SentinelLandUseClassificationKeras.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "VGG.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "_6aL-KEHQcd1",
        "colab_type": "code",
        "outputId": "e6311c3d-1ecc-49ee-bdb9-b5c29b3c088a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        }
      },
      "cell_type": "code",
      "source": [
        "# Load the Drive helper and mount\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3Aietf%3Awg%3Aoauth%3A2.0%3Aoob&scope=email%20https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdocs.test%20https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdrive%20https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdrive.photos.readonly%20https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fpeopleapi.readonly&response_type=code\n",
            "\n",
            "Enter your authorization code:\n",
            "··········\n",
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "tS794aSHBTAQ",
        "colab_type": "code",
        "outputId": "086c59ef-14dd-4df0-96ab-80485cc72c4d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "!ls 'drive/My Drive/PhD/Data/Sat/train'"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "AnnualCrop  HerbaceousVegetation  Industrial  PermanentCrop  River\n",
            "Forest\t    Highway\t\t  Pasture     Residential    SeaLake\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "FNgItgoWQt1h",
        "colab_type": "code",
        "outputId": "f6723f06-ad56-44cf-81c0-f48d01f18c33",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import keras\n",
        "from keras import backend as K\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Activation\n",
        "from keras.layers import Dense, Flatten\n",
        "from keras.optimizers import Adam\n",
        "from keras.metrics import categorical_crossentropy\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.layers.normalization import BatchNormalization\n",
        "from keras.layers.convolutional import *\n",
        "from matplotlib import pyplot as plt\n",
        "from sklearn.metrics import confusion_matrix, precision_score, accuracy_score\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "import itertools\n",
        "import matplotlib.pyplot as plt\n",
        "import PIL\n",
        "from PIL import Image\n",
        "from skimage import transform\n",
        "%matplotlib inline\n",
        "import math"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "Yz5m9BkOEP6Z",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from keras.applications.vgg16 import preprocess_input"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "s--tPiwdQwtj",
        "colab_type": "code",
        "outputId": "3a1e21aa-b5a0-4477-fe66-03cfdf5d011b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        }
      },
      "cell_type": "code",
      "source": [
        "vgg16_model = keras.applications.vgg16.VGG16()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n",
            "Downloading data from https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg16_weights_tf_dim_ordering_tf_kernels.h5\n",
            "553467904/553467096 [==============================] - 37s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "d7Jy_5EkQ6Jw",
        "colab_type": "code",
        "outputId": "dccd2009-929e-48cc-a6ee-a68a490e0c09",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 918
        }
      },
      "cell_type": "code",
      "source": [
        "vgg16_model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_1 (InputLayer)         (None, 224, 224, 3)       0         \n",
            "_________________________________________________________________\n",
            "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "fc1 (Dense)                  (None, 4096)              102764544 \n",
            "_________________________________________________________________\n",
            "fc2 (Dense)                  (None, 4096)              16781312  \n",
            "_________________________________________________________________\n",
            "predictions (Dense)          (None, 1000)              4097000   \n",
            "=================================================================\n",
            "Total params: 138,357,544\n",
            "Trainable params: 138,357,544\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Cj57cVf_RB_W",
        "colab_type": "code",
        "outputId": "a996a12b-fe74-48fe-8bf1-eed5c6aa2cac",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "type(vgg16_model)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "keras.engine.training.Model"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "metadata": {
        "id": "HWdJGzA-SOTj",
        "colab_type": "code",
        "outputId": "622a5b93-0ef9-49be-ab2e-213b84a013e9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 918
        }
      },
      "cell_type": "code",
      "source": [
        "vgg16_model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_1 (InputLayer)         (None, 224, 224, 3)       0         \n",
            "_________________________________________________________________\n",
            "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "fc1 (Dense)                  (None, 4096)              102764544 \n",
            "_________________________________________________________________\n",
            "fc2 (Dense)                  (None, 4096)              16781312  \n",
            "_________________________________________________________________\n",
            "predictions (Dense)          (None, 1000)              4097000   \n",
            "=================================================================\n",
            "Total params: 138,357,544\n",
            "Trainable params: 138,357,544\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "gQEHlgMpSJwQ",
        "colab_type": "code",
        "outputId": "788c881d-d35a-4676-f302-8851df4a5155",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "vgg16_model.layers.pop()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.layers.core.Dense at 0x7ffaa6d3c128>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "metadata": {
        "id": "fCJU7LuiREvM",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = Sequential()\n",
        "for layer in vgg16_model.layers:\n",
        "  layer.trainable = False\n",
        "  model.add(layer)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZIHhD1BeRUmW",
        "colab_type": "code",
        "outputId": "4d8db892-1b9e-47b6-c5ad-47d2bf4744f7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "type(model)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "keras.engine.sequential.Sequential"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "metadata": {
        "id": "a_u5A6HRRVtH",
        "colab_type": "code",
        "outputId": "64c4b87a-3f4a-461f-d73c-75999b22092c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 850
        }
      },
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "fc1 (Dense)                  (None, 4096)              102764544 \n",
            "_________________________________________________________________\n",
            "fc2 (Dense)                  (None, 4096)              16781312  \n",
            "=================================================================\n",
            "Total params: 134,260,544\n",
            "Trainable params: 0\n",
            "Non-trainable params: 134,260,544\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "qRd4vVBvRb9R",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.add(Dense(10, activation=\"softmax\"))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "rRZLe6doRwcD",
        "colab_type": "code",
        "outputId": "b2d72f64-61ed-4926-e18f-e9deaa0d1869",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "fc1 (Dense)                  (None, 4096)              102764544 \n",
            "_________________________________________________________________\n",
            "fc2 (Dense)                  (None, 4096)              16781312  \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 10)                40970     \n",
            "=================================================================\n",
            "Total params: 134,301,514\n",
            "Trainable params: 40,970\n",
            "Non-trainable params: 134,260,544\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "sCnrlf3lCOCk",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.compile(optimizer='rmsprop',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "X5-_7IW2Re8w",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "train_data_dir = 'drive/My Drive/PhD/Data/Sat/train'\n",
        "valid_data_dir = 'drive/My Drive/PhD/Data/Sat/valid'\n",
        "test_data_dir = 'drive/My Drive/PhD/Data/Sat/test'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "xFUc7J8nRg8c",
        "colab_type": "code",
        "outputId": "08d71f19-abc4-4b9b-8607-1a1724df6b2b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "img_width = 224\n",
        "img_height = 224\n",
        "batch_size = 25\n",
        "# dategen = keras.preprocessing.image.ImageDataGenerator(preprocessing_function=preprocess_input)\n",
        "\n",
        "train_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/train',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size)\n",
        "\n",
        "valid_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/valid',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size)\n",
        "test_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/test',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size,\n",
        "    shuffle=False,)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 18900 images belonging to 10 classes.\n",
            "Found 4050 images belonging to 10 classes.\n",
            "Found 4050 images belonging to 10 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "chBWDJNz_tVp",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# https://github.com/smileservices/keras_utils/blob/master/utils.py\n",
        "# plots images with labels within jupyter notebook\n",
        "def plots(ims, figsize=(12,6), rows=10, interp=False, titles=None):\n",
        "    if type(ims[0]) is np.ndarray:\n",
        "        ims = np.array(ims).astype(np.uint8)\n",
        "        if (ims.shape[-1] != 3):\n",
        "            ims = ims.transpose((0,2,3,1))\n",
        "    f = plt.figure(figsize=figsize)\n",
        "    cols = len(ims)//rows if len(ims) % 2 == 0 else len(ims)//rows + 1\n",
        "    for i in range(len(ims)):\n",
        "        sp = f.add_subplot(rows, cols, i+1)\n",
        "        sp.axis('Off')\n",
        "        if titles is not None:\n",
        "            sp.set_title(titles[i], fontsize=16)\n",
        "        plt.imshow(ims[i], interpolation=None if interp else 'none')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "c8wa-c_JG51X",
        "colab_type": "code",
        "outputId": "e11fe07f-686b-48c8-8653-6cf81ad1bde9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        }
      },
      "cell_type": "code",
      "source": [
        "imgs, labels = next(test_generator)\n",
        "plots(imgs, titles = labels)\n",
        "np.argmax(labels[0])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/matplotlib/text.py:1191: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
            "  if s != self._text:\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAp4AAAFWCAYAAADAJZ3XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRORKKPEUERERkSvxfwEvVnkGcuyrVgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 864x432 with 25 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "d8-Sb2V-_MAA",
        "colab_type": "code",
        "outputId": "3452d351-fc37-4d25-aa95-1e5fba95c73a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 394
        }
      },
      "cell_type": "code",
      "source": [
        "model.fit_generator(\n",
        "    train_generator,\n",
        "    steps_per_epoch=train_generator.samples // batch_size,\n",
        "    epochs=10,\n",
        "    validation_data=valid_generator,\n",
        "    validation_steps=valid_generator.samples // batch_size)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "756/756 [==============================] - 247s 327ms/step - loss: 0.4686 - acc: 0.8598 - val_loss: 0.4077 - val_acc: 0.8889\n",
            "Epoch 2/10\n",
            "756/756 [==============================] - 247s 326ms/step - loss: 0.3195 - acc: 0.9067 - val_loss: 0.4679 - val_acc: 0.8758\n",
            "Epoch 3/10\n",
            "756/756 [==============================] - 247s 326ms/step - loss: 0.2628 - acc: 0.9254 - val_loss: 0.3061 - val_acc: 0.9183\n",
            "Epoch 4/10\n",
            "756/756 [==============================] - 247s 326ms/step - loss: 0.2431 - acc: 0.9301 - val_loss: 0.2560 - val_acc: 0.9341\n",
            "Epoch 5/10\n",
            "756/756 [==============================] - 246s 326ms/step - loss: 0.2185 - acc: 0.9387 - val_loss: 0.2901 - val_acc: 0.9262\n",
            "Epoch 6/10\n",
            "756/756 [==============================] - 246s 326ms/step - loss: 0.1945 - acc: 0.9420 - val_loss: 0.2791 - val_acc: 0.9304\n",
            "Epoch 7/10\n",
            "756/756 [==============================] - 247s 326ms/step - loss: 0.1784 - acc: 0.9476 - val_loss: 0.3321 - val_acc: 0.9188\n",
            "Epoch 8/10\n",
            "756/756 [==============================] - 247s 326ms/step - loss: 0.1639 - acc: 0.9531 - val_loss: 0.2906 - val_acc: 0.9358\n",
            "Epoch 9/10\n",
            "756/756 [==============================] - 248s 327ms/step - loss: 0.1519 - acc: 0.9570 - val_loss: 0.3539 - val_acc: 0.9198\n",
            "Epoch 10/10\n",
            "756/756 [==============================] - 248s 327ms/step - loss: 0.1472 - acc: 0.9569 - val_loss: 0.2896 - val_acc: 0.9333\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7ff0ac3107f0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "metadata": {
        "id": "drw_1UahB5eT",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "path = \"/content/drive/My Drive/PhD/Data/landuse_model_18_02_2019.h5\" \n",
        "model.save(path)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "luiY4nlgImUn",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#del model\n",
        "#model = keras.models.load_model('/content/drive/My Drive/PhD/Data/vgg16_trained.h5')\n",
        "#model = keras.models.load_model('/content/drive/My Drive/PhD/Data/vgg16_multiclass_trained.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "z-CN2n6cBAt2",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# first you should re-define the network.\n",
        "#model.load_weights('/content/drive/My Drive/PhD/Data/vgg16_multiclass_trained.h5')\n",
        "model.load_weights('/content/drive/My Drive/PhD/Data/landuse_model_18_02_2019.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "YKDFn_MXEIFF",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "predictions = model.predict_generator(test_generator, steps=test_generator.samples // batch_size)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "q6iDljOSzdsN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# serialize predictions\n",
        "import pickle\n",
        "file_pi = open('/content/drive/My Drive/PhD/Data/predictions_18_02_2019.obj', 'wb') \n",
        "pickle.dump(predictions, file_pi)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "YLb2ZuVOEmJ5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# load serialized predictions\n",
        "import pickle \n",
        "filehandler = open('/content/drive/My Drive/PhD/Data/predictions_18_02_2019.obj', 'r') \n",
        "predictions = pickle.load(filehandler)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "MN2zntNnWyIf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Confusion matrix',\n",
        "                          cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "\n",
        "    print(cm)\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "    plt.tight_layout()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "LeXCaz45W0f8",
        "colab_type": "code",
        "outputId": "5670665a-8fb2-42b7-f7e1-6570dd47fa2b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        }
      },
      "cell_type": "code",
      "source": [
        "labels = test_generator.classes\n",
        "#cm = confusion_matrix(labels, np.round(predictions[:, 0]))\n",
        "cm = confusion_matrix(labels, np.argmax(predictions, axis=1))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-25-58c0f1e0e9f5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtest_generator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclasses\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;31m#cm = confusion_matrix(labels, np.round(predictions[:, 0]))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mcm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconfusion_matrix\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlabels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpredictions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\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[0;31mNameError\u001b[0m: name 'predictions' is not defined"
          ]
        }
      ]
    },
    {
      "metadata": {
        "id": "VOqL4W--XEXk",
        "colab_type": "code",
        "outputId": "61614cd0-5b5b-4b79-9af7-600733dd53e3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 484
        }
      },
      "cell_type": "code",
      "source": [
        "cm_plot_labels = ['Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake']\n",
        "plot_confusion_matrix(cm, cm_plot_labels, title=\"Confusion Matrix\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Confusion matrix, without normalization\n",
            "[[314  11   1   6   1   8  32   2   0   0]\n",
            " [  5 424   0   0   0   6   8   3   0   4]\n",
            " [  1   1 408   1   2  30   3   4   0   0]\n",
            " [  1   0   1 356   6   8   3   0   0   0]\n",
            " [  0   0   1  11 429   8   0   1   0   0]\n",
            " [  2   6  11   0   2 353   1   0   0   0]\n",
            " [ 17   7   1   2   1   0 342   2   1   2]\n",
            " [  3   4   4   1   0   3   7 276   2   0]\n",
            " [  0   0   7   0   1   0   2   5 434   1]\n",
            " [  0   0   0   0   0   0   4   2   0 444]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVAAAAEYCAYAAAAK467YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "ufmAIL7mdgEv",
        "colab_type": "code",
        "outputId": "b3027c3a-3a5e-465d-fea6-be0e3a198eb9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "accuracy_score(labels, np.argmax(predictions, axis=1))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9333333333333333"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 141
        }
      ]
    },
    {
      "metadata": {
        "id": "V42recx_05cn",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#precision_score(labels[:,], np.argmax(predictions, axis=1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "RchOUaGCgilP",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Test Single Images\n"
      ]
    },
    {
      "metadata": {
        "id": "vVtS555xFl2b",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def load(filename):\n",
        "  np_image = Image.open(filename)\n",
        "  np_image = np.array(np_image).astype('float32')/255\n",
        "  np_image = transform.resize(np_image, (224, 224, 3))\n",
        "  np_image = np.expand_dims(np_image, axis=0)\n",
        "  return np_image"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "WBw069I0Ht1y",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "pred = model.predict(load('fatih-1.png'))\n",
        "print(np.argmax(pred*100))\n",
        "res = np.argmax(pred, axis=1)\n",
        "print(pred*100)\n",
        "res[0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Pc7WaCCDjdhV",
        "colab_type": "code",
        "outputId": "c0d04a8b-17ed-46ec-a380-633e69a1ac3c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        }
      },
      "cell_type": "code",
      "source": [
        "model.predict_proba(load('HerbaceousVegetation_2612.jpg'))*100"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/skimage/transform/_warps.py:84: UserWarning: The default mode, 'constant', will be changed to 'reflect' in skimage 0.15.\n",
            "  warn(\"The default mode, 'constant', will be changed to 'reflect' in \"\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[3.8531492e-16, 6.7785423e-04, 9.8759976e-14, 2.2593164e-23,\n",
              "        1.1689909e-25, 3.7096390e-16, 1.2466019e-12, 6.0472352e-08,\n",
              "        6.6771398e-05, 9.9999252e+01]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 79
        }
      ]
    },
    {
      "metadata": {
        "id": "veyumbeugaFp",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "9uGykybCgauj",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Fatih"
      ]
    },
    {
      "metadata": {
        "id": "seuVozXWlfeZ",
        "colab_type": "code",
        "outputId": "0b904783-288d-4ebb-e343-9dc1c41cefb1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "fatih_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/fatih',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size,\n",
        "    shuffle=False,)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 35 images belonging to 10 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HX3j05fGhs0R",
        "colab_type": "code",
        "outputId": "aec79cdf-702c-4182-baf2-86ce0164a77d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        }
      },
      "cell_type": "code",
      "source": [
        "fatih_generator.class_indices"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'AnnualCrop': 1,\n",
              " 'Forest': 8,\n",
              " 'HerbaceousVegetation': 2,\n",
              " 'Highway': 0,\n",
              " 'Industrial': 3,\n",
              " 'Pasture': 7,\n",
              " 'PermanentCrop': 5,\n",
              " 'Residential': 4,\n",
              " 'River': 6,\n",
              " 'SeaLake': 9}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "metadata": {
        "id": "3tav9Yyag7WE",
        "colab_type": "code",
        "outputId": "dd350eab-aa2b-4169-eda9-9f1e658c9e6f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "fatih_generator.samples / batch_size"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.4"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "metadata": {
        "id": "ZbxybXmUrJK6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "fatih_pred = model.predict_generator(fatih_generator, steps=math.ceil(fatih_generator.samples / batch_size))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "gWwyNh_zxxkU",
        "colab_type": "code",
        "outputId": "685cac4c-0a97-4eb1-abcf-1af277aae48f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 612
        }
      },
      "cell_type": "code",
      "source": [
        "for i in range(0, len(fatih_pred)):\n",
        "  print(str(i) + \". \" + fatih_generator.filenames[i] + \"- Pred: \" + str(np.argmax(fatih_pred[i])) + \": \" +\n",
        "        list(fatih_generator.class_indices.keys())[list(fatih_generator.class_indices.values()).index(np.argmax(fatih_pred[i]))])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0. Highway/img0_0.jpg- Pred: 4: Residential\n",
            "1. Highway/img256_128.jpg- Pred: 3: Industrial\n",
            "2. Highway/img256_192.jpg- Pred: 6: River\n",
            "3. Highway/img256_256.jpg- Pred: 6: River\n",
            "4. Highway/img256_64.jpg- Pred: 3: Industrial\n",
            "5. Industrial/img128_192.jpg- Pred: 4: Residential\n",
            "6. Industrial/img128_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "7. Industrial/img64_256.jpg- Pred: 4: Residential\n",
            "8. Industrial/img64_64.jpg- Pred: 4: Residential\n",
            "9. Residential/img0_128.jpg- Pred: 4: Residential\n",
            "10. Residential/img0_192.jpg- Pred: 4: Residential\n",
            "11. Residential/img0_320.jpg- Pred: 3: Industrial\n",
            "12. Residential/img0_384.jpg- Pred: 4: Residential\n",
            "13. Residential/img0_64.jpg- Pred: 4: Residential\n",
            "14. Residential/img128_0.jpg- Pred: 4: Residential\n",
            "15. Residential/img128_128.jpg- Pred: 4: Residential\n",
            "16. Residential/img128_320.jpg- Pred: 4: Residential\n",
            "17. Residential/img128_384.jpg- Pred: 3: Industrial\n",
            "18. Residential/img128_64.jpg- Pred: 4: Residential\n",
            "19. Residential/img192_0.jpg- Pred: 4: Residential\n",
            "20. Residential/img192_128.jpg- Pred: 4: Residential\n",
            "21. Residential/img192_192.jpg- Pred: 4: Residential\n",
            "22. Residential/img192_256.jpg- Pred: 4: Residential\n",
            "23. Residential/img192_320.jpg- Pred: 4: Residential\n",
            "24. Residential/img192_384.jpg- Pred: 4: Residential\n",
            "25. Residential/img192_64.jpg- Pred: 4: Residential\n",
            "26. Residential/img256_0.jpg- Pred: 4: Residential\n",
            "27. Residential/img256_320.jpg- Pred: 6: River\n",
            "28. Residential/img256_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "29. Residential/img64_0.jpg- Pred: 4: Residential\n",
            "30. Residential/img64_128.jpg- Pred: 4: Residential\n",
            "31. Residential/img64_192.jpg- Pred: 4: Residential\n",
            "32. SeaLake/img0_256.jpg- Pred: 6: River\n",
            "33. SeaLake/img64_320.jpg- Pred: 3: Industrial\n",
            "34. SeaLake/img64_384.jpg- Pred: 6: River\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "8pkr_3cEKPzs",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "labels = fatih_generator.classes\n",
        "cm_fatih = confusion_matrix(labels, np.argmax(fatih_pred, axis=1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "7wuQpwoarX2N",
        "colab_type": "code",
        "outputId": "019da9d7-d506-402e-d0f5-7eb25fa72123",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 416
        }
      },
      "cell_type": "code",
      "source": [
        "cm_plot_labels = [\n",
        "    'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake']\n",
        "plot_confusion_matrix(cm_fatih, cm_plot_labels, title=\"Fatih Confusion Matrix\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Confusion matrix, without normalization\n",
            "[[ 0  0  2  1  2  0]\n",
            " [ 0  0  0  0  0  0]\n",
            " [ 0  1  0  3  0  0]\n",
            " [ 0  1  2 19  1  0]\n",
            " [ 0  0  0  0  0  0]\n",
            " [ 0  0  1  0  2  0]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAEYCAYAAADVrdTHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npZlZFf4PdtUVrkx/FmEAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "6YuWAs33UVX3",
        "colab_type": "code",
        "outputId": "308dace9-cf9a-4d4e-d6d7-16291a8a44e8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "accuracy_score(labels, np.argmax(fatih_pred, axis=1))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5428571428571428"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "metadata": {
        "id": "fyrUabtSQrqN",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Güllük"
      ]
    },
    {
      "metadata": {
        "id": "IcYuwMa3yu1-",
        "colab_type": "code",
        "outputId": "c73ae148-f1c1-43b8-c84c-0e507ee50a57",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "gulluk_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/gulluk',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size,\n",
        "    shuffle=False,)\n",
        "\n",
        "gulluk_pred = model.predict_generator(gulluk_generator, steps=math.ceil(gulluk_generator.samples / batch_size))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 1024 images belonging to 10 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "mNdEPgtRQw2g",
        "colab_type": "code",
        "outputId": "9d4874f9-b5cb-4ac9-e95d-765494a57d2c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17425
        }
      },
      "cell_type": "code",
      "source": [
        "for i in range(0, len(gulluk_pred)):\n",
        "  print(str(i) + \". \" + gulluk_generator.filenames[i] + \"- Pred: \" + str(np.argmax(gulluk_pred[i])) + \": \" +\n",
        "        list(gulluk_generator.class_indices.keys())[list(gulluk_generator.class_indices.values()).index(np.argmax(gulluk_pred[i]))])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0. Residential/img0_0.jpg- Pred: 4: Residential\n",
            "1. Residential/img0_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "2. Residential/img0_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "3. Residential/img0_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "4. Residential/img0_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "5. Residential/img0_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "6. Residential/img0_1280.jpg- Pred: 4: Residential\n",
            "7. Residential/img0_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "8. Residential/img0_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "9. Residential/img0_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "10. Residential/img0_1536.jpg- Pred: 0: Highway\n",
            "11. Residential/img0_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "12. Residential/img0_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "13. Residential/img0_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "14. Residential/img0_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "15. Residential/img0_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "16. Residential/img0_192.jpg- Pred: 8: Forest\n",
            "17. Residential/img0_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "18. Residential/img0_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "19. Residential/img0_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "20. Residential/img0_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "21. Residential/img0_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "22. Residential/img0_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "23. Residential/img0_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "24. Residential/img0_576.jpg- Pred: 0: Highway\n",
            "25. Residential/img0_64.jpg- Pred: 8: Forest\n",
            "26. Residential/img0_640.jpg- Pred: 5: PermanentCrop\n",
            "27. Residential/img0_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "28. Residential/img0_768.jpg- Pred: 0: Highway\n",
            "29. Residential/img0_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "30. Residential/img0_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "31. Residential/img0_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "32. Residential/img1024_0.jpg- Pred: 5: PermanentCrop\n",
            "33. Residential/img1024_1024.jpg- Pred: 5: PermanentCrop\n",
            "34. Residential/img1024_1088.jpg- Pred: 0: Highway\n",
            "35. Residential/img1024_1152.jpg- Pred: 0: Highway\n",
            "36. Residential/img1024_1216.jpg- Pred: 7: Pasture\n",
            "37. Residential/img1024_128.jpg- Pred: 6: River\n",
            "38. Residential/img1024_1280.jpg- Pred: 5: PermanentCrop\n",
            "39. Residential/img1024_1344.jpg- Pred: 6: River\n",
            "40. Residential/img1024_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "41. Residential/img1024_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "42. Residential/img1024_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "43. Residential/img1024_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "44. Residential/img1024_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "45. Residential/img1024_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "46. Residential/img1024_1792.jpg- Pred: 3: Industrial\n",
            "47. Residential/img1024_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "48. Residential/img1024_192.jpg- Pred: 7: Pasture\n",
            "49. Residential/img1024_1920.jpg- Pred: 5: PermanentCrop\n",
            "50. Residential/img1024_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "51. Residential/img1024_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "52. Residential/img1024_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "53. Residential/img1024_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "54. Residential/img1024_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "55. Residential/img1024_512.jpg- Pred: 6: River\n",
            "56. Residential/img1024_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "57. Residential/img1024_64.jpg- Pred: 9: SeaLake\n",
            "58. Residential/img1024_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "59. Residential/img1024_704.jpg- Pred: 7: Pasture\n",
            "60. Residential/img1024_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "61. Residential/img1024_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "62. Residential/img1024_896.jpg- Pred: 0: Highway\n",
            "63. Residential/img1024_960.jpg- Pred: 3: Industrial\n",
            "64. Residential/img1088_0.jpg- Pred: 6: River\n",
            "65. Residential/img1088_1024.jpg- Pred: 5: PermanentCrop\n",
            "66. Residential/img1088_1088.jpg- Pred: 5: PermanentCrop\n",
            "67. Residential/img1088_1152.jpg- Pred: 5: PermanentCrop\n",
            "68. Residential/img1088_1216.jpg- Pred: 7: Pasture\n",
            "69. Residential/img1088_128.jpg- Pred: 9: SeaLake\n",
            "70. Residential/img1088_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "71. Residential/img1088_1344.jpg- Pred: 0: Highway\n",
            "72. Residential/img1088_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "73. Residential/img1088_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "74. Residential/img1088_1536.jpg- Pred: 6: River\n",
            "75. Residential/img1088_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "76. Residential/img1088_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "77. Residential/img1088_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "78. Residential/img1088_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "79. Residential/img1088_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "80. Residential/img1088_192.jpg- Pred: 9: SeaLake\n",
            "81. Residential/img1088_1920.jpg- Pred: 3: Industrial\n",
            "82. Residential/img1088_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "83. Residential/img1088_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "84. Residential/img1088_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "85. Residential/img1088_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "86. Residential/img1088_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "87. Residential/img1088_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "88. Residential/img1088_576.jpg- Pred: 7: Pasture\n",
            "89. Residential/img1088_64.jpg- Pred: 9: SeaLake\n",
            "90. Residential/img1088_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "91. Residential/img1088_704.jpg- Pred: 3: Industrial\n",
            "92. Residential/img1088_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "93. Residential/img1088_832.jpg- Pred: 1: AnnualCrop\n",
            "94. Residential/img1088_896.jpg- Pred: 0: Highway\n",
            "95. Residential/img1088_960.jpg- Pred: 5: PermanentCrop\n",
            "96. Residential/img1152_0.jpg- Pred: 9: SeaLake\n",
            "97. Residential/img1152_1024.jpg- Pred: 0: Highway\n",
            "98. Residential/img1152_1088.jpg- Pred: 5: PermanentCrop\n",
            "99. Residential/img1152_1152.jpg- Pred: 5: PermanentCrop\n",
            "100. Residential/img1152_1216.jpg- Pred: 5: PermanentCrop\n",
            "101. Residential/img1152_128.jpg- Pred: 9: SeaLake\n",
            "102. Residential/img1152_1280.jpg- Pred: 0: Highway\n",
            "103. Residential/img1152_1344.jpg- Pred: 0: Highway\n",
            "104. Residential/img1152_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "105. Residential/img1152_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "106. Residential/img1152_1536.jpg- Pred: 0: Highway\n",
            "107. Residential/img1152_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "108. Residential/img1152_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "109. Residential/img1152_1728.jpg- Pred: 7: Pasture\n",
            "110. Residential/img1152_1792.jpg- Pred: 5: PermanentCrop\n",
            "111. Residential/img1152_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "112. Residential/img1152_192.jpg- Pred: 9: SeaLake\n",
            "113. Residential/img1152_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "114. Residential/img1152_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "115. Residential/img1152_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "116. Residential/img1152_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "117. Residential/img1152_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "118. Residential/img1152_448.jpg- Pred: 5: PermanentCrop\n",
            "119. Residential/img1152_512.jpg- Pred: 0: Highway\n",
            "120. Residential/img1152_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "121. Residential/img1152_64.jpg- Pred: 9: SeaLake\n",
            "122. Residential/img1152_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "123. Residential/img1152_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "124. Residential/img1152_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "125. Residential/img1152_832.jpg- Pred: 1: AnnualCrop\n",
            "126. Residential/img1152_896.jpg- Pred: 5: PermanentCrop\n",
            "127. Residential/img1152_960.jpg- Pred: 0: Highway\n",
            "128. Residential/img1216_0.jpg- Pred: 9: SeaLake\n",
            "129. Residential/img1216_1024.jpg- Pred: 5: PermanentCrop\n",
            "130. Residential/img1216_1088.jpg- Pred: 5: PermanentCrop\n",
            "131. Residential/img1216_1152.jpg- Pred: 0: Highway\n",
            "132. Residential/img1216_1216.jpg- Pred: 5: PermanentCrop\n",
            "133. Residential/img1216_128.jpg- Pred: 9: SeaLake\n",
            "134. Residential/img1216_1280.jpg- Pred: 7: Pasture\n",
            "135. Residential/img1216_1344.jpg- Pred: 0: Highway\n",
            "136. Residential/img1216_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "137. Residential/img1216_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "138. Residential/img1216_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "139. Residential/img1216_1600.jpg- Pred: 7: Pasture\n",
            "140. Residential/img1216_1664.jpg- Pred: 7: Pasture\n",
            "141. Residential/img1216_1728.jpg- Pred: 7: Pasture\n",
            "142. Residential/img1216_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "143. Residential/img1216_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "144. Residential/img1216_192.jpg- Pred: 9: SeaLake\n",
            "145. Residential/img1216_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "146. Residential/img1216_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "147. Residential/img1216_256.jpg- Pred: 7: Pasture\n",
            "148. Residential/img1216_320.jpg- Pred: 7: Pasture\n",
            "149. Residential/img1216_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "150. Residential/img1216_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "151. Residential/img1216_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "152. Residential/img1216_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "153. Residential/img1216_64.jpg- Pred: 9: SeaLake\n",
            "154. Residential/img1216_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "155. Residential/img1216_704.jpg- Pred: 7: Pasture\n",
            "156. Residential/img1216_768.jpg- Pred: 7: Pasture\n",
            "157. Residential/img1216_832.jpg- Pred: 7: Pasture\n",
            "158. Residential/img1216_896.jpg- Pred: 0: Highway\n",
            "159. Residential/img1216_960.jpg- Pred: 1: AnnualCrop\n",
            "160. Residential/img1280_0.jpg- Pred: 9: SeaLake\n",
            "161. Residential/img1280_1024.jpg- Pred: 1: AnnualCrop\n",
            "162. Residential/img1280_1088.jpg- Pred: 1: AnnualCrop\n",
            "163. Residential/img1280_1152.jpg- Pred: 6: River\n",
            "164. Residential/img1280_1216.jpg- Pred: 0: Highway\n",
            "165. Residential/img1280_128.jpg- Pred: 9: SeaLake\n",
            "166. Residential/img1280_1280.jpg- Pred: 1: AnnualCrop\n",
            "167. Residential/img1280_1344.jpg- Pred: 7: Pasture\n",
            "168. Residential/img1280_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "169. Residential/img1280_1472.jpg- Pred: 4: Residential\n",
            "170. Residential/img1280_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "171. Residential/img1280_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "172. Residential/img1280_1664.jpg- Pred: 7: Pasture\n",
            "173. Residential/img1280_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "174. Residential/img1280_1792.jpg- Pred: 7: Pasture\n",
            "175. Residential/img1280_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "176. Residential/img1280_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "177. Residential/img1280_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "178. Residential/img1280_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "179. Residential/img1280_256.jpg- Pred: 6: River\n",
            "180. Residential/img1280_320.jpg- Pred: 6: River\n",
            "181. Residential/img1280_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "182. Residential/img1280_448.jpg- Pred: 0: Highway\n",
            "183. Residential/img1280_512.jpg- Pred: 6: River\n",
            "184. Residential/img1280_576.jpg- Pred: 6: River\n",
            "185. Residential/img1280_64.jpg- Pred: 9: SeaLake\n",
            "186. Residential/img1280_640.jpg- Pred: 6: River\n",
            "187. Residential/img1280_704.jpg- Pred: 0: Highway\n",
            "188. Residential/img1280_768.jpg- Pred: 7: Pasture\n",
            "189. Residential/img1280_832.jpg- Pred: 0: Highway\n",
            "190. Residential/img1280_896.jpg- Pred: 7: Pasture\n",
            "191. Residential/img1280_960.jpg- Pred: 0: Highway\n",
            "192. Residential/img128_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "193. Residential/img128_1024.jpg- Pred: 8: Forest\n",
            "194. Residential/img128_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "195. Residential/img128_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "196. Residential/img128_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "197. Residential/img128_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "198. Residential/img128_1280.jpg- Pred: 0: Highway\n",
            "199. Residential/img128_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "200. Residential/img128_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "201. Residential/img128_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "202. Residential/img128_1536.jpg- Pred: 5: PermanentCrop\n",
            "203. Residential/img128_1600.jpg- Pred: 5: PermanentCrop\n",
            "204. Residential/img128_1664.jpg- Pred: 5: PermanentCrop\n",
            "205. Residential/img128_1728.jpg- Pred: 0: Highway\n",
            "206. Residential/img128_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "207. Residential/img128_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "208. Residential/img128_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "209. Residential/img128_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "210. Residential/img128_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "211. Residential/img128_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "212. Residential/img128_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "213. Residential/img128_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "214. Residential/img128_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "215. Residential/img128_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "216. Residential/img128_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "217. Residential/img128_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "218. Residential/img128_640.jpg- Pred: 6: River\n",
            "219. Residential/img128_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "220. Residential/img128_768.jpg- Pred: 7: Pasture\n",
            "221. Residential/img128_832.jpg- Pred: 5: PermanentCrop\n",
            "222. Residential/img128_896.jpg- Pred: 0: Highway\n",
            "223. Residential/img128_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "224. Residential/img1344_0.jpg- Pred: 9: SeaLake\n",
            "225. Residential/img1344_1024.jpg- Pred: 1: AnnualCrop\n",
            "226. Residential/img1344_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "227. Residential/img1344_1152.jpg- Pred: 1: AnnualCrop\n",
            "228. Residential/img1344_1216.jpg- Pred: 1: AnnualCrop\n",
            "229. Residential/img1344_128.jpg- Pred: 9: SeaLake\n",
            "230. Residential/img1344_1280.jpg- Pred: 5: PermanentCrop\n",
            "231. Residential/img1344_1344.jpg- Pred: 7: Pasture\n",
            "232. Residential/img1344_1408.jpg- Pred: 5: PermanentCrop\n",
            "233. Residential/img1344_1472.jpg- Pred: 4: Residential\n",
            "234. Residential/img1344_1536.jpg- Pred: 3: Industrial\n",
            "235. Residential/img1344_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "236. Residential/img1344_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "237. Residential/img1344_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "238. Residential/img1344_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "239. Residential/img1344_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "240. Residential/img1344_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "241. Residential/img1344_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "242. Residential/img1344_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "243. Residential/img1344_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "244. Residential/img1344_320.jpg- Pred: 6: River\n",
            "245. Residential/img1344_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "246. Residential/img1344_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "247. Residential/img1344_512.jpg- Pred: 0: Highway\n",
            "248. Residential/img1344_576.jpg- Pred: 6: River\n",
            "249. Residential/img1344_64.jpg- Pred: 9: SeaLake\n",
            "250. Residential/img1344_640.jpg- Pred: 3: Industrial\n",
            "251. Residential/img1344_704.jpg- Pred: 6: River\n",
            "252. Residential/img1344_768.jpg- Pred: 6: River\n",
            "253. Residential/img1344_832.jpg- Pred: 6: River\n",
            "254. Residential/img1344_896.jpg- Pred: 3: Industrial\n",
            "255. Residential/img1344_960.jpg- Pred: 7: Pasture\n",
            "256. Residential/img1408_0.jpg- Pred: 9: SeaLake\n",
            "257. Residential/img1408_1024.jpg- Pred: 0: Highway\n",
            "258. Residential/img1408_1088.jpg- Pred: 4: Residential\n",
            "259. Residential/img1408_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "260. Residential/img1408_1216.jpg- Pred: 0: Highway\n",
            "261. Residential/img1408_128.jpg- Pred: 6: River\n",
            "262. Residential/img1408_1280.jpg- Pred: 1: AnnualCrop\n",
            "263. Residential/img1408_1344.jpg- Pred: 5: PermanentCrop\n",
            "264. Residential/img1408_1408.jpg- Pred: 5: PermanentCrop\n",
            "265. Residential/img1408_1472.jpg- Pred: 5: PermanentCrop\n",
            "266. Residential/img1408_1536.jpg- Pred: 4: Residential\n",
            "267. Residential/img1408_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "268. Residential/img1408_1664.jpg- Pred: 7: Pasture\n",
            "269. Residential/img1408_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "270. Residential/img1408_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "271. Residential/img1408_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "272. Residential/img1408_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "273. Residential/img1408_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "274. Residential/img1408_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "275. Residential/img1408_256.jpg- Pred: 3: Industrial\n",
            "276. Residential/img1408_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "277. Residential/img1408_384.jpg- Pred: 0: Highway\n",
            "278. Residential/img1408_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "279. Residential/img1408_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "280. Residential/img1408_576.jpg- Pred: 3: Industrial\n",
            "281. Residential/img1408_64.jpg- Pred: 9: SeaLake\n",
            "282. Residential/img1408_640.jpg- Pred: 5: PermanentCrop\n",
            "283. Residential/img1408_704.jpg- Pred: 3: Industrial\n",
            "284. Residential/img1408_768.jpg- Pred: 3: Industrial\n",
            "285. Residential/img1408_832.jpg- Pred: 3: Industrial\n",
            "286. Residential/img1408_896.jpg- Pred: 4: Residential\n",
            "287. Residential/img1408_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "288. Residential/img1472_0.jpg- Pred: 9: SeaLake\n",
            "289. Residential/img1472_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "290. Residential/img1472_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "291. Residential/img1472_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "292. Residential/img1472_1216.jpg- Pred: 7: Pasture\n",
            "293. Residential/img1472_128.jpg- Pred: 6: River\n",
            "294. Residential/img1472_1280.jpg- Pred: 6: River\n",
            "295. Residential/img1472_1344.jpg- Pred: 0: Highway\n",
            "296. Residential/img1472_1408.jpg- Pred: 0: Highway\n",
            "297. Residential/img1472_1472.jpg- Pred: 4: Residential\n",
            "298. Residential/img1472_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "299. Residential/img1472_1600.jpg- Pred: 8: Forest\n",
            "300. Residential/img1472_1664.jpg- Pred: 8: Forest\n",
            "301. Residential/img1472_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "302. Residential/img1472_1792.jpg- Pred: 6: River\n",
            "303. Residential/img1472_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "304. Residential/img1472_192.jpg- Pred: 3: Industrial\n",
            "305. Residential/img1472_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "306. Residential/img1472_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "307. Residential/img1472_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "308. Residential/img1472_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "309. Residential/img1472_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "310. Residential/img1472_448.jpg- Pred: 6: River\n",
            "311. Residential/img1472_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "312. Residential/img1472_576.jpg- Pred: 5: PermanentCrop\n",
            "313. Residential/img1472_64.jpg- Pred: 9: SeaLake\n",
            "314. Residential/img1472_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "315. Residential/img1472_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "316. Residential/img1472_768.jpg- Pred: 0: Highway\n",
            "317. Residential/img1472_832.jpg- Pred: 0: Highway\n",
            "318. Residential/img1472_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "319. Residential/img1472_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "320. Residential/img1536_0.jpg- Pred: 9: SeaLake\n",
            "321. Residential/img1536_1024.jpg- Pred: 6: River\n",
            "322. Residential/img1536_1088.jpg- Pred: 5: PermanentCrop\n",
            "323. Residential/img1536_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "324. Residential/img1536_1216.jpg- Pred: 7: Pasture\n",
            "325. Residential/img1536_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "326. Residential/img1536_1280.jpg- Pred: 0: Highway\n",
            "327. Residential/img1536_1344.jpg- Pred: 1: AnnualCrop\n",
            "328. Residential/img1536_1408.jpg- Pred: 1: AnnualCrop\n",
            "329. Residential/img1536_1472.jpg- Pred: 1: AnnualCrop\n",
            "330. Residential/img1536_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "331. Residential/img1536_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "332. Residential/img1536_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "333. Residential/img1536_1728.jpg- Pred: 0: Highway\n",
            "334. Residential/img1536_1792.jpg- Pred: 6: River\n",
            "335. Residential/img1536_1856.jpg- Pred: 6: River\n",
            "336. Residential/img1536_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "337. Residential/img1536_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "338. Residential/img1536_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "339. Residential/img1536_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "340. Residential/img1536_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "341. Residential/img1536_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "342. Residential/img1536_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "343. Residential/img1536_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "344. Residential/img1536_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "345. Residential/img1536_64.jpg- Pred: 6: River\n",
            "346. Residential/img1536_640.jpg- Pred: 0: Highway\n",
            "347. Residential/img1536_704.jpg- Pred: 0: Highway\n",
            "348. Residential/img1536_768.jpg- Pred: 3: Industrial\n",
            "349. Residential/img1536_832.jpg- Pred: 0: Highway\n",
            "350. Residential/img1536_896.jpg- Pred: 7: Pasture\n",
            "351. Residential/img1536_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "352. Residential/img1600_0.jpg- Pred: 9: SeaLake\n",
            "353. Residential/img1600_1024.jpg- Pred: 5: PermanentCrop\n",
            "354. Residential/img1600_1088.jpg- Pred: 5: PermanentCrop\n",
            "355. Residential/img1600_1152.jpg- Pred: 5: PermanentCrop\n",
            "356. Residential/img1600_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "357. Residential/img1600_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "358. Residential/img1600_1280.jpg- Pred: 1: AnnualCrop\n",
            "359. Residential/img1600_1344.jpg- Pred: 6: River\n",
            "360. Residential/img1600_1408.jpg- Pred: 6: River\n",
            "361. Residential/img1600_1472.jpg- Pred: 1: AnnualCrop\n",
            "362. Residential/img1600_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "363. Residential/img1600_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "364. Residential/img1600_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "365. Residential/img1600_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "366. Residential/img1600_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "367. Residential/img1600_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "368. Residential/img1600_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "369. Residential/img1600_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "370. Residential/img1600_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "371. Residential/img1600_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "372. Residential/img1600_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "373. Residential/img1600_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "374. Residential/img1600_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "375. Residential/img1600_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "376. Residential/img1600_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "377. Residential/img1600_64.jpg- Pred: 3: Industrial\n",
            "378. Residential/img1600_640.jpg- Pred: 7: Pasture\n",
            "379. Residential/img1600_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "380. Residential/img1600_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "381. Residential/img1600_832.jpg- Pred: 5: PermanentCrop\n",
            "382. Residential/img1600_896.jpg- Pred: 1: AnnualCrop\n",
            "383. Residential/img1600_960.jpg- Pred: 0: Highway\n",
            "384. Residential/img1664_0.jpg- Pred: 3: Industrial\n",
            "385. Residential/img1664_1024.jpg- Pred: 5: PermanentCrop\n",
            "386. Residential/img1664_1088.jpg- Pred: 5: PermanentCrop\n",
            "387. Residential/img1664_1152.jpg- Pred: 5: PermanentCrop\n",
            "388. Residential/img1664_1216.jpg- Pred: 1: AnnualCrop\n",
            "389. Residential/img1664_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "390. Residential/img1664_1280.jpg- Pred: 1: AnnualCrop\n",
            "391. Residential/img1664_1344.jpg- Pred: 0: Highway\n",
            "392. Residential/img1664_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "393. Residential/img1664_1472.jpg- Pred: 0: Highway\n",
            "394. Residential/img1664_1536.jpg- Pred: 4: Residential\n",
            "395. Residential/img1664_1600.jpg- Pred: 4: Residential\n",
            "396. Residential/img1664_1664.jpg- Pred: 4: Residential\n",
            "397. Residential/img1664_1728.jpg- Pred: 4: Residential\n",
            "398. Residential/img1664_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "399. Residential/img1664_1856.jpg- Pred: 5: PermanentCrop\n",
            "400. Residential/img1664_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "401. Residential/img1664_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "402. Residential/img1664_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "403. Residential/img1664_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "404. Residential/img1664_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "405. Residential/img1664_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "406. Residential/img1664_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "407. Residential/img1664_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "408. Residential/img1664_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "409. Residential/img1664_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "410. Residential/img1664_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "411. Residential/img1664_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "412. Residential/img1664_768.jpg- Pred: 6: River\n",
            "413. Residential/img1664_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "414. Residential/img1664_896.jpg- Pred: 0: Highway\n",
            "415. Residential/img1664_960.jpg- Pred: 0: Highway\n",
            "416. Residential/img1728_0.jpg- Pred: 3: Industrial\n",
            "417. Residential/img1728_1024.jpg- Pred: 5: PermanentCrop\n",
            "418. Residential/img1728_1088.jpg- Pred: 5: PermanentCrop\n",
            "419. Residential/img1728_1152.jpg- Pred: 6: River\n",
            "420. Residential/img1728_1216.jpg- Pred: 1: AnnualCrop\n",
            "421. Residential/img1728_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "422. Residential/img1728_1280.jpg- Pred: 6: River\n",
            "423. Residential/img1728_1344.jpg- Pred: 5: PermanentCrop\n",
            "424. Residential/img1728_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "425. Residential/img1728_1472.jpg- Pred: 4: Residential\n",
            "426. Residential/img1728_1536.jpg- Pred: 5: PermanentCrop\n",
            "427. Residential/img1728_1600.jpg- Pred: 5: PermanentCrop\n",
            "428. Residential/img1728_1664.jpg- Pred: 5: PermanentCrop\n",
            "429. Residential/img1728_1728.jpg- Pred: 5: PermanentCrop\n",
            "430. Residential/img1728_1792.jpg- Pred: 5: PermanentCrop\n",
            "431. Residential/img1728_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "432. Residential/img1728_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "433. Residential/img1728_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "434. Residential/img1728_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "435. Residential/img1728_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "436. Residential/img1728_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "437. Residential/img1728_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "438. Residential/img1728_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "439. Residential/img1728_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "440. Residential/img1728_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "441. Residential/img1728_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "442. Residential/img1728_640.jpg- Pred: 0: Highway\n",
            "443. Residential/img1728_704.jpg- Pred: 6: River\n",
            "444. Residential/img1728_768.jpg- Pred: 7: Pasture\n",
            "445. Residential/img1728_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "446. Residential/img1728_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "447. Residential/img1728_960.jpg- Pred: 0: Highway\n",
            "448. Residential/img1792_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "449. Residential/img1792_1024.jpg- Pred: 6: River\n",
            "450. Residential/img1792_1088.jpg- Pred: 3: Industrial\n",
            "451. Residential/img1792_1152.jpg- Pred: 0: Highway\n",
            "452. Residential/img1792_1216.jpg- Pred: 6: River\n",
            "453. Residential/img1792_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "454. Residential/img1792_1280.jpg- Pred: 6: River\n",
            "455. Residential/img1792_1344.jpg- Pred: 6: River\n",
            "456. Residential/img1792_1408.jpg- Pred: 5: PermanentCrop\n",
            "457. Residential/img1792_1472.jpg- Pred: 5: PermanentCrop\n",
            "458. Residential/img1792_1536.jpg- Pred: 0: Highway\n",
            "459. Residential/img1792_1600.jpg- Pred: 5: PermanentCrop\n",
            "460. Residential/img1792_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "461. Residential/img1792_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "462. Residential/img1792_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "463. Residential/img1792_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "464. Residential/img1792_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "465. Residential/img1792_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "466. Residential/img1792_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "467. Residential/img1792_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "468. Residential/img1792_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "469. Residential/img1792_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "470. Residential/img1792_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "471. Residential/img1792_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "472. Residential/img1792_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "473. Residential/img1792_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "474. Residential/img1792_640.jpg- Pred: 6: River\n",
            "475. Residential/img1792_704.jpg- Pred: 6: River\n",
            "476. Residential/img1792_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "477. Residential/img1792_832.jpg- Pred: 6: River\n",
            "478. Residential/img1792_896.jpg- Pred: 6: River\n",
            "479. Residential/img1792_960.jpg- Pred: 6: River\n",
            "480. Residential/img1856_0.jpg- Pred: 6: River\n",
            "481. Residential/img1856_1024.jpg- Pred: 0: Highway\n",
            "482. Residential/img1856_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "483. Residential/img1856_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "484. Residential/img1856_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "485. Residential/img1856_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "486. Residential/img1856_1280.jpg- Pred: 0: Highway\n",
            "487. Residential/img1856_1344.jpg- Pred: 0: Highway\n",
            "488. Residential/img1856_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "489. Residential/img1856_1472.jpg- Pred: 7: Pasture\n",
            "490. Residential/img1856_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "491. Residential/img1856_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "492. Residential/img1856_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "493. Residential/img1856_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "494. Residential/img1856_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "495. Residential/img1856_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "496. Residential/img1856_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "497. Residential/img1856_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "498. Residential/img1856_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "499. Residential/img1856_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "500. Residential/img1856_320.jpg- Pred: 7: Pasture\n",
            "501. Residential/img1856_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "502. Residential/img1856_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "503. Residential/img1856_512.jpg- Pred: 6: River\n",
            "504. Residential/img1856_576.jpg- Pred: 0: Highway\n",
            "505. Residential/img1856_64.jpg- Pred: 7: Pasture\n",
            "506. Residential/img1856_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "507. Residential/img1856_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "508. Residential/img1856_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "509. Residential/img1856_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "510. Residential/img1856_896.jpg- Pred: 6: River\n",
            "511. Residential/img1856_960.jpg- Pred: 0: Highway\n",
            "512. Residential/img1920_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "513. Residential/img1920_1024.jpg- Pred: 7: Pasture\n",
            "514. Residential/img1920_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "515. Residential/img1920_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "516. Residential/img1920_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "517. Residential/img1920_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "518. Residential/img1920_1280.jpg- Pred: 6: River\n",
            "519. Residential/img1920_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "520. Residential/img1920_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "521. Residential/img1920_1472.jpg- Pred: 5: PermanentCrop\n",
            "522. Residential/img1920_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "523. Residential/img1920_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "524. Residential/img1920_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "525. Residential/img1920_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "526. Residential/img1920_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "527. Residential/img1920_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "528. Residential/img1920_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "529. Residential/img1920_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "530. Residential/img1920_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "531. Residential/img1920_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "532. Residential/img1920_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "533. Residential/img1920_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "534. Residential/img1920_448.jpg- Pred: 0: Highway\n",
            "535. Residential/img1920_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "536. Residential/img1920_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "537. Residential/img1920_64.jpg- Pred: 9: SeaLake\n",
            "538. Residential/img1920_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "539. Residential/img1920_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "540. Residential/img1920_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "541. Residential/img1920_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "542. Residential/img1920_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "543. Residential/img1920_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "544. Residential/img192_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "545. Residential/img192_1024.jpg- Pred: 6: River\n",
            "546. Residential/img192_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "547. Residential/img192_1152.jpg- Pred: 3: Industrial\n",
            "548. Residential/img192_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "549. Residential/img192_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "550. Residential/img192_1280.jpg- Pred: 0: Highway\n",
            "551. Residential/img192_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "552. Residential/img192_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "553. Residential/img192_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "554. Residential/img192_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "555. Residential/img192_1600.jpg- Pred: 5: PermanentCrop\n",
            "556. Residential/img192_1664.jpg- Pred: 5: PermanentCrop\n",
            "557. Residential/img192_1728.jpg- Pred: 5: PermanentCrop\n",
            "558. Residential/img192_1792.jpg- Pred: 0: Highway\n",
            "559. Residential/img192_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "560. Residential/img192_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "561. Residential/img192_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "562. Residential/img192_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "563. Residential/img192_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "564. Residential/img192_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "565. Residential/img192_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "566. Residential/img192_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "567. Residential/img192_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "568. Residential/img192_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "569. Residential/img192_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "570. Residential/img192_640.jpg- Pred: 0: Highway\n",
            "571. Residential/img192_704.jpg- Pred: 5: PermanentCrop\n",
            "572. Residential/img192_768.jpg- Pred: 5: PermanentCrop\n",
            "573. Residential/img192_832.jpg- Pred: 5: PermanentCrop\n",
            "574. Residential/img192_896.jpg- Pred: 7: Pasture\n",
            "575. Residential/img192_960.jpg- Pred: 0: Highway\n",
            "576. Residential/img1984_0.jpg- Pred: 0: Highway\n",
            "577. Residential/img1984_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "578. Residential/img1984_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "579. Residential/img1984_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "580. Residential/img1984_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "581. Residential/img1984_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "582. Residential/img1984_1280.jpg- Pred: 6: River\n",
            "583. Residential/img1984_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "584. Residential/img1984_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "585. Residential/img1984_1472.jpg- Pred: 0: Highway\n",
            "586. Residential/img1984_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "587. Residential/img1984_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "588. Residential/img1984_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "589. Residential/img1984_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "590. Residential/img1984_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "591. Residential/img1984_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "592. Residential/img1984_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "593. Residential/img1984_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "594. Residential/img1984_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "595. Residential/img1984_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "596. Residential/img1984_320.jpg- Pred: 0: Highway\n",
            "597. Residential/img1984_384.jpg- Pred: 3: Industrial\n",
            "598. Residential/img1984_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "599. Residential/img1984_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "600. Residential/img1984_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "601. Residential/img1984_64.jpg- Pred: 0: Highway\n",
            "602. Residential/img1984_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "603. Residential/img1984_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "604. Residential/img1984_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "605. Residential/img1984_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "606. Residential/img1984_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "607. Residential/img1984_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "608. Residential/img256_0.jpg- Pred: 8: Forest\n",
            "609. Residential/img256_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "610. Residential/img256_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "611. Residential/img256_1152.jpg- Pred: 1: AnnualCrop\n",
            "612. Residential/img256_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "613. Residential/img256_128.jpg- Pred: 8: Forest\n",
            "614. Residential/img256_1280.jpg- Pred: 1: AnnualCrop\n",
            "615. Residential/img256_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "616. Residential/img256_1408.jpg- Pred: 6: River\n",
            "617. Residential/img256_1472.jpg- Pred: 7: Pasture\n",
            "618. Residential/img256_1536.jpg- Pred: 6: River\n",
            "619. Residential/img256_1600.jpg- Pred: 6: River\n",
            "620. Residential/img256_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "621. Residential/img256_1728.jpg- Pred: 5: PermanentCrop\n",
            "622. Residential/img256_1792.jpg- Pred: 5: PermanentCrop\n",
            "623. Residential/img256_1856.jpg- Pred: 4: Residential\n",
            "624. Residential/img256_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "625. Residential/img256_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "626. Residential/img256_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "627. Residential/img256_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "628. Residential/img256_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "629. Residential/img256_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "630. Residential/img256_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "631. Residential/img256_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "632. Residential/img256_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "633. Residential/img256_64.jpg- Pred: 6: River\n",
            "634. Residential/img256_640.jpg- Pred: 5: PermanentCrop\n",
            "635. Residential/img256_704.jpg- Pred: 7: Pasture\n",
            "636. Residential/img256_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "637. Residential/img256_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "638. Residential/img256_896.jpg- Pred: 7: Pasture\n",
            "639. Residential/img256_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "640. Residential/img320_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "641. Residential/img320_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "642. Residential/img320_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "643. Residential/img320_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "644. Residential/img320_1216.jpg- Pred: 3: Industrial\n",
            "645. Residential/img320_128.jpg- Pred: 8: Forest\n",
            "646. Residential/img320_1280.jpg- Pred: 0: Highway\n",
            "647. Residential/img320_1344.jpg- Pred: 6: River\n",
            "648. Residential/img320_1408.jpg- Pred: 3: Industrial\n",
            "649. Residential/img320_1472.jpg- Pred: 3: Industrial\n",
            "650. Residential/img320_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "651. Residential/img320_1600.jpg- Pred: 9: SeaLake\n",
            "652. Residential/img320_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "653. Residential/img320_1728.jpg- Pred: 5: PermanentCrop\n",
            "654. Residential/img320_1792.jpg- Pred: 4: Residential\n",
            "655. Residential/img320_1856.jpg- Pred: 3: Industrial\n",
            "656. Residential/img320_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "657. Residential/img320_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "658. Residential/img320_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "659. Residential/img320_256.jpg- Pred: 8: Forest\n",
            "660. Residential/img320_320.jpg- Pred: 8: Forest\n",
            "661. Residential/img320_384.jpg- Pred: 7: Pasture\n",
            "662. Residential/img320_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "663. Residential/img320_512.jpg- Pred: 7: Pasture\n",
            "664. Residential/img320_576.jpg- Pred: 0: Highway\n",
            "665. Residential/img320_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "666. Residential/img320_640.jpg- Pred: 5: PermanentCrop\n",
            "667. Residential/img320_704.jpg- Pred: 5: PermanentCrop\n",
            "668. Residential/img320_768.jpg- Pred: 5: PermanentCrop\n",
            "669. Residential/img320_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "670. Residential/img320_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "671. Residential/img320_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "672. Residential/img384_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "673. Residential/img384_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "674. Residential/img384_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "675. Residential/img384_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "676. Residential/img384_1216.jpg- Pred: 5: PermanentCrop\n",
            "677. Residential/img384_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "678. Residential/img384_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "679. Residential/img384_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "680. Residential/img384_1408.jpg- Pred: 6: River\n",
            "681. Residential/img384_1472.jpg- Pred: 0: Highway\n",
            "682. Residential/img384_1536.jpg- Pred: 0: Highway\n",
            "683. Residential/img384_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "684. Residential/img384_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "685. Residential/img384_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "686. Residential/img384_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "687. Residential/img384_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "688. Residential/img384_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "689. Residential/img384_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "690. Residential/img384_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "691. Residential/img384_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "692. Residential/img384_320.jpg- Pred: 8: Forest\n",
            "693. Residential/img384_384.jpg- Pred: 7: Pasture\n",
            "694. Residential/img384_448.jpg- Pred: 8: Forest\n",
            "695. Residential/img384_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "696. Residential/img384_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "697. Residential/img384_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "698. Residential/img384_640.jpg- Pred: 0: Highway\n",
            "699. Residential/img384_704.jpg- Pred: 7: Pasture\n",
            "700. Residential/img384_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "701. Residential/img384_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "702. Residential/img384_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "703. Residential/img384_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "704. Residential/img448_0.jpg- Pred: 8: Forest\n",
            "705. Residential/img448_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "706. Residential/img448_1088.jpg- Pred: 5: PermanentCrop\n",
            "707. Residential/img448_1152.jpg- Pred: 5: PermanentCrop\n",
            "708. Residential/img448_1216.jpg- Pred: 5: PermanentCrop\n",
            "709. Residential/img448_128.jpg- Pred: 0: Highway\n",
            "710. Residential/img448_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "711. Residential/img448_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "712. Residential/img448_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "713. Residential/img448_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "714. Residential/img448_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "715. Residential/img448_1600.jpg- Pred: 6: River\n",
            "716. Residential/img448_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "717. Residential/img448_1728.jpg- Pred: 5: PermanentCrop\n",
            "718. Residential/img448_1792.jpg- Pred: 6: River\n",
            "719. Residential/img448_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "720. Residential/img448_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "721. Residential/img448_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "722. Residential/img448_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "723. Residential/img448_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "724. Residential/img448_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "725. Residential/img448_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "726. Residential/img448_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "727. Residential/img448_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "728. Residential/img448_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "729. Residential/img448_64.jpg- Pred: 0: Highway\n",
            "730. Residential/img448_640.jpg- Pred: 0: Highway\n",
            "731. Residential/img448_704.jpg- Pred: 5: PermanentCrop\n",
            "732. Residential/img448_768.jpg- Pred: 0: Highway\n",
            "733. Residential/img448_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "734. Residential/img448_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "735. Residential/img448_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "736. Residential/img512_0.jpg- Pred: 8: Forest\n",
            "737. Residential/img512_1024.jpg- Pred: 5: PermanentCrop\n",
            "738. Residential/img512_1088.jpg- Pred: 0: Highway\n",
            "739. Residential/img512_1152.jpg- Pred: 5: PermanentCrop\n",
            "740. Residential/img512_1216.jpg- Pred: 0: Highway\n",
            "741. Residential/img512_128.jpg- Pred: 7: Pasture\n",
            "742. Residential/img512_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "743. Residential/img512_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "744. Residential/img512_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "745. Residential/img512_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "746. Residential/img512_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "747. Residential/img512_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "748. Residential/img512_1664.jpg- Pred: 4: Residential\n",
            "749. Residential/img512_1728.jpg- Pred: 2: HerbaceousVegetation\n",
            "750. Residential/img512_1792.jpg- Pred: 5: PermanentCrop\n",
            "751. Residential/img512_1856.jpg- Pred: 4: Residential\n",
            "752. Residential/img512_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "753. Residential/img512_1920.jpg- Pred: 4: Residential\n",
            "754. Residential/img512_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "755. Residential/img512_256.jpg- Pred: 8: Forest\n",
            "756. Residential/img512_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "757. Residential/img512_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "758. Residential/img512_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "759. Residential/img512_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "760. Residential/img512_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "761. Residential/img512_64.jpg- Pred: 8: Forest\n",
            "762. Residential/img512_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "763. Residential/img512_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "764. Residential/img512_768.jpg- Pred: 0: Highway\n",
            "765. Residential/img512_832.jpg- Pred: 1: AnnualCrop\n",
            "766. Residential/img512_896.jpg- Pred: 5: PermanentCrop\n",
            "767. Residential/img512_960.jpg- Pred: 7: Pasture\n",
            "768. Residential/img576_0.jpg- Pred: 8: Forest\n",
            "769. Residential/img576_1024.jpg- Pred: 7: Pasture\n",
            "770. Residential/img576_1088.jpg- Pred: 5: PermanentCrop\n",
            "771. Residential/img576_1152.jpg- Pred: 5: PermanentCrop\n",
            "772. Residential/img576_1216.jpg- Pred: 0: Highway\n",
            "773. Residential/img576_128.jpg- Pred: 8: Forest\n",
            "774. Residential/img576_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "775. Residential/img576_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "776. Residential/img576_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "777. Residential/img576_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "778. Residential/img576_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "779. Residential/img576_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "780. Residential/img576_1664.jpg- Pred: 5: PermanentCrop\n",
            "781. Residential/img576_1728.jpg- Pred: 4: Residential\n",
            "782. Residential/img576_1792.jpg- Pred: 2: HerbaceousVegetation\n",
            "783. Residential/img576_1856.jpg- Pred: 3: Industrial\n",
            "784. Residential/img576_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "785. Residential/img576_1920.jpg- Pred: 0: Highway\n",
            "786. Residential/img576_1984.jpg- Pred: 5: PermanentCrop\n",
            "787. Residential/img576_256.jpg- Pred: 0: Highway\n",
            "788. Residential/img576_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "789. Residential/img576_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "790. Residential/img576_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "791. Residential/img576_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "792. Residential/img576_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "793. Residential/img576_64.jpg- Pred: 7: Pasture\n",
            "794. Residential/img576_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "795. Residential/img576_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "796. Residential/img576_768.jpg- Pred: 1: AnnualCrop\n",
            "797. Residential/img576_832.jpg- Pred: 1: AnnualCrop\n",
            "798. Residential/img576_896.jpg- Pred: 1: AnnualCrop\n",
            "799. Residential/img576_960.jpg- Pred: 1: AnnualCrop\n",
            "800. Residential/img640_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "801. Residential/img640_1024.jpg- Pred: 7: Pasture\n",
            "802. Residential/img640_1088.jpg- Pred: 0: Highway\n",
            "803. Residential/img640_1152.jpg- Pred: 5: PermanentCrop\n",
            "804. Residential/img640_1216.jpg- Pred: 5: PermanentCrop\n",
            "805. Residential/img640_128.jpg- Pred: 8: Forest\n",
            "806. Residential/img640_1280.jpg- Pred: 0: Highway\n",
            "807. Residential/img640_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "808. Residential/img640_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "809. Residential/img640_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "810. Residential/img640_1536.jpg- Pred: 4: Residential\n",
            "811. Residential/img640_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "812. Residential/img640_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "813. Residential/img640_1728.jpg- Pred: 4: Residential\n",
            "814. Residential/img640_1792.jpg- Pred: 4: Residential\n",
            "815. Residential/img640_1856.jpg- Pred: 3: Industrial\n",
            "816. Residential/img640_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "817. Residential/img640_1920.jpg- Pred: 4: Residential\n",
            "818. Residential/img640_1984.jpg- Pred: 5: PermanentCrop\n",
            "819. Residential/img640_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "820. Residential/img640_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "821. Residential/img640_384.jpg- Pred: 8: Forest\n",
            "822. Residential/img640_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "823. Residential/img640_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "824. Residential/img640_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "825. Residential/img640_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "826. Residential/img640_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "827. Residential/img640_704.jpg- Pred: 0: Highway\n",
            "828. Residential/img640_768.jpg- Pred: 7: Pasture\n",
            "829. Residential/img640_832.jpg- Pred: 1: AnnualCrop\n",
            "830. Residential/img640_896.jpg- Pred: 1: AnnualCrop\n",
            "831. Residential/img640_960.jpg- Pred: 6: River\n",
            "832. Residential/img64_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "833. Residential/img64_1024.jpg- Pred: 7: Pasture\n",
            "834. Residential/img64_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "835. Residential/img64_1152.jpg- Pred: 2: HerbaceousVegetation\n",
            "836. Residential/img64_1216.jpg- Pred: 2: HerbaceousVegetation\n",
            "837. Residential/img64_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "838. Residential/img64_1280.jpg- Pred: 2: HerbaceousVegetation\n",
            "839. Residential/img64_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "840. Residential/img64_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "841. Residential/img64_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "842. Residential/img64_1536.jpg- Pred: 5: PermanentCrop\n",
            "843. Residential/img64_1600.jpg- Pred: 3: Industrial\n",
            "844. Residential/img64_1664.jpg- Pred: 5: PermanentCrop\n",
            "845. Residential/img64_1728.jpg- Pred: 4: Residential\n",
            "846. Residential/img64_1792.jpg- Pred: 4: Residential\n",
            "847. Residential/img64_1856.jpg- Pred: 2: HerbaceousVegetation\n",
            "848. Residential/img64_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "849. Residential/img64_1920.jpg- Pred: 2: HerbaceousVegetation\n",
            "850. Residential/img64_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "851. Residential/img64_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "852. Residential/img64_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "853. Residential/img64_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "854. Residential/img64_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "855. Residential/img64_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "856. Residential/img64_576.jpg- Pred: 7: Pasture\n",
            "857. Residential/img64_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "858. Residential/img64_640.jpg- Pred: 0: Highway\n",
            "859. Residential/img64_704.jpg- Pred: 7: Pasture\n",
            "860. Residential/img64_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "861. Residential/img64_832.jpg- Pred: 0: Highway\n",
            "862. Residential/img64_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "863. Residential/img64_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "864. Residential/img704_0.jpg- Pred: 8: Forest\n",
            "865. Residential/img704_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "866. Residential/img704_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "867. Residential/img704_1152.jpg- Pred: 5: PermanentCrop\n",
            "868. Residential/img704_1216.jpg- Pred: 5: PermanentCrop\n",
            "869. Residential/img704_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "870. Residential/img704_1280.jpg- Pred: 5: PermanentCrop\n",
            "871. Residential/img704_1344.jpg- Pred: 2: HerbaceousVegetation\n",
            "872. Residential/img704_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "873. Residential/img704_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "874. Residential/img704_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "875. Residential/img704_1600.jpg- Pred: 8: Forest\n",
            "876. Residential/img704_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "877. Residential/img704_1728.jpg- Pred: 4: Residential\n",
            "878. Residential/img704_1792.jpg- Pred: 5: PermanentCrop\n",
            "879. Residential/img704_1856.jpg- Pred: 3: Industrial\n",
            "880. Residential/img704_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "881. Residential/img704_1920.jpg- Pred: 5: PermanentCrop\n",
            "882. Residential/img704_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "883. Residential/img704_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "884. Residential/img704_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "885. Residential/img704_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "886. Residential/img704_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "887. Residential/img704_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "888. Residential/img704_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "889. Residential/img704_64.jpg- Pred: 8: Forest\n",
            "890. Residential/img704_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "891. Residential/img704_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "892. Residential/img704_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "893. Residential/img704_832.jpg- Pred: 0: Highway\n",
            "894. Residential/img704_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "895. Residential/img704_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "896. Residential/img768_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "897. Residential/img768_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "898. Residential/img768_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "899. Residential/img768_1152.jpg- Pred: 6: River\n",
            "900. Residential/img768_1216.jpg- Pred: 5: PermanentCrop\n",
            "901. Residential/img768_128.jpg- Pred: 7: Pasture\n",
            "902. Residential/img768_1280.jpg- Pred: 5: PermanentCrop\n",
            "903. Residential/img768_1344.jpg- Pred: 7: Pasture\n",
            "904. Residential/img768_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "905. Residential/img768_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "906. Residential/img768_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "907. Residential/img768_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "908. Residential/img768_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "909. Residential/img768_1728.jpg- Pred: 4: Residential\n",
            "910. Residential/img768_1792.jpg- Pred: 3: Industrial\n",
            "911. Residential/img768_1856.jpg- Pred: 4: Residential\n",
            "912. Residential/img768_192.jpg- Pred: 0: Highway\n",
            "913. Residential/img768_1920.jpg- Pred: 5: PermanentCrop\n",
            "914. Residential/img768_1984.jpg- Pred: 3: Industrial\n",
            "915. Residential/img768_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "916. Residential/img768_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "917. Residential/img768_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "918. Residential/img768_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "919. Residential/img768_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "920. Residential/img768_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "921. Residential/img768_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "922. Residential/img768_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "923. Residential/img768_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "924. Residential/img768_768.jpg- Pred: 7: Pasture\n",
            "925. Residential/img768_832.jpg- Pred: 7: Pasture\n",
            "926. Residential/img768_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "927. Residential/img768_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "928. Residential/img832_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "929. Residential/img832_1024.jpg- Pred: 9: SeaLake\n",
            "930. Residential/img832_1088.jpg- Pred: 0: Highway\n",
            "931. Residential/img832_1152.jpg- Pred: 6: River\n",
            "932. Residential/img832_1216.jpg- Pred: 1: AnnualCrop\n",
            "933. Residential/img832_128.jpg- Pred: 0: Highway\n",
            "934. Residential/img832_1280.jpg- Pred: 5: PermanentCrop\n",
            "935. Residential/img832_1344.jpg- Pred: 6: River\n",
            "936. Residential/img832_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "937. Residential/img832_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "938. Residential/img832_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "939. Residential/img832_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "940. Residential/img832_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "941. Residential/img832_1728.jpg- Pred: 5: PermanentCrop\n",
            "942. Residential/img832_1792.jpg- Pred: 5: PermanentCrop\n",
            "943. Residential/img832_1856.jpg- Pred: 5: PermanentCrop\n",
            "944. Residential/img832_192.jpg- Pred: 6: River\n",
            "945. Residential/img832_1920.jpg- Pred: 4: Residential\n",
            "946. Residential/img832_1984.jpg- Pred: 5: PermanentCrop\n",
            "947. Residential/img832_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "948. Residential/img832_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "949. Residential/img832_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "950. Residential/img832_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "951. Residential/img832_512.jpg- Pred: 2: HerbaceousVegetation\n",
            "952. Residential/img832_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "953. Residential/img832_64.jpg- Pred: 7: Pasture\n",
            "954. Residential/img832_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "955. Residential/img832_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "956. Residential/img832_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "957. Residential/img832_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "958. Residential/img832_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "959. Residential/img832_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "960. Residential/img896_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "961. Residential/img896_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "962. Residential/img896_1088.jpg- Pred: 2: HerbaceousVegetation\n",
            "963. Residential/img896_1152.jpg- Pred: 5: PermanentCrop\n",
            "964. Residential/img896_1216.jpg- Pred: 5: PermanentCrop\n",
            "965. Residential/img896_128.jpg- Pred: 6: River\n",
            "966. Residential/img896_1280.jpg- Pred: 0: Highway\n",
            "967. Residential/img896_1344.jpg- Pred: 0: Highway\n",
            "968. Residential/img896_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "969. Residential/img896_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "970. Residential/img896_1536.jpg- Pred: 2: HerbaceousVegetation\n",
            "971. Residential/img896_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "972. Residential/img896_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "973. Residential/img896_1728.jpg- Pred: 4: Residential\n",
            "974. Residential/img896_1792.jpg- Pred: 5: PermanentCrop\n",
            "975. Residential/img896_1856.jpg- Pred: 5: PermanentCrop\n",
            "976. Residential/img896_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "977. Residential/img896_1920.jpg- Pred: 0: Highway\n",
            "978. Residential/img896_1984.jpg- Pred: 0: Highway\n",
            "979. Residential/img896_256.jpg- Pred: 0: Highway\n",
            "980. Residential/img896_320.jpg- Pred: 2: HerbaceousVegetation\n",
            "981. Residential/img896_384.jpg- Pred: 7: Pasture\n",
            "982. Residential/img896_448.jpg- Pred: 2: HerbaceousVegetation\n",
            "983. Residential/img896_512.jpg- Pred: 8: Forest\n",
            "984. Residential/img896_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "985. Residential/img896_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "986. Residential/img896_640.jpg- Pred: 2: HerbaceousVegetation\n",
            "987. Residential/img896_704.jpg- Pred: 2: HerbaceousVegetation\n",
            "988. Residential/img896_768.jpg- Pred: 2: HerbaceousVegetation\n",
            "989. Residential/img896_832.jpg- Pred: 2: HerbaceousVegetation\n",
            "990. Residential/img896_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "991. Residential/img896_960.jpg- Pred: 2: HerbaceousVegetation\n",
            "992. Residential/img960_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "993. Residential/img960_1024.jpg- Pred: 2: HerbaceousVegetation\n",
            "994. Residential/img960_1088.jpg- Pred: 0: Highway\n",
            "995. Residential/img960_1152.jpg- Pred: 0: Highway\n",
            "996. Residential/img960_1216.jpg- Pred: 1: AnnualCrop\n",
            "997. Residential/img960_128.jpg- Pred: 6: River\n",
            "998. Residential/img960_1280.jpg- Pred: 6: River\n",
            "999. Residential/img960_1344.jpg- Pred: 0: Highway\n",
            "1000. Residential/img960_1408.jpg- Pred: 2: HerbaceousVegetation\n",
            "1001. Residential/img960_1472.jpg- Pred: 2: HerbaceousVegetation\n",
            "1002. Residential/img960_1536.jpg- Pred: 6: River\n",
            "1003. Residential/img960_1600.jpg- Pred: 2: HerbaceousVegetation\n",
            "1004. Residential/img960_1664.jpg- Pred: 2: HerbaceousVegetation\n",
            "1005. Residential/img960_1728.jpg- Pred: 3: Industrial\n",
            "1006. Residential/img960_1792.jpg- Pred: 5: PermanentCrop\n",
            "1007. Residential/img960_1856.jpg- Pred: 5: PermanentCrop\n",
            "1008. Residential/img960_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "1009. Residential/img960_1920.jpg- Pred: 5: PermanentCrop\n",
            "1010. Residential/img960_1984.jpg- Pred: 2: HerbaceousVegetation\n",
            "1011. Residential/img960_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "1012. Residential/img960_320.jpg- Pred: 1: AnnualCrop\n",
            "1013. Residential/img960_384.jpg- Pred: 2: HerbaceousVegetation\n",
            "1014. Residential/img960_448.jpg- Pred: 6: River\n",
            "1015. Residential/img960_512.jpg- Pred: 6: River\n",
            "1016. Residential/img960_576.jpg- Pred: 2: HerbaceousVegetation\n",
            "1017. Residential/img960_64.jpg- Pred: 9: SeaLake\n",
            "1018. Residential/img960_640.jpg- Pred: 9: SeaLake\n",
            "1019. Residential/img960_704.jpg- Pred: 9: SeaLake\n",
            "1020. Residential/img960_768.jpg- Pred: 6: River\n",
            "1021. Residential/img960_832.jpg- Pred: 0: Highway\n",
            "1022. Residential/img960_896.jpg- Pred: 2: HerbaceousVegetation\n",
            "1023. Residential/img960_960.jpg- Pred: 2: HerbaceousVegetation\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "TusLU-NyTSzK",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Siteler"
      ]
    },
    {
      "metadata": {
        "id": "9-trd0etTCik",
        "colab_type": "code",
        "outputId": "f0de109c-eedb-4458-e343-42660a936820",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "siteler_generator = ImageDataGenerator().flow_from_directory(\n",
        "    'drive/My Drive/PhD/Data/Sat/siteler',\n",
        "    target_size=(img_width, img_height),\n",
        "    classes=\n",
        "    [\n",
        "        'Highway',\n",
        "        'AnnualCrop',\n",
        "        'HerbaceousVegetation',\n",
        "        'Industrial',\n",
        "        'Residential',\n",
        "        'PermanentCrop',\n",
        "        'River',\n",
        "        'Pasture',\n",
        "        'Forest',\n",
        "        'SeaLake'\n",
        "    ],\n",
        "    batch_size=batch_size,\n",
        "    shuffle=False,)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 21 images belonging to 10 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "pUTZTcCEUPGS",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "siteler_pred = model.predict_generator(siteler_generator, steps=math.ceil(siteler_generator.samples / batch_size))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "q6xFUWSlhf4-",
        "colab_type": "code",
        "outputId": "9e074b70-022d-4082-e4d5-488ebbfcf04a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "siteler_pred[0]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([4.4296906e-04, 1.5280736e-16, 1.3828640e-04, 9.9941373e-01,\n",
              "       2.5311121e-22, 4.8604770e-06, 9.9239706e-08, 2.0476219e-16,\n",
              "       4.4385372e-34, 1.2066972e-21], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "metadata": {
        "id": "3cdBW0tlTfuq",
        "colab_type": "code",
        "outputId": "38b7d0c0-4c8f-4542-8bdc-57669708ea01",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        }
      },
      "cell_type": "code",
      "source": [
        "for i in range(0, len(siteler_pred)):\n",
        "  print(str(i) + \". \" + siteler_generator.filenames[i] + \"- Pred: \" + str(np.argmax(siteler_pred[i])) + \": \" \n",
        "        + list(siteler_generator.class_indices.keys())[list(siteler_generator.class_indices.values()).index(np.argmax(siteler_pred[i]))])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0. Industrial/armada.jpg- Pred: 3: Industrial\n",
            "1. Industrial/asti.jpg- Pred: 3: Industrial\n",
            "2. Industrial/cepa.jpg- Pred: 5: PermanentCrop\n",
            "3. Industrial/teknokent.jpg- Pred: 3: Industrial\n",
            "4. Industrial/ulus.jpg- Pred: 4: Residential\n",
            "5. Industrial/yurtlar.jpg- Pred: 2: HerbaceousVegetation\n",
            "6. Residential/img0_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "7. Residential/img0_128.jpg- Pred: 2: HerbaceousVegetation\n",
            "8. Residential/img0_192.jpg- Pred: 5: PermanentCrop\n",
            "9. Residential/img0_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "10. Residential/img0_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "11. Residential/img128_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "12. Residential/img128_128.jpg- Pred: 3: Industrial\n",
            "13. Residential/img128_192.jpg- Pred: 2: HerbaceousVegetation\n",
            "14. Residential/img128_256.jpg- Pred: 3: Industrial\n",
            "15. Residential/img128_64.jpg- Pred: 2: HerbaceousVegetation\n",
            "16. Residential/img64_0.jpg- Pred: 2: HerbaceousVegetation\n",
            "17. Residential/img64_128.jpg- Pred: 4: Residential\n",
            "18. Residential/img64_192.jpg- Pred: 4: Residential\n",
            "19. Residential/img64_256.jpg- Pred: 2: HerbaceousVegetation\n",
            "20. Residential/img64_64.jpg- Pred: 2: HerbaceousVegetation\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "1LaHMMqKUTaI",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}