Plant_Breed_Final_Solution (3).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Plant_Breed_Final_Solution.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "YaBr-erZncce",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from google.colab import drive\n",
        "from sklearn import preprocessing"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gF5meo6Pnd-y",
        "colab_type": "code",
        "outputId": "4e3f0928-ef91-4e11-8e59-0d649af5b1de",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "drive.mount('/content/drive')"
      ],
      "execution_count": 187,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "--D8rvSffrll",
        "colab_type": "text"
      },
      "source": [
        "# Read the images and generate the train and test dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ivJgjcGSnf-C",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import os\n",
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XuE6AcnM2Z1W",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#import pdb; pdb.set_trace()\n",
        "from zipfile import ZipFile\n",
        "with ZipFile('train.zip', 'r') as z:\n",
        "  z.extractall()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OudDEvpd2djj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from zipfile import ZipFile\n",
        "with ZipFile('test.zip', 'r') as z:\n",
        "  z.extractall()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QsPoSsZ3nn2B",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test=[]\n",
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed/test')\n",
        "import cv2\n",
        "for i in os.listdir():\n",
        "  try:\n",
        "    dummy = cv2.imread(i)\n",
        "    dummy = cv2.resize(dummy,(128,128))\n",
        "    x_test.append(dummy)\n",
        "  except Exception as e:\n",
        "    print(e)\n",
        "    "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CzKDiNr_6F_C",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test_orig = x_test"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4CQVYV_4aP-b",
        "colab_type": "code",
        "outputId": "5afb1341-0d11-431a-c3cd-f36081e38580",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "os.listdir()"
      ],
      "execution_count": 193,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['ef74dbcad.png',\n",
              " '699d3c707.png',\n",
              " '8ab8a958b.png',\n",
              " '486e59179.png',\n",
              " '7fdb7202d.png',\n",
              " '0a64e3e6c.png',\n",
              " 'c10db7ae2.png',\n",
              " '56112b92c.png',\n",
              " '9d79a1f0c.png',\n",
              " 'abf8b0772.png',\n",
              " 'f8318faf1.png',\n",
              " 'a254d71f6.png',\n",
              " '32b42c120.png',\n",
              " 'b4c3df835.png',\n",
              " '4a337a4a9.png',\n",
              " '98062cd87.png',\n",
              " '7d3045fc3.png',\n",
              " 'b7ad92859.png',\n",
              " '085974290.png',\n",
              " 'ede6b84b4.png',\n",
              " '90b595f12.png',\n",
              " '3d65168c2.png',\n",
              " '96f14d90c.png',\n",
              " 'e5064f6be.png',\n",
              " '5b63dcc21.png',\n",
              " '1c680883c.png',\n",
              " '756dd5070.png',\n",
              " '24d78df74.png',\n",
              " '8303b27ed.png',\n",
              " 'd2f0f326e.png',\n",
              " 'd488a4fe1.png',\n",
              " '55920f07f.png',\n",
              " 'c88ebfb47.png',\n",
              " 'fdea6b119.png',\n",
              " 'cb496f36e.png',\n",
              " '8cf2e3e6c.png',\n",
              " 'af45e222a.png',\n",
              " '1e095a7e1.png',\n",
              " '1364b297a.png',\n",
              " '338c7e907.png',\n",
              " '38c054379.png',\n",
              " '1694a70e4.png',\n",
              " '79e5ea8fa.png',\n",
              " '01291174f.png',\n",
              " 'c7b07431e.png',\n",
              " 'a060c1cf8.png',\n",
              " '25fa8d109.png',\n",
              " 'fe29629fb.png',\n",
              " 'd9c50616e.png',\n",
              " '67e185673.png',\n",
              " 'd14aa43f3.png',\n",
              " 'bb1d1bfd3.png',\n",
              " 'fea3da57c.png',\n",
              " 'dcd7ff249.png',\n",
              " '37c3108d6.png',\n",
              " '59c6a9f95.png',\n",
              " '721b2c47a.png',\n",
              " '4ea7493d5.png',\n",
              " '86676d627.png',\n",
              " 'dabe3e5be.png',\n",
              " 'cf3a8b2fd.png',\n",
              " '521b27a17.png',\n",
              " '8b043093d.png',\n",
              " '770a265f5.png',\n",
              " '026716f9b.png',\n",
              " 'a7bd7cadb.png',\n",
              " '2df78338c.png',\n",
              " '39b740f7e.png',\n",
              " '3185294c8.png',\n",
              " '589e643b8.png',\n",
              " 'a8de7c1b7.png',\n",
              " '8faadb6a8.png',\n",
              " 'a35fd6fbb.png',\n",
              " '75cb95e91.png',\n",
              " 'bea23d9f8.png',\n",
              " '5b3beec58.png',\n",
              " '3526b05cc.png',\n",
              " '59f62ad1d.png',\n",
              " '1623fb9e1.png',\n",
              " 'b130a0632.png',\n",
              " '2d5058a59.png',\n",
              " 'b3e08b037.png',\n",
              " 'd0cdc768f.png',\n",
              " 'c74c5b7fc.png',\n",
              " '90d119d25.png',\n",
              " '1758a1baf.png',\n",
              " 'e478c452c.png',\n",
              " '9aa5587fe.png',\n",
              " '592473c83.png',\n",
              " '6b721f68e.png',\n",
              " 'cc74feadc.png',\n",
              " '99569b224.png',\n",
              " 'bf66b9cd2.png',\n",
              " 'da9ef7858.png',\n",
              " 'b39c71707.png',\n",
              " 'c50335991.png',\n",
              " '17529c555.png',\n",
              " '239bdf640.png',\n",
              " '675ec1b0b.png',\n",
              " 'e14afa235.png',\n",
              " '8ece6efec.png',\n",
              " 'd41d87796.png',\n",
              " '6db684fff.png',\n",
              " '8db450ce3.png',\n",
              " '7d4cd07ad.png',\n",
              " 'b29339405.png',\n",
              " 'ac3193f78.png',\n",
              " 'a2d25b4f3.png',\n",
              " '1926e82fd.png',\n",
              " '48d97c645.png',\n",
              " 'fc6f686fb.png',\n",
              " 'e5e3dccff.png',\n",
              " 'b47691c08.png',\n",
              " '65e97117e.png',\n",
              " 'e5297b675.png',\n",
              " '7beb2766f.png',\n",
              " 'a3b375b34.png',\n",
              " '6bce55e05.png',\n",
              " '7f31c7f42.png',\n",
              " '39d569be4.png',\n",
              " 'c35efa095.png',\n",
              " '9643fc5f4.png',\n",
              " '4287d810c.png',\n",
              " '67ce3eaa6.png',\n",
              " '88ac6df54.png',\n",
              " '8b27bfd2b.png',\n",
              " '4823c3ffa.png',\n",
              " '5a6bf96f6.png',\n",
              " '59b2c6f2b.png',\n",
              " '9c777333d.png',\n",
              " 'b0acaff4a.png',\n",
              " '1312065a5.png',\n",
              " '3f64c2c1b.png',\n",
              " '606647f64.png',\n",
              " 'c7051c902.png',\n",
              " '728eabae1.png',\n",
              " '55251925f.png',\n",
              " '7f9e9565d.png',\n",
              " '391dcd7fd.png',\n",
              " 'ff65bc002.png',\n",
              " '1fefb54b7.png',\n",
              " 'd668409ff.png',\n",
              " '0625f063b.png',\n",
              " 'a83820a2c.png',\n",
              " '5bbc0a255.png',\n",
              " '36ed4f215.png',\n",
              " '406162ef9.png',\n",
              " '668c1007c.png',\n",
              " '862b8e7a0.png',\n",
              " '7cabd68cc.png',\n",
              " 'cae684f8f.png',\n",
              " 'c60e91e07.png',\n",
              " '65e262a6d.png',\n",
              " 'f9ea23fb5.png',\n",
              " 'ab0f67743.png',\n",
              " '615d2b0a9.png',\n",
              " 'd0152bd7c.png',\n",
              " '9c0c5b731.png',\n",
              " 'f48916a8c.png',\n",
              " 'aa7d098d1.png',\n",
              " '56a01b835.png',\n",
              " 'e783f5a4f.png',\n",
              " '54b3afd58.png',\n",
              " '099b961ec.png',\n",
              " '785a73ab8.png',\n",
              " '79d93bc96.png',\n",
              " 'a4b61a4ea.png',\n",
              " '79fba50db.png',\n",
              " 'a55d26a4c.png',\n",
              " '37e545a60.png',\n",
              " '36d62bf36.png',\n",
              " 'e80a259c5.png',\n",
              " '23e480e64.png',\n",
              " '3a909ead8.png',\n",
              " '3edf5e9ef.png',\n",
              " '71334c634.png',\n",
              " '4c5ab9b68.png',\n",
              " 'e1a0e3202.png',\n",
              " '8dbb8e1b9.png',\n",
              " 'c06e7c748.png',\n",
              " '4b155fb07.png',\n",
              " '6d6eb3830.png',\n",
              " 'fbf88b6be.png',\n",
              " 'dce2f6612.png',\n",
              " 'a85fc8c9a.png',\n",
              " '00c47e980.png',\n",
              " 'c5e88cd42.png',\n",
              " 'ef02b4ee7.png',\n",
              " '1541bdb2e.png',\n",
              " '8301b0547.png',\n",
              " '32a8c8a1d.png',\n",
              " '9c8b08a24.png',\n",
              " '5bc6595f6.png',\n",
              " 'b1cd2a91e.png',\n",
              " 'a2c89c367.png',\n",
              " '241e6935a.png',\n",
              " '61b044411.png',\n",
              " '8311740de.png',\n",
              " '08d591441.png',\n",
              " '855955aaf.png',\n",
              " '87608f7aa.png',\n",
              " '14bb43eee.png',\n",
              " 'ce3d280eb.png',\n",
              " '5883b423d.png',\n",
              " '65489944f.png',\n",
              " '7506c0c02.png',\n",
              " '1dc7c45df.png',\n",
              " 'bb1c84bbc.png',\n",
              " 'e19ad6ac9.png',\n",
              " '8ca6140ca.png',\n",
              " '31fcdc161.png',\n",
              " '653193c1a.png',\n",
              " '30ad31220.png',\n",
              " '99036c51d.png',\n",
              " '205df1df3.png',\n",
              " '8e4eaeec0.png',\n",
              " '77ccb8b2a.png',\n",
              " '4bcdaa5e2.png',\n",
              " '98da6ef4e.png',\n",
              " '4392d93cf.png',\n",
              " '03ef36742.png',\n",
              " 'd01873fdd.png',\n",
              " 'c8f50f0c3.png',\n",
              " '837ac0270.png',\n",
              " '8f523520c.png',\n",
              " 'a1da8be3c.png',\n",
              " '9c32a797e.png',\n",
              " 'da4ed3a28.png',\n",
              " '9326bda1b.png',\n",
              " 'b573b7a56.png',\n",
              " 'f4ad9d950.png',\n",
              " '071cb3ece.png',\n",
              " 'fda0b5c38.png',\n",
              " '647689543.png',\n",
              " 'cbe761896.png',\n",
              " 'a935ca110.png',\n",
              " '85431c075.png',\n",
              " 'f85ed9b6d.png',\n",
              " 'b7192c70f.png',\n",
              " '3827436f3.png',\n",
              " 'cd6adba97.png',\n",
              " 'c1ecff98b.png',\n",
              " '604dd663f.png',\n",
              " 'df11d56a7.png',\n",
              " '25cf6eb73.png',\n",
              " 'eaf0815e2.png',\n",
              " 'e901b0f28.png',\n",
              " '34dd57ca9.png',\n",
              " '78b1bf91a.png',\n",
              " '2ea664465.png',\n",
              " 'f3fcfff1b.png',\n",
              " '165681fd9.png',\n",
              " '0437393b1.png',\n",
              " 'd2f422ccb.png',\n",
              " 'f23faf9c1.png',\n",
              " '3da774107.png',\n",
              " '007b3da8b.png',\n",
              " '82b5f4d33.png',\n",
              " '8bc0261c9.png',\n",
              " '46c14fde2.png',\n",
              " '892e9d6c6.png',\n",
              " 'b5c7fd009.png',\n",
              " 'c2de6020a.png',\n",
              " 'caa2fbd79.png',\n",
              " '2a5064f19.png',\n",
              " '5b3000b9a.png',\n",
              " 'b828443ff.png',\n",
              " 'dc4cd56a3.png',\n",
              " 'd350a25fa.png',\n",
              " '248436078.png',\n",
              " 'bb7621cb3.png',\n",
              " '377283a21.png',\n",
              " '5ca0205f9.png',\n",
              " 'd09d24c58.png',\n",
              " '3f92d8039.png',\n",
              " 'fea355851.png',\n",
              " '1b490196c.png',\n",
              " 'abc331628.png',\n",
              " 'f25996db8.png',\n",
              " '26852751a.png',\n",
              " '223e4af09.png',\n",
              " '060450d79.png',\n",
              " '04814f36d.png',\n",
              " '060f1dc84.png',\n",
              " 'e471f1d3a.png',\n",
              " '39858776a.png',\n",
              " 'fe9e87b78.png',\n",
              " '6c874918c.png',\n",
              " 'b4f7c9214.png',\n",
              " '5e6a237f2.png',\n",
              " 'df7cb5f87.png',\n",
              " '2e86f1085.png',\n",
              " '897e5a8de.png',\n",
              " 'a8da9c08d.png',\n",
              " 'c832e4302.png',\n",
              " '558aa7deb.png',\n",
              " 'ef7a5651d.png',\n",
              " '4049a6ea6.png',\n",
              " 'd102e1a15.png',\n",
              " '3d38a87bc.png',\n",
              " '87f627bf9.png',\n",
              " '1191ba346.png',\n",
              " 'a38b8a581.png',\n",
              " '88d8a4508.png',\n",
              " 'b6f3d8b5d.png',\n",
              " '1d48b7564.png',\n",
              " '11d3f68ff.png',\n",
              " 'de0b79659.png',\n",
              " 'cfd8165e9.png',\n",
              " '43ede9de9.png',\n",
              " '8e6ec1ca6.png',\n",
              " '19e58cc5e.png',\n",
              " '2d9c798f9.png',\n",
              " '74068643d.png',\n",
              " '0d117d910.png',\n",
              " '24d36c52c.png',\n",
              " 'd7017f701.png',\n",
              " 'b30ab4659.png',\n",
              " '172f9b10b.png',\n",
              " '5af1d74ee.png',\n",
              " '55a852f40.png',\n",
              " '16fd2e01a.png',\n",
              " 'fda39e16f.png',\n",
              " '0021e90e4.png',\n",
              " 'da713c465.png',\n",
              " '0ae6668fa.png',\n",
              " '969a851be.png',\n",
              " 'a74bf916d.png',\n",
              " '686dc7ec8.png',\n",
              " '4f83143e1.png',\n",
              " '6a47821f9.png',\n",
              " '8455169fe.png',\n",
              " '6edc76e7c.png',\n",
              " '288564c76.png',\n",
              " '126a71ce0.png',\n",
              " '0c51bf229.png',\n",
              " '0dba99002.png',\n",
              " '93d76fd5d.png',\n",
              " 'ce15eee52.png',\n",
              " 'f445fe6fb.png',\n",
              " 'a93f940d6.png',\n",
              " '3eebd36c6.png',\n",
              " 'f4234cf4f.png',\n",
              " 'fd925f542.png',\n",
              " 'be2499cf4.png',\n",
              " '780bd2a2c.png',\n",
              " 'fef2ade8c.png',\n",
              " '593896f83.png',\n",
              " '6982a9d30.png',\n",
              " 'a3d0031fd.png',\n",
              " 'e96e57a90.png',\n",
              " '2406d6c99.png',\n",
              " '599c82eea.png',\n",
              " 'c4ed8ed38.png',\n",
              " '29f49cd0b.png',\n",
              " 'f6d250856.png',\n",
              " '446f7da01.png',\n",
              " 'd17f48d3b.png',\n",
              " 'b62dca166.png',\n",
              " 'b145ba9d4.png',\n",
              " '995c7ab1e.png',\n",
              " '4e18ab737.png',\n",
              " 'b98327bf4.png',\n",
              " 'c5e419015.png',\n",
              " 'a006a475c.png',\n",
              " 'bd789d151.png',\n",
              " 'ad12382d4.png',\n",
              " 'd6d80a321.png',\n",
              " 'e84464f5a.png',\n",
              " '24a058589.png',\n",
              " '47f9e5d91.png',\n",
              " 'afcf6abd5.png',\n",
              " '3281183f9.png',\n",
              " '2a667e099.png',\n",
              " 'a544fc46d.png',\n",
              " '79dafec17.png',\n",
              " '827279bad.png',\n",
              " 'eef131644.png',\n",
              " 'ba3ce6b3e.png',\n",
              " 'd3331e071.png',\n",
              " 'b687160f5.png',\n",
              " '0911d3dee.png',\n",
              " '44e8b8833.png',\n",
              " '20817c846.png',\n",
              " '0885e7690.png',\n",
              " 'c85ef220d.png',\n",
              " 'd563be369.png',\n",
              " 'aecfaed64.png',\n",
              " '948cdb277.png',\n",
              " '9b4800b42.png',\n",
              " '089ad62a7.png',\n",
              " '13b9fa92d.png',\n",
              " '5c3cd7ea2.png',\n",
              " '31f3dd81f.png',\n",
              " 'ac75d3326.png',\n",
              " '1d56351b2.png',\n",
              " 'afa446484.png',\n",
              " '5779fe8b4.png',\n",
              " '115f93ecc.png',\n",
              " 'a2b703e21.png',\n",
              " 'da5255450.png',\n",
              " 'b9062c1c8.png',\n",
              " '7d22abf91.png',\n",
              " 'd689256be.png',\n",
              " '9df3275da.png',\n",
              " 'dabea05f4.png',\n",
              " '6dd095129.png',\n",
              " 'd09275360.png',\n",
              " 'c63da993b.png',\n",
              " '16467a950.png',\n",
              " 'f0ffa00bd.png',\n",
              " 'da231c97f.png',\n",
              " '8b9144917.png',\n",
              " '632156793.png',\n",
              " '063363305.png',\n",
              " '953496deb.png',\n",
              " '48ef6a2ff.png',\n",
              " '9516e56c4.png',\n",
              " 'e88bf0db9.png',\n",
              " 'e4d5ec761.png',\n",
              " 'efe19dc32.png',\n",
              " '490c4f9c8.png',\n",
              " '599691cd9.png',\n",
              " '752101fdf.png',\n",
              " 'b026bf8ca.png',\n",
              " '9baf94467.png',\n",
              " 'a74d475c2.png',\n",
              " '3fbf1a417.png',\n",
              " '52a87abe5.png',\n",
              " '8c98a6e9b.png',\n",
              " '1bf9b94a6.png',\n",
              " 'cadab6616.png',\n",
              " '17d5e5ac4.png',\n",
              " 'aee6fa3df.png',\n",
              " '7e9cf1c46.png',\n",
              " '19618ad6a.png',\n",
              " '48231e475.png',\n",
              " '1c52ea820.png',\n",
              " '3b73c3b61.png',\n",
              " '33448fe39.png',\n",
              " '19b51843a.png',\n",
              " '4e1190d78.png',\n",
              " 'c0461776c.png',\n",
              " '0c45ace27.png',\n",
              " '4bbfd1e05.png',\n",
              " '851c90831.png',\n",
              " '1d0cbd819.png',\n",
              " 'd93c7ab6d.png',\n",
              " '93079d970.png',\n",
              " '279df95f2.png',\n",
              " '7c85b0265.png',\n",
              " '856f2910a.png',\n",
              " '534e74d83.png',\n",
              " '3e9f41817.png',\n",
              " '976e4e079.png',\n",
              " 'ace8761dd.png',\n",
              " 'd2fd9df40.png',\n",
              " 'b31292706.png',\n",
              " '7615e52d3.png',\n",
              " '54c8bb900.png',\n",
              " '29bab7cad.png',\n",
              " '71b232519.png',\n",
              " '4e69d100a.png',\n",
              " '37297a64c.png',\n",
              " 'c7ae30f3a.png',\n",
              " 'e52493d0b.png',\n",
              " 'f2dc546ca.png',\n",
              " 'fd87b36ae.png',\n",
              " '32c86784b.png',\n",
              " 'b6a3f7876.png',\n",
              " '7f46a71db.png',\n",
              " 'e9cd91682.png',\n",
              " 'a276c65f7.png',\n",
              " '8874bba69.png',\n",
              " '187668bde.png',\n",
              " 'd84d37a61.png',\n",
              " '1459e96a0.png',\n",
              " 'edfdb4aeb.png',\n",
              " '3d67c434b.png',\n",
              " '9a3f20121.png',\n",
              " '0bf7bfb05.png',\n",
              " '5315c2dca.png',\n",
              " '0751c0bbc.png',\n",
              " '1f5e5554e.png',\n",
              " 'b7a7f6390.png',\n",
              " '4c8005bbc.png',\n",
              " 'c6c8d4ba0.png',\n",
              " 'cfb18d262.png',\n",
              " '0ad9e7dfb.png',\n",
              " '20e562fd5.png',\n",
              " '780defa2e.png',\n",
              " '1f3f44563.png',\n",
              " 'e3f50adfc.png',\n",
              " '86c309150.png',\n",
              " 'bffc08672.png',\n",
              " '6ba4ef411.png',\n",
              " 'c64370a72.png',\n",
              " 'e82017baa.png',\n",
              " 'acdb75e00.png',\n",
              " 'dfb1d9012.png',\n",
              " 'e0ec5b6a1.png',\n",
              " 'cd5f0db1c.png',\n",
              " '9d3cb4745.png',\n",
              " '219fd68d5.png',\n",
              " '6680836dd.png',\n",
              " '406ecb5c5.png',\n",
              " '41f1c3cdb.png',\n",
              " '35a90f8d0.png',\n",
              " '6b9d6f8c9.png',\n",
              " '65d08b894.png',\n",
              " '64fe8beb9.png',\n",
              " 'bf3924a57.png',\n",
              " '3f826b318.png',\n",
              " '0e8492cb1.png',\n",
              " 'c0f5d9ac8.png',\n",
              " 'fd253a74e.png',\n",
              " 'c7eb96871.png',\n",
              " '12625488b.png',\n",
              " 'a5db42f7d.png',\n",
              " '8e3ed0a25.png',\n",
              " '22fbf13d6.png',\n",
              " '0ee4ad224.png',\n",
              " '0ebf8f2f4.png',\n",
              " '502dff972.png',\n",
              " 'e15fce4f2.png',\n",
              " '26e7ae885.png',\n",
              " '66ab0e8d0.png',\n",
              " 'bb64660b7.png',\n",
              " '9fab816f2.png',\n",
              " 'b341d0aab.png',\n",
              " '043449b0b.png',\n",
              " '4b032563b.png',\n",
              " 'ef9676433.png',\n",
              " 'c069fc3fa.png',\n",
              " 'c5078bac5.png',\n",
              " '466bb6d3b.png',\n",
              " '37714071b.png',\n",
              " 'c75a82234.png',\n",
              " '8585f9718.png',\n",
              " '550a8b7e6.png',\n",
              " 'b215531dd.png',\n",
              " '3efa1f66c.png',\n",
              " '963544aa0.png',\n",
              " 'beebe5f4e.png',\n",
              " '2b55a2da2.png',\n",
              " '03566743d.png',\n",
              " '0caeda5df.png',\n",
              " '539961189.png',\n",
              " '3fbd0fc6a.png',\n",
              " 'bd72d4d8a.png',\n",
              " '53e6e9000.png',\n",
              " 'ae90f2827.png',\n",
              " '23bc8ec4f.png',\n",
              " 'e73e308be.png',\n",
              " 'cf46d09c5.png',\n",
              " 'a19c3faca.png',\n",
              " '41e07778c.png',\n",
              " '2693e5c65.png',\n",
              " 'fb022edf9.png',\n",
              " '24c94a6ca.png',\n",
              " '3dd52bd2a.png',\n",
              " 'e19673dc9.png',\n",
              " 'a8b431a3e.png',\n",
              " '122913909.png',\n",
              " '16357b436.png',\n",
              " 'f1e87cba7.png',\n",
              " 'f9f35cbd4.png',\n",
              " 'a800caead.png',\n",
              " '0c27cf05f.png',\n",
              " 'a0f37c726.png',\n",
              " '74d810f87.png',\n",
              " '8cfd98117.png',\n",
              " 'f33e9d918.png',\n",
              " 'e1809cef2.png',\n",
              " 'd89db156f.png',\n",
              " '4ac29bbf0.png',\n",
              " '71f5323c5.png',\n",
              " '2bd74d2da.png',\n",
              " '0fb233ad6.png',\n",
              " 'e7077322d.png',\n",
              " '20ea96bcc.png',\n",
              " '8a30b2de3.png',\n",
              " '3eda9cbb6.png',\n",
              " '664194d19.png',\n",
              " '19fdf19fb.png',\n",
              " '17a78fb44.png',\n",
              " 'cec5bf198.png',\n",
              " 'e98e5d1d5.png',\n",
              " '866be78b0.png',\n",
              " 'cb76a7766.png',\n",
              " '5dcad9a53.png',\n",
              " '7b52585da.png',\n",
              " '0086a6340.png',\n",
              " '2f0004a7f.png',\n",
              " '8c9953903.png',\n",
              " '8e2e5604e.png',\n",
              " '6be169e41.png',\n",
              " 'b3d6fdb80.png',\n",
              " 'c26ccf73c.png',\n",
              " 'adb7a032c.png',\n",
              " '33748968f.png',\n",
              " '36839d5e9.png',\n",
              " '809eb0b82.png',\n",
              " '00ef713a8.png',\n",
              " '523e5505c.png',\n",
              " '60fea2ef6.png',\n",
              " 'c10ccbd82.png',\n",
              " '8104422bb.png',\n",
              " 'e5881dd33.png',\n",
              " '6df8e31ea.png',\n",
              " '410e6f702.png',\n",
              " 'f9b6bfb00.png',\n",
              " '52dc7a4d6.png',\n",
              " '6da892be6.png',\n",
              " 'b03397525.png',\n",
              " 'c6b76307d.png',\n",
              " '78750e0ff.png',\n",
              " '5a38ac566.png',\n",
              " 'cf90fc52d.png',\n",
              " '98d819587.png',\n",
              " 'a8388a37f.png',\n",
              " '59358cd44.png',\n",
              " '60f0bc617.png',\n",
              " '53ceb4657.png',\n",
              " '800a8c17e.png',\n",
              " '2d992d1fb.png',\n",
              " '07e62f903.png',\n",
              " '96ecad7a1.png',\n",
              " 'e5368474f.png',\n",
              " '824f5d4e5.png',\n",
              " '7691014a1.png',\n",
              " '0f6cbe5e8.png',\n",
              " '2126dc71b.png',\n",
              " '1cfd91582.png',\n",
              " 'ce42adffb.png',\n",
              " '9b9911f20.png',\n",
              " '0c4199daa.png',\n",
              " '97844bfd5.png',\n",
              " '6edb96d45.png',\n",
              " 'a890ac088.png',\n",
              " '1be0713da.png',\n",
              " '8e29abce1.png',\n",
              " '4c7838de4.png',\n",
              " '0d31e6602.png',\n",
              " '76555b064.png',\n",
              " 'b2706e2b3.png',\n",
              " '429211ee6.png',\n",
              " '8d6acbe9b.png',\n",
              " 'a9d2eab61.png',\n",
              " '1d321253f.png',\n",
              " '8d6991365.png',\n",
              " 'd6d31dcbe.png',\n",
              " '4bbf1f6ea.png',\n",
              " 'fadc6adbc.png',\n",
              " 'd5f7dd60a.png',\n",
              " 'df521c0c0.png',\n",
              " '7696badea.png',\n",
              " '35cf9fa01.png',\n",
              " 'f4e7733d4.png',\n",
              " '808cf55c6.png',\n",
              " '76dbd1054.png',\n",
              " 'cc3d2a59a.png',\n",
              " 'dd9f36df7.png',\n",
              " '9cce7328c.png',\n",
              " 'a1e0a6c02.png',\n",
              " '03a2ee656.png',\n",
              " '4e9d3765f.png',\n",
              " 'd515398fd.png',\n",
              " 'e9d48d664.png',\n",
              " '5ca2687a4.png',\n",
              " '177d7e2a4.png',\n",
              " 'a0b393945.png',\n",
              " 'ef3e232ad.png',\n",
              " '02cfeb38d.png',\n",
              " '2ff5cb348.png',\n",
              " 'ef65533d5.png',\n",
              " 'c0bc3997b.png',\n",
              " '966ae5ad9.png',\n",
              " 'e721c6ac8.png',\n",
              " '5bdcfa329.png',\n",
              " '590f5aea6.png',\n",
              " 'aaf4da98f.png',\n",
              " '63c07d340.png',\n",
              " 'eec1079a1.png',\n",
              " 'e15472085.png',\n",
              " 'fa9f3a8f9.png',\n",
              " '1376f3b63.png',\n",
              " 'fc3e58836.png',\n",
              " 'f1f7c833f.png',\n",
              " '506347cfe.png',\n",
              " '6908fb540.png',\n",
              " 'f4caf74f9.png',\n",
              " 'ec08a5d56.png',\n",
              " '42e7ed442.png',\n",
              " '3abb502fb.png',\n",
              " '1f290e016.png',\n",
              " '711b46fba.png',\n",
              " '8a4d3a1b1.png',\n",
              " 'ffc6f8527.png',\n",
              " '116b136de.png',\n",
              " '03e322a29.png',\n",
              " 'd6c8c3c48.png',\n",
              " '808578ed5.png',\n",
              " '003d61042.png',\n",
              " '3ebbe9ca4.png',\n",
              " 'dc55449b2.png',\n",
              " '958bb9e96.png',\n",
              " 'f66ae4070.png',\n",
              " 'fea1d13d6.png',\n",
              " 'a85b48a95.png',\n",
              " '78c5fba1d.png',\n",
              " '8cf909eb3.png',\n",
              " 'f4021df6c.png',\n",
              " 'e4a76885b.png',\n",
              " '97b2f0a10.png',\n",
              " '22e79540f.png',\n",
              " '456d507c0.png',\n",
              " 'bfab3e3d0.png',\n",
              " '50de8a115.png',\n",
              " '8170d33c1.png',\n",
              " '5296835a0.png',\n",
              " '71e73a8a0.png',\n",
              " '754b1adf8.png',\n",
              " 'ab35453cb.png',\n",
              " '06d12f6fa.png',\n",
              " '80e299ae9.png',\n",
              " 'a8c8a1db0.png',\n",
              " '6a41bf95b.png',\n",
              " '35ebe165c.png',\n",
              " 'e921021a8.png',\n",
              " 'e1abb4ff9.png',\n",
              " '5ee9d0a5b.png',\n",
              " 'be341dbdc.png',\n",
              " '5bd71f445.png',\n",
              " '74fd477eb.png',\n",
              " '73260a4ee.png',\n",
              " '4f44ca525.png',\n",
              " '05341a8a6.png',\n",
              " '91e469b4a.png',\n",
              " 'bebcaab66.png',\n",
              " '69d1669f8.png',\n",
              " 'e6f1211a2.png',\n",
              " 'fa5fd1384.png',\n",
              " 'f593c9cf0.png',\n",
              " '7a38416be.png',\n",
              " '20f983a71.png',\n",
              " '7b21ba6ba.png',\n",
              " '444473900.png',\n",
              " 'fba8fc78a.png',\n",
              " '8916793ce.png',\n",
              " '61dd2cdc5.png',\n",
              " '592cc5b89.png',\n",
              " '8a32d0bfa.png',\n",
              " 'bb20fce02.png',\n",
              " '618de3d7a.png',\n",
              " '148bbda66.png',\n",
              " '1b6a6494d.png',\n",
              " '60ee66ddd.png',\n",
              " '0c5f6c493.png',\n",
              " 'f351ce097.png',\n",
              " '5f04aed97.png',\n",
              " '3bbef3ecb.png',\n",
              " '659412b1a.png',\n",
              " 'c0d9e170b.png',\n",
              " '8a8d6c712.png',\n",
              " 'aad8375e0.png',\n",
              " '270b939cf.png',\n",
              " '25a4c427e.png',\n",
              " '00d090cde.png',\n",
              " '59e1cea8d.png',\n",
              " 'dd5ec63d9.png',\n",
              " 'bdde957ec.png',\n",
              " '92292055d.png',\n",
              " '24dbc3b21.png',\n",
              " '39c8fde99.png',\n",
              " '5817b766d.png',\n",
              " 'b944a49ca.png',\n",
              " '47b7d8e17.png',\n",
              " '258b1183c.png',\n",
              " '33317fc2a.png',\n",
              " 'd8f4923f8.png',\n",
              " '1821eb11a.png',\n",
              " '835dc5447.png',\n",
              " '5eb9c26a6.png',\n",
              " '55fed435f.png',\n",
              " 'cbba27d89.png',\n",
              " 'a169b71e7.png',\n",
              " '2dd5cfba9.png',\n",
              " '86f08e6d1.png',\n",
              " '974959ec1.png',\n",
              " '721be0a4a.png',\n",
              " '29ce426a1.png',\n",
              " '6049234e6.png',\n",
              " '2f246d688.png']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 193
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-gBOjq7abqmF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_test = os.listdir()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bNHegsjZnxWQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_train = []\n",
        "y_train = []\n",
        "#y_test = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZSdUcumZp-LV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import os\n",
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4t42yJdqqAJV",
        "colab_type": "code",
        "outputId": "85e392ab-22ca-4160-bc44-f1266de66674",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 218
        }
      },
      "source": [
        "ls"
      ],
      "execution_count": 197,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Plant_Breed10.csv  Plant_Breed21.csv  Plant_Breed32.csv  Plant_Breed43.csv\n",
            "Plant_Breed11.csv  Plant_Breed22.csv  Plant_Breed33.csv  Plant_Breed44.csv\n",
            "Plant_Breed12.csv  Plant_Breed23.csv  Plant_Breed34.csv  Plant_Breed4.csv\n",
            "Plant_Breed13.csv  Plant_Breed24.csv  Plant_Breed35.csv  Plant_Breed5.csv\n",
            "Plant_Breed14.csv  Plant_Breed25.csv  Plant_Breed36.csv  Plant_Breed6.csv\n",
            "Plant_Breed15.csv  Plant_Breed26.csv  Plant_Breed37.csv  Plant_Breed7.csv\n",
            "Plant_Breed16.csv  Plant_Breed27.csv  Plant_Breed38.csv  Plant_Breed8.csv\n",
            "Plant_Breed17.csv  Plant_Breed28.csv  Plant_Breed39.csv  Plant_Breed9.csv\n",
            "Plant_Breed18.csv  Plant_Breed29.csv  Plant_Breed3.csv   \u001b[0m\u001b[01;34mtest\u001b[0m/\n",
            "Plant_Breed19.csv  Plant_Breed2.csv   Plant_Breed40.csv  test.zip\n",
            "Plant_Breed1.csv   Plant_Breed30.csv  Plant_Breed41.csv  \u001b[01;34mtrain\u001b[0m/\n",
            "Plant_Breed20.csv  Plant_Breed31.csv  Plant_Breed42.csv  train.zip\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zrEWFArOqFXs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed/train')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PfhxmZSxqNeg",
        "colab_type": "code",
        "outputId": "edfafa6e-4e7a-4e93-8aca-a292605bfe0f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 218
        }
      },
      "source": [
        "os.listdir()"
      ],
      "execution_count": 199,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Black-grass',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Maize',\n",
              " 'Common Chickweed',\n",
              " 'Common wheat']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 199
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J3j0GUFiMky_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "l = os.listdir()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PXDS4G3bMrYO",
        "colab_type": "code",
        "outputId": "91990efb-fbae-408e-eae9-5e3e91a2dc9e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "le = preprocessing.LabelEncoder()\n",
        "le.fit(l)\n",
        "\n",
        "list(le.classes_)\n",
        "\n",
        "le.transform(l) "
      ],
      "execution_count": 201,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 5, 10,  2,  0, 11,  9,  1,  6,  8,  7,  3,  4])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 201
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8A7CxdukWjDD",
        "colab_type": "code",
        "outputId": "7f0e6dbd-bd26-44c5-d495-862b249444d0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 218
        }
      },
      "source": [
        "list(le.inverse_transform([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]))"
      ],
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Black-grass',\n",
              " 'Charlock',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Common wheat',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Sugar beet']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 202
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8BYcVPZUqOMi",
        "colab_type": "code",
        "outputId": "fc71cf20-05c9-444f-af07-8a7ed5bf6a48",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 218
        }
      },
      "source": [
        "x_train = []\n",
        "y_train = []\n",
        "import cv2\n",
        "for i in os.listdir():\n",
        "    print(i)\n",
        "    if (os.path.isdir(i)):\n",
        "            for j in os.listdir(i):\n",
        "                try:\n",
        "                    dummy = cv2.imread('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed/train/' + i + \"/\" + j)\n",
        "                    dummy = cv2.resize(dummy,(128,128))\n",
        "                    x_train.append(dummy)\n",
        "                    y_train.append(i)\n",
        "                except Exception as e:\n",
        "                    print(e)\n"
      ],
      "execution_count": 203,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fat Hen\n",
            "Small-flowered Cranesbill\n",
            "Cleavers\n",
            "Black-grass\n",
            "Sugar beet\n",
            "Shepherds Purse\n",
            "Charlock\n",
            "Loose Silky-bent\n",
            "Scentless Mayweed\n",
            "Maize\n",
            "Common Chickweed\n",
            "Common wheat\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KVTWzHBLvLkR",
        "colab_type": "code",
        "outputId": "acd8fd8e-1403-4234-c477-1275bf3ff559",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "x_train[0].shape"
      ],
      "execution_count": 204,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(128, 128, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 204
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jX9jD090uIxX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import pandas as pd\n",
        "dum = pd.get_dummies(y_train)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2LniWcretNjt",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "encoded_labels = dum\n",
        "y_train = dum"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yW1MRzNzscJD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "y_train = np.array(y_train)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i_4kF4C9qV73",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_train = np.array(x_train)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "69ZkfU51usBT",
        "colab_type": "code",
        "outputId": "3e05f700-4b24-414a-ab18-a6fc270163c9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "x_train[0].shape"
      ],
      "execution_count": 209,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(128, 128, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 209
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u3Yqo5_8LG7y",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test = np.array(x_test)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IkIsdd5oLMbZ",
        "colab_type": "code",
        "outputId": "3671f266-53de-422d-c306-5dcc9f9d7352",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "x_test[0].shape"
      ],
      "execution_count": 211,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(128, 128, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 211
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bUhC9eIKf8m2",
        "colab_type": "text"
      },
      "source": [
        "#Divide the data set into Train and validation data sets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VzBQIPBAuQwN",
        "colab_type": "code",
        "outputId": "6f810b21-76f2-4786-8d0e-eb8335b2c735",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_train2, x_val, y_train2, y_val = train_test_split(x_train, y_train, test_size=0.2, random_state=7)\n",
        "print (len(x_train2))\n",
        "print (len(x_val))"
      ],
      "execution_count": 212,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "3800\n",
            "950\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zOXrhr_huSPR",
        "colab_type": "code",
        "outputId": "757a709e-5e96-4669-8d62-936b85c69704",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "x_train2[0].shape"
      ],
      "execution_count": 213,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(128, 128, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 213
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dMocdt9JuVOJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_train2 = x_train2.reshape(x_train2.shape[0],128,128,3)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k1KiJod-uWkR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_val = x_val.reshape(x_val.shape[0],128,128,3)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TyWC85VZvS5_",
        "colab_type": "code",
        "outputId": "d28b103e-114b-434b-b844-bcfb82902635",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "print (x_train2.shape)\n",
        "print (x_val.shape)"
      ],
      "execution_count": 216,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(3800, 128, 128, 3)\n",
            "(950, 128, 128, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DKgWlwHSvUc2",
        "colab_type": "code",
        "outputId": "fd16995a-9832-43b2-9253-878ee6a90ad7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "print(y_train2.shape)\n",
        "print(y_val.shape)"
      ],
      "execution_count": 217,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(3800, 12)\n",
            "(950, 12)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9XSAoGRfgTdt",
        "colab_type": "text"
      },
      "source": [
        "#Initialize & build the model "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gJxwoWFevVsM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Convolution2D, Dropout, Dense"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UdofRIJXvXXU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.layers import BatchNormalization\n",
        "from keras.layers import MaxPooling2D\n",
        "from keras.layers import Flatten\n",
        "from keras.optimizers import adam\n",
        "from keras.optimizers import adamax\n",
        "from keras.utils import np_utils"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i0PInUEOvZDn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = Sequential()\n",
        "model.add(BatchNormalization(input_shape = (128,128,3)))\n",
        "model.add(Convolution2D(32, (3,3), activation ='relu', input_shape = (128, 128, 3)))\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=64, kernel_size=4, padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=128, kernel_size=3, padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=128, kernel_size=2, padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Flatten()) \n",
        "\n",
        "# fully connected layer\n",
        "model.add(Dense(units=128,activation = 'relu'))\n",
        "model.add(Dense(units = 64, activation = 'relu'))\n",
        "#model.add(Dropout(0.25))\n",
        "model.add(Dense(units = 32, activation = 'relu'))\n",
        "#model.add(Dropout(0.25))\n",
        "model.add(Dense(units = 12, activation = 'softmax')) \n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DR6o7TgKgfkm",
        "colab_type": "text"
      },
      "source": [
        "#Optimize the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3V2UmrqIva95",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "optimizer = adam(lr=0.001)\n",
        "model.compile(optimizer='adam', loss = 'categorical_crossentropy',metrics = ['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xMnStHPbvdS9",
        "colab_type": "code",
        "outputId": "94146f77-6356-4289-e6cc-f8cafd16a065",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 605
        }
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 222,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "batch_normalization_19 (Batc (None, 128, 128, 3)       12        \n",
            "_________________________________________________________________\n",
            "conv2d_73 (Conv2D)           (None, 126, 126, 32)      896       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_73 (MaxPooling (None, 63, 63, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_74 (Conv2D)           (None, 63, 63, 64)        32832     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_74 (MaxPooling (None, 31, 31, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_75 (Conv2D)           (None, 31, 31, 128)       73856     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_75 (MaxPooling (None, 15, 15, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_76 (Conv2D)           (None, 15, 15, 128)       65664     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_76 (MaxPooling (None, 7, 7, 128)         0         \n",
            "_________________________________________________________________\n",
            "flatten_19 (Flatten)         (None, 6272)              0         \n",
            "_________________________________________________________________\n",
            "dense_73 (Dense)             (None, 128)               802944    \n",
            "_________________________________________________________________\n",
            "dense_74 (Dense)             (None, 64)                8256      \n",
            "_________________________________________________________________\n",
            "dense_75 (Dense)             (None, 32)                2080      \n",
            "_________________________________________________________________\n",
            "dense_76 (Dense)             (None, 12)                396       \n",
            "=================================================================\n",
            "Total params: 986,936\n",
            "Trainable params: 986,930\n",
            "Non-trainable params: 6\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aJw9iI46_8OK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Store Training Results\n",
        "early_stopping = keras.callbacks.EarlyStopping(monitor='val_acc', patience=5, verbose=1, mode='auto')\n",
        "callback_list = [early_stopping]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KcGlfQxFgp9c",
        "colab_type": "text"
      },
      "source": [
        "# Predict the accuracy for both train and validation data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vdsV52_lvfBr",
        "colab_type": "code",
        "outputId": "6c427dbb-3296-481b-ace0-fd0a5de00fb0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 722
        }
      },
      "source": [
        "#batch_size = 64\n",
        "batch_size = 32\n",
        "nb_epochs = 10\n",
        "history = model.fit(x_train,y_train,\n",
        "                    epochs=50, \n",
        "                    validation_data=(x_val,y_val),\n",
        "                    verbose = 1,\n",
        "                    initial_epoch=0,\n",
        "                    callbacks=callback_list)"
      ],
      "execution_count": 224,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 4750 samples, validate on 950 samples\n",
            "Epoch 1/50\n",
            "4750/4750 [==============================] - 8s 2ms/step - loss: 1.8064 - acc: 0.3882 - val_loss: 1.2456 - val_acc: 0.5663\n",
            "Epoch 2/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 1.0221 - acc: 0.6516 - val_loss: 0.6534 - val_acc: 0.7947\n",
            "Epoch 3/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.7249 - acc: 0.7579 - val_loss: 0.4531 - val_acc: 0.8600\n",
            "Epoch 4/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.5337 - acc: 0.8164 - val_loss: 0.3882 - val_acc: 0.8768\n",
            "Epoch 5/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.4231 - acc: 0.8516 - val_loss: 0.3122 - val_acc: 0.8947\n",
            "Epoch 6/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.3541 - acc: 0.8735 - val_loss: 0.2770 - val_acc: 0.9053\n",
            "Epoch 7/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.2829 - acc: 0.8939 - val_loss: 0.2627 - val_acc: 0.9032\n",
            "Epoch 8/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.2461 - acc: 0.9120 - val_loss: 0.1088 - val_acc: 0.9726\n",
            "Epoch 9/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1764 - acc: 0.9358 - val_loss: 0.1776 - val_acc: 0.9358\n",
            "Epoch 10/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1373 - acc: 0.9493 - val_loss: 0.1084 - val_acc: 0.9558\n",
            "Epoch 11/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1558 - acc: 0.9438 - val_loss: 0.1503 - val_acc: 0.9432\n",
            "Epoch 12/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1455 - acc: 0.9474 - val_loss: 0.0905 - val_acc: 0.9674\n",
            "Epoch 13/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0931 - acc: 0.9665 - val_loss: 0.0503 - val_acc: 0.9821\n",
            "Epoch 14/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1045 - acc: 0.9608 - val_loss: 0.0512 - val_acc: 0.9811\n",
            "Epoch 15/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0684 - acc: 0.9749 - val_loss: 0.0222 - val_acc: 0.9926\n",
            "Epoch 16/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0950 - acc: 0.9667 - val_loss: 0.0642 - val_acc: 0.9768\n",
            "Epoch 17/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.1065 - acc: 0.9602 - val_loss: 0.0298 - val_acc: 0.9916\n",
            "Epoch 18/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0448 - acc: 0.9842 - val_loss: 0.0502 - val_acc: 0.9863\n",
            "Epoch 19/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0803 - acc: 0.9747 - val_loss: 0.0462 - val_acc: 0.9842\n",
            "Epoch 20/50\n",
            "4750/4750 [==============================] - 5s 1ms/step - loss: 0.0795 - acc: 0.9735 - val_loss: 0.1291 - val_acc: 0.9505\n",
            "Epoch 00020: early stopping\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yXO4aBr-8fH0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "### Now, to the above model, lets add Data Augmentation \n",
        "\n",
        "### Import the ImageDataGenrator from keras and fit the training images\n",
        "\n",
        "from keras.preprocessing.image import ImageDataGenerator"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zd6ISQVEqWGh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_train_255 = x_train / 255"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TJf_vhh080UC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "datagen = ImageDataGenerator(\n",
        "    featurewise_center=False,\n",
        "    featurewise_std_normalization=False,\n",
        "    rotation_range=10, # random;y rotate images in range (degrees, 0 to 180)\n",
        "    width_shift_range=0.1, # randomly shift images horizontally (fraction of total width)\n",
        "    height_shift_range=0.1, # randomly shift images horizontally (fraction of total height)\n",
        "    horizontal_flip=True)\n",
        "\n",
        "# compute quantities required for featurewise normalization\n",
        "# (std, mean, and principal components if ZCA whitening is applied)\n",
        "datagen.fit(x_train_255)\n",
        "\n",
        "# fits the model on batches with real-time data augmentation:\n",
        "#model.fit_generator(datagen.flow(x_train, y_train, batch_size=32),\n",
        "#                    steps_per_epoch=len(x_train) / 32, epochs=epochs)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uDb5hDiPPCts",
        "colab_type": "code",
        "outputId": "2c62d768-1c16-41a2-d509-28f94b3cdb8c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 109
        }
      },
      "source": [
        "#### Showing 5 versions of the first image in training dataset using image datagenerator.flow()\n",
        "\n",
        "from matplotlib import pyplot as plt\n",
        "gen = datagen.flow(x_train_255[0:1], batch_size=1)\n",
        "for i in range(1, 6):\n",
        "    plt.subplot(1,5,i)\n",
        "    plt.axis(\"off\")\n",
        "    plt.imshow(gen.next().squeeze(), cmap='gray')\n",
        "    plt.plot()\n",
        "plt.show()"
      ],
      "execution_count": 274,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAABcCAYAAACV3PW/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndtkntP8HsraOzuvWpoIAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 5 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RohTk6hu1kWu",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = Sequential()\n",
        "model.add(BatchNormalization(input_shape = (128,128,3)))\n",
        "\n",
        "model.add(Convolution2D(32, (3,3), activation ='relu', input_shape = (128, 128, 3)))\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=64, kernel_size=4, padding='same', activation='relu'))\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=128, kernel_size=3, padding='same', activation='relu'))\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Convolution2D(filters=128, kernel_size=2, padding='same', activation='relu'))\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=2))\n",
        "#model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Flatten()) \n",
        "\n",
        "# fully connected layer\n",
        "model.add(Dense(units=128,activation = 'relu'))\n",
        "\n",
        "model.add(Dense(units = 64, activation = 'relu'))\n",
        "\n",
        "#model.add(Dropout(0.25))\n",
        "model.add(Dense(units = 32, activation = 'relu'))\n",
        "\n",
        "#model.add(Dropout(0.25))\n",
        "model.add(Dense(units = 12, activation = 'softmax')) \n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jhk4ZfKn1j8z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "optimizer = adam(lr=0.001)\n",
        "model.compile(optimizer='adam', loss = 'categorical_crossentropy',metrics = ['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fVmj05ksANA_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Store Training Results\n",
        "early_stopping = keras.callbacks.EarlyStopping(monitor='val_acc', patience=10, verbose=1, mode='auto')\n",
        "callback_list = [early_stopping]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v_owvftX1jCf",
        "colab_type": "code",
        "outputId": "55cf0fe5-4975-444d-d7e6-f9e8e6e7b482",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#batch_size = 64\n",
        "batch_size = 32\n",
        "nb_epochs = 20\n",
        "history = model.fit_generator(datagen.flow(x_train,y_train,batch_size=32),\n",
        "                    samples_per_epoch=x_train.shape[0], nb_epoch=100, validation_data=(x_val, y_val),\n",
        "                             callbacks=callback_list)"
      ],
      "execution_count": 280,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "\r  1/148 [..............................] - ETA: 19s - loss: 0.0856 - acc: 0.9688"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:5: UserWarning: The semantics of the Keras 2 argument `steps_per_epoch` is not the same as the Keras 1 argument `samples_per_epoch`. `steps_per_epoch` is the number of batches to draw from the generator at each epoch. Basically steps_per_epoch = samples_per_epoch/batch_size. Similarly `nb_val_samples`->`validation_steps` and `val_samples`->`steps` arguments have changed. Update your method calls accordingly.\n",
            "  \"\"\"\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:5: UserWarning: Update your `fit_generator` call to the Keras 2 API: `fit_generator(<keras_pre..., validation_data=(array([[[..., callbacks=[<keras.ca..., steps_per_epoch=148, epochs=100)`\n",
            "  \"\"\"\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "148/148 [==============================] - 16s 109ms/step - loss: 0.1027 - acc: 0.9647 - val_loss: 0.0544 - val_acc: 0.9821\n",
            "Epoch 2/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.1047 - acc: 0.9632 - val_loss: 0.0707 - val_acc: 0.9747\n",
            "Epoch 3/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.1083 - acc: 0.9613 - val_loss: 0.0370 - val_acc: 0.9842\n",
            "Epoch 4/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0963 - acc: 0.9691 - val_loss: 0.0354 - val_acc: 0.9916\n",
            "Epoch 5/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0736 - acc: 0.9735 - val_loss: 0.0435 - val_acc: 0.9832\n",
            "Epoch 6/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0909 - acc: 0.9715 - val_loss: 0.0381 - val_acc: 0.9884\n",
            "Epoch 7/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.1040 - acc: 0.9637 - val_loss: 0.0509 - val_acc: 0.9842\n",
            "Epoch 8/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0722 - acc: 0.9751 - val_loss: 0.0636 - val_acc: 0.9747\n",
            "Epoch 9/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0813 - acc: 0.9734 - val_loss: 0.0466 - val_acc: 0.9811\n",
            "Epoch 10/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.0798 - acc: 0.9729 - val_loss: 0.0363 - val_acc: 0.9895\n",
            "Epoch 11/100\n",
            "148/148 [==============================] - 16s 106ms/step - loss: 0.1034 - acc: 0.9630 - val_loss: 0.0292 - val_acc: 0.9905\n",
            "Epoch 12/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.0730 - acc: 0.9721 - val_loss: 0.0221 - val_acc: 0.9937\n",
            "Epoch 13/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0824 - acc: 0.9716 - val_loss: 0.0329 - val_acc: 0.9863\n",
            "Epoch 14/100\n",
            "148/148 [==============================] - 17s 113ms/step - loss: 0.0693 - acc: 0.9764 - val_loss: 0.0221 - val_acc: 0.9916\n",
            "Epoch 15/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0701 - acc: 0.9755 - val_loss: 0.0619 - val_acc: 0.9832\n",
            "Epoch 16/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0680 - acc: 0.9776 - val_loss: 0.0823 - val_acc: 0.9705\n",
            "Epoch 17/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0566 - acc: 0.9807 - val_loss: 0.0756 - val_acc: 0.9684\n",
            "Epoch 18/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0897 - acc: 0.9706 - val_loss: 0.0637 - val_acc: 0.9768\n",
            "Epoch 19/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0637 - acc: 0.9795 - val_loss: 0.0403 - val_acc: 0.9842\n",
            "Epoch 20/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0916 - acc: 0.9696 - val_loss: 0.0388 - val_acc: 0.9874\n",
            "Epoch 21/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0825 - acc: 0.9709 - val_loss: 0.0484 - val_acc: 0.9842\n",
            "Epoch 22/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0768 - acc: 0.9738 - val_loss: 0.0585 - val_acc: 0.9789\n",
            "Epoch 23/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0731 - acc: 0.9766 - val_loss: 0.0188 - val_acc: 0.9916\n",
            "Epoch 24/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0740 - acc: 0.9751 - val_loss: 0.0564 - val_acc: 0.9821\n",
            "Epoch 25/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0779 - acc: 0.9711 - val_loss: 0.0395 - val_acc: 0.9874\n",
            "Epoch 26/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0794 - acc: 0.9733 - val_loss: 0.0510 - val_acc: 0.9842\n",
            "Epoch 27/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0549 - acc: 0.9816 - val_loss: 0.0208 - val_acc: 0.9916\n",
            "Epoch 28/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0555 - acc: 0.9823 - val_loss: 0.0311 - val_acc: 0.9916\n",
            "Epoch 29/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0713 - acc: 0.9766 - val_loss: 0.0411 - val_acc: 0.9842\n",
            "Epoch 30/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0780 - acc: 0.9721 - val_loss: 0.0375 - val_acc: 0.9863\n",
            "Epoch 31/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0647 - acc: 0.9764 - val_loss: 0.0848 - val_acc: 0.9695\n",
            "Epoch 32/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0775 - acc: 0.9745 - val_loss: 0.0366 - val_acc: 0.9884\n",
            "Epoch 33/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0696 - acc: 0.9769 - val_loss: 0.0250 - val_acc: 0.9916\n",
            "Epoch 34/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0620 - acc: 0.9782 - val_loss: 0.0153 - val_acc: 0.9947\n",
            "Epoch 35/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0572 - acc: 0.9816 - val_loss: 0.0242 - val_acc: 0.9947\n",
            "Epoch 36/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0532 - acc: 0.9842 - val_loss: 0.0142 - val_acc: 0.9947\n",
            "Epoch 37/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0426 - acc: 0.9859 - val_loss: 0.0222 - val_acc: 0.9937\n",
            "Epoch 38/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0447 - acc: 0.9847 - val_loss: 0.0218 - val_acc: 0.9905\n",
            "Epoch 39/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0749 - acc: 0.9759 - val_loss: 0.0399 - val_acc: 0.9853\n",
            "Epoch 40/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0427 - acc: 0.9854 - val_loss: 0.0267 - val_acc: 0.9937\n",
            "Epoch 41/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0795 - acc: 0.9755 - val_loss: 0.2246 - val_acc: 0.9463\n",
            "Epoch 42/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0703 - acc: 0.9764 - val_loss: 0.0551 - val_acc: 0.9779\n",
            "Epoch 43/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0567 - acc: 0.9824 - val_loss: 0.0106 - val_acc: 0.9968\n",
            "Epoch 44/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0622 - acc: 0.9806 - val_loss: 0.0387 - val_acc: 0.9832\n",
            "Epoch 45/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0544 - acc: 0.9829 - val_loss: 0.0252 - val_acc: 0.9884\n",
            "Epoch 46/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0467 - acc: 0.9852 - val_loss: 0.0357 - val_acc: 0.9884\n",
            "Epoch 47/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0712 - acc: 0.9795 - val_loss: 0.0316 - val_acc: 0.9916\n",
            "Epoch 48/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.0460 - acc: 0.9848 - val_loss: 0.0127 - val_acc: 0.9958\n",
            "Epoch 49/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.0834 - acc: 0.9740 - val_loss: 0.0186 - val_acc: 0.9926\n",
            "Epoch 50/100\n",
            "148/148 [==============================] - 16s 107ms/step - loss: 0.0746 - acc: 0.9768 - val_loss: 0.0397 - val_acc: 0.9863\n",
            "Epoch 51/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0322 - acc: 0.9897 - val_loss: 0.0239 - val_acc: 0.9916\n",
            "Epoch 52/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0427 - acc: 0.9854 - val_loss: 0.0308 - val_acc: 0.9905\n",
            "Epoch 53/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0853 - acc: 0.9746 - val_loss: 0.0547 - val_acc: 0.9842\n",
            "Epoch 54/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0748 - acc: 0.9755 - val_loss: 0.0212 - val_acc: 0.9937\n",
            "Epoch 55/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0522 - acc: 0.9825 - val_loss: 0.0331 - val_acc: 0.9884\n",
            "Epoch 56/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0489 - acc: 0.9865 - val_loss: 0.0396 - val_acc: 0.9853\n",
            "Epoch 57/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0668 - acc: 0.9789 - val_loss: 0.0224 - val_acc: 0.9916\n",
            "Epoch 58/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0457 - acc: 0.9848 - val_loss: 0.0311 - val_acc: 0.9874\n",
            "Epoch 59/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0393 - acc: 0.9862 - val_loss: 0.0546 - val_acc: 0.9821\n",
            "Epoch 60/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0346 - acc: 0.9882 - val_loss: 0.0339 - val_acc: 0.9884\n",
            "Epoch 61/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0761 - acc: 0.9782 - val_loss: 0.0257 - val_acc: 0.9895\n",
            "Epoch 62/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0485 - acc: 0.9840 - val_loss: 0.0104 - val_acc: 0.9968\n",
            "Epoch 63/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0525 - acc: 0.9842 - val_loss: 0.0264 - val_acc: 0.9895\n",
            "Epoch 64/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0490 - acc: 0.9840 - val_loss: 0.0122 - val_acc: 0.9937\n",
            "Epoch 65/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0461 - acc: 0.9846 - val_loss: 0.0137 - val_acc: 0.9968\n",
            "Epoch 66/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0450 - acc: 0.9859 - val_loss: 0.0276 - val_acc: 0.9884\n",
            "Epoch 67/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0563 - acc: 0.9846 - val_loss: 0.0567 - val_acc: 0.9853\n",
            "Epoch 68/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0679 - acc: 0.9816 - val_loss: 0.0787 - val_acc: 0.9737\n",
            "Epoch 69/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0558 - acc: 0.9783 - val_loss: 0.0098 - val_acc: 0.9989\n",
            "Epoch 70/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0417 - acc: 0.9875 - val_loss: 0.0454 - val_acc: 0.9853\n",
            "Epoch 71/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0549 - acc: 0.9808 - val_loss: 0.0155 - val_acc: 0.9958\n",
            "Epoch 72/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0457 - acc: 0.9854 - val_loss: 0.0221 - val_acc: 0.9916\n",
            "Epoch 73/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0552 - acc: 0.9835 - val_loss: 0.0187 - val_acc: 0.9937\n",
            "Epoch 74/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0448 - acc: 0.9862 - val_loss: 0.0144 - val_acc: 0.9937\n",
            "Epoch 75/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0364 - acc: 0.9875 - val_loss: 0.0148 - val_acc: 0.9937\n",
            "Epoch 76/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0440 - acc: 0.9860 - val_loss: 0.0167 - val_acc: 0.9937\n",
            "Epoch 77/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0406 - acc: 0.9854 - val_loss: 0.0848 - val_acc: 0.9821\n",
            "Epoch 78/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0636 - acc: 0.9801 - val_loss: 0.0166 - val_acc: 0.9937\n",
            "Epoch 79/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0369 - acc: 0.9864 - val_loss: 0.0175 - val_acc: 0.9916\n",
            "Epoch 80/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0474 - acc: 0.9839 - val_loss: 0.0826 - val_acc: 0.9821\n",
            "Epoch 81/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0393 - acc: 0.9848 - val_loss: 0.0307 - val_acc: 0.9874\n",
            "Epoch 82/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0424 - acc: 0.9869 - val_loss: 0.0062 - val_acc: 0.9979\n",
            "Epoch 83/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0608 - acc: 0.9837 - val_loss: 0.0325 - val_acc: 0.9884\n",
            "Epoch 84/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0244 - acc: 0.9924 - val_loss: 0.0175 - val_acc: 0.9947\n",
            "Epoch 85/100\n",
            "148/148 [==============================] - 16s 106ms/step - loss: 0.0485 - acc: 0.9846 - val_loss: 0.0497 - val_acc: 0.9842\n",
            "Epoch 86/100\n",
            "148/148 [==============================] - 16s 106ms/step - loss: 0.0590 - acc: 0.9840 - val_loss: 0.0952 - val_acc: 0.9726\n",
            "Epoch 87/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0658 - acc: 0.9804 - val_loss: 0.0155 - val_acc: 0.9958\n",
            "Epoch 88/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0316 - acc: 0.9900 - val_loss: 0.0211 - val_acc: 0.9926\n",
            "Epoch 89/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0226 - acc: 0.9922 - val_loss: 0.0285 - val_acc: 0.9895\n",
            "Epoch 90/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0350 - acc: 0.9901 - val_loss: 0.0431 - val_acc: 0.9853\n",
            "Epoch 91/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0807 - acc: 0.9776 - val_loss: 0.0607 - val_acc: 0.9811\n",
            "Epoch 92/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0527 - acc: 0.9844 - val_loss: 0.0253 - val_acc: 0.9895\n",
            "Epoch 93/100\n",
            "148/148 [==============================] - 16s 110ms/step - loss: 0.0393 - acc: 0.9883 - val_loss: 0.0953 - val_acc: 0.9758\n",
            "Epoch 94/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0375 - acc: 0.9873 - val_loss: 0.0937 - val_acc: 0.9779\n",
            "Epoch 95/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0614 - acc: 0.9822 - val_loss: 0.0157 - val_acc: 0.9947\n",
            "Epoch 96/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0266 - acc: 0.9913 - val_loss: 0.0120 - val_acc: 0.9968\n",
            "Epoch 97/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0395 - acc: 0.9865 - val_loss: 0.0117 - val_acc: 0.9968\n",
            "Epoch 98/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0529 - acc: 0.9825 - val_loss: 0.0156 - val_acc: 0.9958\n",
            "Epoch 99/100\n",
            "148/148 [==============================] - 16s 109ms/step - loss: 0.0355 - acc: 0.9886 - val_loss: 0.0410 - val_acc: 0.9905\n",
            "Epoch 100/100\n",
            "148/148 [==============================] - 16s 108ms/step - loss: 0.0653 - acc: 0.9827 - val_loss: 0.0367 - val_acc: 0.9842\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QRD-f5FjPmAM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Train accuracy is 98.27%\n",
        "#Validation accuracy is 98.42%"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_hfYgAypHWPi",
        "colab_type": "code",
        "outputId": "71f3ee7d-f0ca-4dde-c57a-46347330e938",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# PREDICTIONS\n",
        "y_pred = model.predict(x_test)\n",
        "y_pred.shape"
      ],
      "execution_count": 281,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(794, 12)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 281
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eJ9HuGW5aNbk",
        "colab_type": "code",
        "outputId": "b05fb83d-c93c-4405-9bfb-11e3c78306be",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        }
      },
      "source": [
        "y_pred"
      ],
      "execution_count": 282,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[3.6448392e-13, 1.4113970e-09, 3.3140435e-07, ..., 1.8290573e-04,\n",
              "        7.6725826e-09, 2.3261659e-09],\n",
              "       [4.0690979e-19, 3.3680473e-14, 9.9431705e-16, ..., 3.1080826e-16,\n",
              "        6.0283536e-16, 1.0000000e+00],\n",
              "       [2.7375915e-18, 1.6869157e-18, 7.6691342e-17, ..., 6.2357037e-23,\n",
              "        1.2881265e-23, 5.6420456e-24],\n",
              "       ...,\n",
              "       [3.1138283e-14, 4.6888171e-16, 3.4316643e-11, ..., 4.2501921e-07,\n",
              "        1.3253286e-10, 1.3015803e-09],\n",
              "       [3.5052240e-04, 5.5467337e-04, 3.6700645e-03, ..., 1.0312398e-06,\n",
              "        8.2449278e-06, 8.2776087e-06],\n",
              "       [4.0190676e-12, 5.4061141e-15, 1.7095019e-14, ..., 3.4158063e-13,\n",
              "        1.0000000e+00, 2.7167993e-15]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 282
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FzX8G6Jqvg-c",
        "colab_type": "code",
        "outputId": "a2f5afdc-243c-44d3-821f-b3255ebdb252",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "y_class = np.argmax(y_pred, axis = 1) \n",
        "y_class.shape"
      ],
      "execution_count": 283,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(794,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 283
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gwcw16rJd6AI",
        "colab_type": "code",
        "outputId": "a7401b5c-305d-45b4-d51c-4ab8987c3e37",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 806
        }
      },
      "source": [
        "y_class"
      ],
      "execution_count": 284,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 8, 11,  5,  3,  8,  0,  7, 10, 10,  8,  8,  6, 10,  7,  5,  9, 11,\n",
              "        9,  8,  2,  3,  3, 11, 11,  3,  0,  3,  3,  6,  1,  4,  7, 10,  6,\n",
              "        6,  3,  8, 10,  2,  1, 10,  8,  4,  5,  8,  7,  4,  8, 11, 10, 10,\n",
              "        2, 11,  8,  9,  5,  5, 11,  5, 10,  6,  6, 10, 10,  6,  5,  3,  6,\n",
              "        8,  9,  6, 11,  3,  1,  8,  6,  8,  0,  5,  2,  3,  5,  4,  9,  8,\n",
              "        3, 11,  6,  0,  0,  0,  4,  7,  1,  7,  5,  8,  1, 10,  5,  8,  8,\n",
              "        8,  3,  3,  6,  3, 10,  6, 10, 11,  4,  8,  6,  6,  8,  6,  2,  6,\n",
              "        6,  8,  1,  3, 10,  6, 10,  6,  7,  3,  3,  0, 10,  4,  3,  3,  1,\n",
              "        3,  8,  0,  1,  5,  0,  3, 11,  3,  3,  1,  1, 11,  6,  1,  6,  1,\n",
              "        5,  0, 11,  5,  6,  8,  3,  2, 11,  2,  9,  3,  0,  8,  3,  3,  3,\n",
              "        4, 10,  9, 10,  0,  7,  2,  6,  1,  8,  7,  9,  1,  3, 10, 11,  3,\n",
              "       11,  7,  1,  6,  6,  7,  2,  1, 11,  8,  5, 11, 10,  1,  4,  3,  8,\n",
              "        9,  8,  6,  6,  1,  6,  7,  4,  1,  6,  5,  3, 10,  2,  8,  5,  8,\n",
              "        6,  4,  8,  9,  7,  2,  1,  3,  6, 11, 11,  4,  6,  1,  0,  6,  9,\n",
              "        4,  3, 10,  6,  9, 11, 10, 10,  5,  6, 11, 11,  2,  9,  5,  3,  6,\n",
              "        5, 11,  9,  6,  6,  8,  6, 10,  3,  4,  7,  7,  6,  3,  2,  5,  7,\n",
              "        1,  6,  6,  6,  9,  4, 10,  1,  2,  3,  8,  2,  1,  2,  8,  0,  0,\n",
              "       11,  9,  5,  2,  1,  5,  0,  8,  6,  2,  8,  3,  2, 10,  7,  8,  6,\n",
              "        6,  5,  1,  1, 10,  3, 10,  6,  3,  3,  5,  2,  3, 11,  3, 10,  0,\n",
              "       10,  0,  6,  9,  9,  4,  7,  6,  4,  5, 11,  1,  7, 11, 10,  1,  1,\n",
              "        8,  1,  5,  3,  0, 11, 11,  3, 10,  6,  9,  9,  8,  1,  5,  9,  5,\n",
              "        8,  5,  2, 11,  2,  9,  8,  9,  3,  6,  8,  2,  1,  3, 10,  8,  5,\n",
              "        2,  6,  8,  6,  8,  5, 10,  3, 10,  8,  8,  3,  5,  9,  2,  1,  3,\n",
              "       11,  8,  6,  6,  3, 10,  5,  6, 10,  3, 11, 11,  2, 10,  0,  3, 11,\n",
              "       11,  6, 11,  8, 10,  2, 10,  5,  4,  0, 10,  0,  1,  5, 10, 10,  6,\n",
              "       11,  3,  2,  5,  6,  8,  9,  6,  6,  3, 11,  8,  5,  0,  6,  3,  0,\n",
              "        3,  3,  2,  6,  0,  6, 11, 11,  6, 10,  7,  6,  3,  3,  6,  3,  2,\n",
              "        1,  3,  0, 10,  3,  8,  2,  2,  6,  3,  6, 11,  3,  6,  6,  8,  3,\n",
              "        2,  8,  5,  6,  0,  6, 11, 10,  5,  0,  0,  8,  7, 11,  3,  7,  2,\n",
              "       11,  8,  2, 11,  6,  6,  0, 11, 10,  4,  9, 11,  1,  5,  8, 11, 11,\n",
              "        8,  3,  3, 11, 10,  2,  2,  5,  7, 10,  6,  8,  7, 10,  8,  5,  1,\n",
              "       11,  5,  3, 11,  3,  1,  5,  6,  7,  9,  8,  0, 10,  2,  0,  0,  8,\n",
              "        1,  5,  4,  3,  3,  8,  4,  3,  3,  5,  3,  9,  2,  7, 11, 11, 10,\n",
              "        1, 11,  7,  6,  6,  3, 11,  8, 10,  6,  8,  8,  8, 10,  3,  0,  6,\n",
              "       10,  8,  5, 11,  8, 10,  4,  4,  2,  0,  3,  5, 10,  0,  3,  4,  5,\n",
              "       11,  9,  9,  3,  6,  2,  8,  1,  3,  5, 11,  8,  3,  5,  1, 10,  2,\n",
              "        3, 10,  6,  2,  1,  6,  7,  9,  5, 10, 10,  3,  5,  6,  7, 10,  8,\n",
              "       11,  1,  7,  3,  1,  0,  8,  0,  9,  2,  1,  2,  8,  3, 10,  6,  4,\n",
              "        5,  3,  0,  5,  0,  8,  9,  1,  8,  6,  0,  9, 10,  9,  8, 11, 10,\n",
              "        6,  2, 10,  1,  6,  9,  7,  3, 10,  3,  6, 11,  6,  8,  0,  0,  6,\n",
              "        1, 11, 11,  1,  3,  2,  0,  8,  9,  1, 11,  1, 10,  9,  0,  8,  9,\n",
              "        6,  4, 11, 11, 10,  1,  5,  1, 11,  1,  7,  9,  8, 11,  1,  5,  7,\n",
              "       10,  2,  2, 11,  9,  7, 10, 11, 10,  2,  9,  9,  9,  8,  0,  5, 11,\n",
              "        6,  6,  3,  3,  5,  3,  8,  1,  8,  1, 11,  9, 11,  1,  5, 11,  5,\n",
              "        0, 10,  4,  6, 10,  8,  1,  1, 11,  2,  6,  6, 10,  8,  6, 10,  2,\n",
              "        9,  3,  6,  6,  2, 11,  8,  3, 11,  6,  9,  6, 10,  2,  5,  3,  8,\n",
              "       10,  6,  7,  7,  1,  9,  5,  4,  6,  3,  5, 10])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 284
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cBYWPJMkd8Pw",
        "colab_type": "code",
        "outputId": "5a255425-f4e7-484e-838d-e56ecf7b52be",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "list(le.inverse_transform(y_class))"
      ],
      "execution_count": 285,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Maize',\n",
              " 'Fat Hen',\n",
              " 'Shepherds Purse',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Common wheat',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Common wheat',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Maize',\n",
              " 'Common wheat',\n",
              " 'Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Charlock',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Fat Hen',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Common wheat',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Common wheat',\n",
              " 'Maize',\n",
              " 'Charlock',\n",
              " 'Maize',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Sugar beet',\n",
              " 'Common wheat',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common wheat',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Charlock',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Sugar beet',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Cleavers',\n",
              " 'Shepherds Purse',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Common wheat',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Shepherds Purse',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Maize',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Scentless Mayweed',\n",
              " 'Maize',\n",
              " 'Shepherds Purse',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Maize',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Common wheat',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Common wheat',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Common wheat',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Maize',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Common wheat',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Common wheat',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Cleavers',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Common wheat',\n",
              " 'Maize',\n",
              " 'Maize',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Fat Hen',\n",
              " 'Maize',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Common wheat',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Maize',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Fat Hen',\n",
              " 'Charlock',\n",
              " 'Charlock',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Shepherds Purse',\n",
              " 'Common wheat',\n",
              " 'Maize',\n",
              " 'Loose Silky-bent',\n",
              " 'Common wheat',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Maize',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Charlock',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Cleavers',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Shepherds Purse',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Cleavers',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Fat Hen',\n",
              " 'Common wheat',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Maize',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Cleavers',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Maize',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Maize',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Black-grass',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common wheat',\n",
              " 'Shepherds Purse',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Cleavers',\n",
              " 'Fat Hen',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Common Chickweed',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Common wheat',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Common wheat',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Shepherds Purse',\n",
              " 'Cleavers',\n",
              " 'Maize',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Maize',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common wheat',\n",
              " 'Common wheat',\n",
              " 'Cleavers',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Black-grass',\n",
              " 'Common Chickweed',\n",
              " 'Common wheat',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Shepherds Purse',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Charlock',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Maize',\n",
              " 'Common Chickweed',\n",
              " 'Charlock',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Shepherds Purse',\n",
              " 'Cleavers',\n",
              " 'Charlock',\n",
              " 'Cleavers',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Common wheat',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Black-grass',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Charlock',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Black-grass',\n",
              " 'Shepherds Purse',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Maize',\n",
              " 'Common Chickweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Black-grass',\n",
              " 'Loose Silky-bent',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Common Chickweed',\n",
              " 'Cleavers',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Shepherds Purse',\n",
              " 'Black-grass',\n",
              " 'Scentless Mayweed',\n",
              " 'Shepherds Purse',\n",
              " 'Loose Silky-bent',\n",
              " 'Common wheat',\n",
              " 'Sugar beet',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Maize',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Maize',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Sugar beet',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Shepherds Purse',\n",
              " 'Shepherds Purse',\n",
              " 'Shepherds Purse',\n",
              " 'Scentless Mayweed',\n",
              " 'Black-grass',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Shepherds Purse',\n",
              " 'Sugar beet',\n",
              " 'Charlock',\n",
              " 'Fat Hen',\n",
              " 'Sugar beet',\n",
              " 'Fat Hen',\n",
              " 'Black-grass',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Common wheat',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Charlock',\n",
              " 'Charlock',\n",
              " 'Sugar beet',\n",
              " 'Cleavers',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Scentless Mayweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Shepherds Purse',\n",
              " 'Common Chickweed',\n",
              " 'Loose Silky-bent',\n",
              " 'Loose Silky-bent',\n",
              " 'Cleavers',\n",
              " 'Sugar beet',\n",
              " 'Scentless Mayweed',\n",
              " 'Common Chickweed',\n",
              " 'Sugar beet',\n",
              " 'Loose Silky-bent',\n",
              " 'Shepherds Purse',\n",
              " 'Loose Silky-bent',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Cleavers',\n",
              " 'Fat Hen',\n",
              " 'Common Chickweed',\n",
              " 'Scentless Mayweed',\n",
              " 'Small-flowered Cranesbill',\n",
              " 'Loose Silky-bent',\n",
              " 'Maize',\n",
              " 'Maize',\n",
              " 'Charlock',\n",
              " 'Shepherds Purse',\n",
              " 'Fat Hen',\n",
              " 'Common wheat',\n",
              " 'Loose Silky-bent',\n",
              " 'Common Chickweed',\n",
              " 'Fat Hen',\n",
              " 'Small-flowered Cranesbill']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 285
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vIqZUFslb3F7",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_class_label = list(le.inverse_transform(y_class))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7dtehWYmiirS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9w2V3qj9XeIH",
        "colab_type": "code",
        "outputId": "8178b863-a24e-4c93-dddd-1c4e02f50122",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#import pdb; pdb.set_trace()\n",
        "submission = pd.DataFrame({\n",
        "        \"file\": y_test,\n",
        "        \"species\": y_class_label\n",
        "    })\n",
        "submission.to_csv('Plant_Breed45.csv', index=False)\n",
        "print('Exported')"
      ],
      "execution_count": 288,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Exported\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c66KgTveP3BQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# After numerous runs the best Test score submitted to Kaggle was 96.221%"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O-Xb5s4PcJpk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "os.chdir('/content/drive/My Drive/ComputerVision_with_CNNs/Plant_Breed')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B13VDVmGex4-",
        "colab_type": "code",
        "outputId": "c33497a5-d59a-477b-9dfe-b4c2f2f8a5e5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "ls"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Plant_Breed1.csv  Plant_Breed3.csv  Plant_Breed5.csv  \u001b[0m\u001b[01;34mtest\u001b[0m/     \u001b[01;34mtrain\u001b[0m/\n",
            "Plant_Breed2.csv  Plant_Breed4.csv  Plant_Breed6.csv  test.zip  train.zip\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "efBFtC4Cey4t",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}