test_model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "test_model.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "bgyOXyZVclmU"
      },
      "source": [
        "from glob import glob\n",
        "import torch\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import json\n",
        "from torchvision import datasets, models, transforms\n",
        "from PIL import Image\n",
        "from torch.autograd import Variable\n",
        "from torch import nn, optim, cuda\n",
        "import cv2\n",
        "from skimage.transform import rescale, resize\n",
        "from skimage import measure\n",
        "from shapely.geometry import Polygon, MultiPolygon\n",
        "from shutil import copyfile"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0e9lHCZuc7rY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "953d97ac-10ec-4c0a-f9b9-b47c5589e568"
      },
      "source": [
        "device = torch.device(\"cuda\" if torch.cuda.is_available() \n",
        "                                  else \"cpu\")\n",
        "print(device)\n",
        "\n",
        "from google.colab import drive\n",
        "\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "cuda\n",
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MrW_DsLnJFf7"
      },
      "source": [
        "# **Get Model**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BXYso5irdB34"
      },
      "source": [
        "model_version = \"alex_8\"\n",
        "params_file = model_version + '_params.pth' #model version to use\n",
        "\n",
        "#get model\n",
        "model_dir = '/content/drive/My Drive/Lake detection/models/'\n",
        "\n",
        "params = torch.load(model_dir + params_file)\n",
        "model_name = params['name']\n",
        "model = torch.load(model_dir + model_name)\n",
        "\n",
        "for param in model.parameters():\n",
        "    param.requires_grad = False\n",
        "\n",
        "model.eval();\n",
        "\n",
        "preprocess = transforms.Compose([transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406], \n",
        "                             [0.229, 0.224, 0.225])\n",
        "                                     ])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DUA66qpFOHGo"
      },
      "source": [
        "def get_prediction(probs, th):\n",
        "\n",
        "    if np.asarray(probs.cpu()[0])[4] > th:\n",
        "        predicted = 4\n",
        "    else:\n",
        "        pred = np.asarray(probs.cpu()[0])\n",
        "        pred[4] = 0\n",
        "        predicted = np.argmax(pred)\n",
        "\n",
        "    return predicted"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Udu0k7XJKUJ"
      },
      "source": [
        "# **Classify tiles**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DMawRV8XTHX1"
      },
      "source": [
        "th = 0.7\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ElOIW_2LwI6T",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "adbf7d30-77f7-4a4e-d140-6b7eff766892"
      },
      "source": [
        "#ice test\n",
        "clas = 6\n",
        "types = {0:'smooth ice', 1:'textured ice', 2:'remnant', 3:'water', 4:'subsurface', 5:'mountains', 6:'land'}\n",
        "print(types[clas])\n",
        "ice_imgs = glob('/content/drive/My Drive/Lake detection/greenland/test/' + str(clas) + '/*.jpg')\n",
        "ice = 0\n",
        "ice2 = 0\n",
        "sub = 0\n",
        "rem = 0\n",
        "wat = 0 \n",
        "mtn = 0\n",
        "land = 0\n",
        "\n",
        "for img_path in ice_imgs:\n",
        "    img = Image.open(img_path)\n",
        "    img = preprocess(img)\n",
        "    img2 = np.expand_dims(img, 0) #Convert 2D image to 1D vector\n",
        "    img2 = torch.from_numpy(img2)\n",
        "\n",
        "    #classify image\n",
        "    inputs = Variable(img2).to(device)\n",
        "    outputs = model(inputs) #labels\n",
        "    probabilites = torch.softmax(outputs.data, 1) #get probability for each class\n",
        "    pred = get_prediction(probabilites, th)\n",
        "    if pred == 0:\n",
        "      ice = ice + 1\n",
        "    elif pred == 1:\n",
        "      ice2 = ice2 + 1\n",
        "    elif pred == 2:\n",
        "      rem = rem + 1\n",
        "    elif pred == 3:\n",
        "      wat = wat + 1\n",
        "    elif pred == 4:\n",
        "      sub = sub + 1\n",
        "    elif pred == 5:\n",
        "      mtn = mtn + 1\n",
        "    elif pred == 6:\n",
        "      land = land + 1\n",
        "\n",
        "print(f'classified as smooth ice: {ice}')\n",
        "print(f'classified as textured ice: {ice2}')\n",
        "print(f'classified as subsurface: {sub}')\n",
        "print(f'classified as remnant: {rem}')\n",
        "print(f'classified as water: {wat}')\n",
        "print(f'classified as mtn: {mtn}')\n",
        "print(f'classified as land: {land}')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "land\n",
            "classified as smooth ice: 1\n",
            "classified as textured ice: 1\n",
            "classified as subsurface: 0\n",
            "classified as remnant: 1\n",
            "classified as water: 0\n",
            "classified as mtn: 0\n",
            "classified as land: 31\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "35rYpaF1ADUn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "31661b1b-4455-4450-8564-421167de732c"
      },
      "source": [
        "precision = 184/(184+14)\n",
        "recall = 184/(184+25)\n",
        "F1 = 2*precision*recall/(precision + recall)\n",
        "print(precision, recall, F1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.9292929292929293 0.8803827751196173 0.9041769041769042\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9tsvDR015kgl"
      },
      "source": [
        "y_true = list()\n",
        "y_pred = list()\n",
        "for i in range(172):\n",
        "  y_true.append(0)\n",
        "  y_pred.append(0)\n",
        "for i in range(6):\n",
        "  y_true.append(1)\n",
        "  y_pred.append(0)\n",
        "y_true.append(2)\n",
        "y_pred.append(0)\n",
        "y_true.append(6)\n",
        "y_pred.append(0)\n",
        "for i in range(16):\n",
        "  y_true.append(0)\n",
        "  y_pred.append(1)\n",
        "for i in range(110):\n",
        "  y_true.append(1)\n",
        "  y_pred.append(1)\n",
        "for i in range(17):\n",
        "  y_true.append(2)\n",
        "  y_pred.append(1)\n",
        "for i in range(16):\n",
        "  y_true.append(4)\n",
        "  y_pred.append(1)\n",
        "y_true.append(5)\n",
        "y_pred.append(1)\n",
        "y_true.append(6)\n",
        "y_pred.append(1)\n",
        "y_true.append(0)\n",
        "y_pred.append(2)\n",
        "for i in range(12):\n",
        "  y_true.append(1)\n",
        "  y_pred.append(2)\n",
        "for i in range(195):\n",
        "  y_true.append(2)\n",
        "  y_pred.append(2)\n",
        "for i in range(7):\n",
        "  y_true.append(4)\n",
        "  y_pred.append(2)\n",
        "for i in range(14):\n",
        "  y_true.append(3)\n",
        "  y_pred.append(3)\n",
        "y_true.append(6)\n",
        "y_pred.append(2)\n",
        "for i in range(2):\n",
        "  y_true.append(0)\n",
        "  y_pred.append(4)\n",
        "for i in range(7):\n",
        "  y_true.append(1)\n",
        "  y_pred.append(4)\n",
        "for i in range(5):\n",
        "  y_true.append(2)\n",
        "  y_pred.append(4)\n",
        "for i in range(184):\n",
        "  y_true.append(4)\n",
        "  y_pred.append(4)\n",
        "for i in range(3):\n",
        "  y_true.append(1)\n",
        "  y_pred.append(5)\n",
        "for i in range(2):\n",
        "  y_true.append(4)\n",
        "  y_pred.append(5)\n",
        "y_true.append(2)\n",
        "y_pred.append(5)\n",
        "for i in range(76):\n",
        "  y_true.append(5)\n",
        "  y_pred.append(5)\n",
        "for i in range(2):\n",
        "  y_true.append(2)\n",
        "  y_pred.append(6)\n",
        "for i in range(4):\n",
        "  y_true.append(5)\n",
        "  y_pred.append(6)\n",
        "for i in range(31):\n",
        "  y_true.append(6)\n",
        "  y_pred.append(6)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wbRmvp1y_nY0"
      },
      "source": [
        "import seaborn as sns\n",
        "from sklearn.metrics import confusion_matrix"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 742
        },
        "id": "lKMs0QkA9ZTV",
        "outputId": "130c9824-7cd9-42b4-d96b-c908225c87bb"
      },
      "source": [
        "\n",
        "labels = ['smooth \\nice', 'textured \\nice', 'remnant \\nlake', 'ocean', 'buried \\nlake', 'land', 'mountain']\n",
        "c = confusion_matrix(y_true, y_pred, normalize = 'pred')\n",
        "\n",
        "fig, ax = plt.subplots(1,1,figsize = (12,10))\n",
        "sns.heatmap(c, annot=True, fmt = \".1%\", cmap = 'Greens',cbar = False,square = True,linecolor = 'gray', linewidths = 1, annot_kws={'fontsize':'large'})\n",
        "ax.set_xlabel('Predicted', fontsize = 16)\n",
        "ax.set_ylabel('Actual', fontsize = 16)\n",
        "ax.set_xticklabels(labels, fontsize = 12)\n",
        "ax.set_yticklabels(labels, fontsize = 12, rotation = 00, va=\"center\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(0, 0.5, 'smooth \\nice'),\n",
              " Text(0, 1.5, 'textured \\nice'),\n",
              " Text(0, 2.5, 'remnant \\nlake'),\n",
              " Text(0, 3.5, 'ocean'),\n",
              " Text(0, 4.5, 'buried \\nlake'),\n",
              " Text(0, 5.5, 'land'),\n",
              " Text(0, 6.5, 'mountain')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 77
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZQjQyz8K9_Yc"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}