2\n",
        "\n",
        "One possible solution is to decrease the threshold value till 1.\n",
        "\n",
        "$f(v) = 1$ if $v >= 1$\n",
        "\n",
        "Testing AND function for different input combinations:\n",
        "\n",
        "1. $ x_1 = 0, x_2 = 0: $<br>$\n",
        "v = 1*0 + 1*0 = 0$<br>$\n",
        "v < 1, f(v) = 0 $<br><br>\n",
        "\n",
        "2. $ x_1 = 0, x_2 = 1: $<br>$\n",
        "v = 1*0 + 1*1 = 1$<br>$\n",
        "v = 1, f(v) = 1 $<br><br>\n",
        "\n",
        "3. $ x_1 = 1, x_2 = 0: $<br>$\n",
        "v = 1*1 + 1*0 = 1$<br>$\n",
        "v = 1, f(v) = 1 $<br><br>\n",
        "\n",
        "4. $ x_1 = 1, x_2 = 1: $<br>$\n",
        "v = 1*1 + 1*1 = 2$<br>$\n",
        "v >= 1, f(v) = 1 $<br><br>\n",
        "\n"
      ],
      "metadata": {
        "id": "wsLyd7jESsEw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Question 3\n",
        "\n",
        "It is not possible to implement the XOR neither by a single unit nor by a single–layer feed–forward network. The solution was found using a feed–forward network with a hidden layer. The XOR network uses two hidden nodes and one output node."
      ],
      "metadata": {
        "id": "afewuUsWSv--"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**task3**"
      ],
      "metadata": {
        "id": "MQBC2z_IS9Eh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "class CustomCNN(nn.Module):\n",
        "    def __init__(self):\n",
        "        super(CustomCNN, self).__init__()\n",
        "        self.conv1 = nn.Conv2d(in_channels=1, out_channels=32, kernel_size=3, stride=1, padding=1)\n",
        "        self.conv2 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=3, stride=1, padding=1)\n",
        "        self.conv3 = nn.Conv2d(in_channels=64, out_channels=128, kernel_size=3, stride=1, padding=1)\n",
        "        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)\n",
        "        self.fc1 = nn.Linear(128 * 3 * 3, 512)\n",
        "        self.fc2 = nn.Linear(512, 10)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.pool(F.relu(self.conv1(x)))\n",
        "        x = self.pool(F.relu(self.conv2(x)))\n",
        "        x = self.pool(F.relu(self.conv3(x)))\n",
        "        x = x.view(-1,128 * 3 * 3)  # Flatten the tensor for the fully connected layers\n",
        "        x = nn.functional.relu(self.fc1(x))\n",
        "        x = self.fc2(x)\n",
        "        return nn.functional.log_softmax(x, dim=1)\n"
      ],
      "metadata": {
        "id": "VBG1Pw1tSdYH"
      },
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create an instance of the CustomCNN model\n",
        "model = CustomCNN()\n",
        "\n",
        "# Count the number of parameters\n",
        "num_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n",
        "print(f\"There are {num_params} parameters in this CustomCNN.\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CbmGgn6CTGrO",
        "outputId": "61204815-f7fe-4421-a5a7-93404c2482e8"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "There are 688138 parameters in this CustomCNN.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "ckFPCydkgMSR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "This CNN model has:\n",
        "- convolutional layers: 3 convolutional blocks (conv1, conv2, conv3)\n",
        "- linear layers: 2 linear layers (fc1, fc2)\n",
        "\n",
        "\n",
        "__input shape__\n",
        "\n",
        "For the Fashion MNIST dataset, the images are grayscale, meaning they have a single channel (1 color channel), and each image has a resolution of 28x28 pixels.\n",
        "\n",
        "Therefore, the input shape for the CNN model when working with Fashion MNIST images is `(1, 28, 28)`. This signifies a single channel with an image resolution of 28x28 pixels.\n",
        "\n"
      ],
      "metadata": {
        "id": "rnq16T2ATLoC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import torch.optim as optim\n",
        "\n",
        "\n",
        "# Load Fashion MNIST dataset\n",
        "transform = transforms.Compose([transforms.ToTensor()])\n",
        "trainset = torchvision.datasets.FashionMNIST(root='./data', train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=64, shuffle=True)\n",
        "\n",
        "# Define the criterion (loss function) and optimizer\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
        "\n",
        "# Training loop\n",
        "epochs = 6\n",
        "for epoch in range(epochs):\n",
        "    running_loss = 0.0\n",
        "    for i, data in enumerate(trainloader, 0):\n",
        "        inputs, labels = data\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # Forward, backward, and optimize\n",
        "        outputs = model(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        running_loss += loss.item()\n",
        "        if i % 300 == 299:  # Print every 100 mini-batches\n",
        "            print(f\"Epoch {epoch + 1}, Batch {i + 1}: Loss {running_loss / 100:.3f}\")\n",
        "            running_loss = 0.0\n",
        "\n",
        "print(\"Finished Training\")\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CBMkTR5dYcL6",
        "outputId": "3373a87a-6aff-4981-85b6-0a01363683b8"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1, Batch 300: Loss 0.605\n",
            "Epoch 1, Batch 600: Loss 0.630\n",
            "Epoch 1, Batch 900: Loss 0.630\n",
            "Epoch 2, Batch 300: Loss 0.532\n",
            "Epoch 2, Batch 600: Loss 0.536\n",
            "Epoch 2, Batch 900: Loss 0.531\n",
            "Epoch 3, Batch 300: Loss 0.447\n",
            "Epoch 3, Batch 600: Loss 0.469\n",
            "Epoch 3, Batch 900: Loss 0.482\n",
            "Epoch 4, Batch 300: Loss 0.398\n",
            "Epoch 4, Batch 600: Loss 0.396\n",
            "Epoch 4, Batch 900: Loss 0.433\n",
            "Epoch 5, Batch 300: Loss 0.346\n",
            "Epoch 5, Batch 600: Loss 0.370\n",
            "Epoch 5, Batch 900: Loss 0.347\n",
            "Epoch 6, Batch 300: Loss 0.265\n",
            "Epoch 6, Batch 600: Loss 0.310\n",
            "Epoch 6, Batch 900: Loss 0.331\n",
            "Finished Training\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "\n",
        "# Load Fashion MNIST test dataset\n",
        "testset = torchvision.datasets.FashionMNIST(root='./data', train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=4, shuffle=True)\n",
        "\n",
        "# Function to display a single image from the dataset\n",
        "def imshow(img):\n",
        "    img = img / 2 + 0.5  # Unnormalize\n",
        "    npimg = img.numpy()\n",
        "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
        "    plt.show()\n",
        "\n",
        "# Display a random image from the test set\n",
        "dataiter = iter(testloader)\n",
        "images, labels = dataiter.next()\n",
        "imshow(torchvision.utils.make_grid(images))\n",
        "\n",
        "# Make a prediction for the displayed image\n",
        "outputs = model(images)\n",
        "_, predicted = torch.max(outputs, 1)\n",
        "\n",
        "# Display predicted probabilities\n",
        "print('Predicted Probabilities:', torch.nn.functional.softmax(outputs, dim=1))\n",
        "print('Predicted Classes:', predicted)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 245
        },
        "id": "b-km8zjlcjZm",
        "outputId": "4899e5a7-5817-4bb8-9cae-aad75b4f1adb"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "error",
          "ename": "AttributeError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-49-36875066b8fd>\u001b[0m in \u001b[0;36m<cell line: 18>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     16\u001b[0m \u001b[0;31m# Display a random image from the test set\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0mdataiter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0miter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtestloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 18\u001b[0;31m \u001b[0mimages\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdataiter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     19\u001b[0m \u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtorchvision\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmake_grid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimages\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mAttributeError\u001b[0m: '_SingleProcessDataLoaderIter' object has no attribute 'next'"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
        "\n",
        "\n",
        "\n",
        "model.eval()\n",
        "predicted_labels = []\n",
        "true_labels = []\n",
        "\n",
        "# Iterate through the test set and make predictions\n",
        "with torch.no_grad():\n",
        "    for data in testloader:\n",
        "        images, labels = data\n",
        "        outputs = model(images)\n",
        "        _, predicted = torch.max(outputs, 1)\n",
        "        predicted_labels.extend(predicted.tolist())\n",
        "        true_labels.extend(labels.tolist())\n",
        "\n",
        "# Calculate evaluation metrics\n",
        "accuracy = accuracy_score(true_labels, predicted_labels)\n",
        "precision = precision_score(true_labels, predicted_labels, average='weighted')\n",
        "recall = recall_score(true_labels, predicted_labels, average='weighted')\n",
        "f1 = f1_score(true_labels, predicted_labels, average='weighted')\n",
        "\n",
        "print(f\"Accuracy: {accuracy:.4f}\")\n",
        "print(f\"Precision: {precision:.4f}\")\n",
        "print(f\"Recall: {recall:.4f}\")\n",
        "print(f\"F1 Score: {f1:.4f}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l8iBTDByTM1m",
        "outputId": "cfaf2a30-dbfa-4d80-a43c-2c2b9aaa2d06"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9156\n",
            "Precision: 0.9173\n",
            "Recall: 0.9156\n",
            "F1 Score: 0.9159\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "as we saw above CNN model give more accurate result than FNN model.\n",
        "CNN model outperformed the FNN model on the Fashion MNIST dataset, especially in terms of accuracy and robustness in handling image classification tasks."
      ],
      "metadata": {
        "id": "urbQUgdliIwx"
      }
    }
  ]
}