Flowers_R_Kefras.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Flowers R Kefras.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNW71doObtYUzBnvBda+fBw",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "ir",
      "display_name": "R"
    },
    "language_info": {
      "name": "R"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jnmaloof/DeepLearning_JulinMaloof/blob/main/Flowers_R_Kefras.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E2_VZb_8K3RM"
      },
      "source": [
        "# Colab notebook for trying to classify flowers\n",
        "\n",
        "This is an R script.  See this [info on how to start an R script in colab](https://towardsdatascience.com/how-to-use-r-in-google-colab-b6e02d736497)\n",
        "\n",
        "\n",
        "Dataset is from kaggle [here](https://www.kaggle.com/rishitchs/final-flowers-course-project-dataset)\n",
        "\n",
        "## Installations and setup"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7NKERgVaLB6e",
        "outputId": "7b2c8d77-701b-4ab1-ab66-e3f59779a83f"
      },
      "source": [
        "install.packages(\"keras\")"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Installing package into ‘/usr/local/lib/R/site-library’\n",
            "(as ‘lib’ is unspecified)\n",
            "\n",
            "also installing the dependencies ‘png’, ‘config’, ‘reticulate’, ‘tensorflow’, ‘tfruns’, ‘zeallot’\n",
            "\n",
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L2v0gVg8STxI",
        "outputId": "d01dbef5-a0e7-4489-f72f-870db4b1f7b5"
      },
      "source": [
        "library(keras)\n",
        "library(tidyverse)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Warning message in system(\"timedatectl\", intern = TRUE):\n",
            "“running command 'timedatectl' had status 1”\n",
            "── \u001b[1mAttaching packages\u001b[22m ─────────────────────────────────────── tidyverse 1.3.1 ──\n",
            "\n",
            "\u001b[32m✔\u001b[39m \u001b[34mggplot2\u001b[39m 3.3.3     \u001b[32m✔\u001b[39m \u001b[34mpurrr  \u001b[39m 0.3.4\n",
            "\u001b[32m✔\u001b[39m \u001b[34mtibble \u001b[39m 3.1.1     \u001b[32m✔\u001b[39m \u001b[34mdplyr  \u001b[39m 1.0.5\n",
            "\u001b[32m✔\u001b[39m \u001b[34mtidyr  \u001b[39m 1.1.3     \u001b[32m✔\u001b[39m \u001b[34mstringr\u001b[39m 1.4.0\n",
            "\u001b[32m✔\u001b[39m \u001b[34mreadr  \u001b[39m 1.4.0     \u001b[32m✔\u001b[39m \u001b[34mforcats\u001b[39m 0.5.1\n",
            "\n",
            "── \u001b[1mConflicts\u001b[22m ────────────────────────────────────────── tidyverse_conflicts() ──\n",
            "\u001b[31m✖\u001b[39m \u001b[34mdplyr\u001b[39m::\u001b[32mfilter()\u001b[39m masks \u001b[34mstats\u001b[39m::filter()\n",
            "\u001b[31m✖\u001b[39m \u001b[34mdplyr\u001b[39m::\u001b[32mlag()\u001b[39m    masks \u001b[34mstats\u001b[39m::lag()\n",
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sU308un81Sy-"
      },
      "source": [
        "## get data from google drive\n",
        "\n",
        "If you upload your zipped file to google drive, then you can download it using the comands here. \n",
        "\n",
        "Your other alternative is to upload it from your laptop, but that is slower.\n",
        "\n",
        "Either way, the data will have to be uploaded/downloaded every time, because we can't mount the google drive to an R runtime.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fLYjNxGN1hHW",
        "outputId": "d4cdb443-5d1e-48cf-c5c8-e153be526831"
      },
      "source": [
        "install.packages(\"googledrive\")\n",
        "library(\"googledrive\")\n",
        "\n",
        "# this next step is required to authenticate google drive.\n",
        "# Check if is running in Colab and redefine is_interactive()\n",
        "if (file.exists(\"/usr/local/lib/python3.7/dist-packages/google/colab/_ipython.py\")) {\n",
        "  install.packages(\"R.utils\")\n",
        "  library(\"R.utils\")\n",
        "  library(\"httr\")\n",
        "  my_check <- function() {return(TRUE)}\n",
        "  reassignInPackage(\"is_interactive\", pkgName = \"httr\", my_check) \n",
        "  options(rlang_interactive=TRUE)\n",
        "}\n",
        "\n",
        "drive_auth(use_oob = TRUE, cache = TRUE)\n"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Installing package into ‘/usr/local/lib/R/site-library’\n",
            "(as ‘lib’ is unspecified)\n",
            "\n",
            "Installing package into ‘/usr/local/lib/R/site-library’\n",
            "(as ‘lib’ is unspecified)\n",
            "\n",
            "also installing the dependencies ‘R.oo’, ‘R.methodsS3’\n",
            "\n",
            "\n",
            "Loading required package: R.oo\n",
            "\n",
            "Loading required package: R.methodsS3\n",
            "\n",
            "R.methodsS3 v1.8.1 (2020-08-26 16:20:06 UTC) successfully loaded. See ?R.methodsS3 for help.\n",
            "\n",
            "R.oo v1.24.0 (2020-08-26 16:11:58 UTC) successfully loaded. See ?R.oo for help.\n",
            "\n",
            "\n",
            "Attaching package: ‘R.oo’\n",
            "\n",
            "\n",
            "The following object is masked from ‘package:R.methodsS3’:\n",
            "\n",
            "    throw\n",
            "\n",
            "\n",
            "The following object is masked from ‘package:keras’:\n",
            "\n",
            "    compile\n",
            "\n",
            "\n",
            "The following objects are masked from ‘package:methods’:\n",
            "\n",
            "    getClasses, getMethods\n",
            "\n",
            "\n",
            "The following objects are masked from ‘package:base’:\n",
            "\n",
            "    attach, detach, load, save\n",
            "\n",
            "\n",
            "R.utils v2.10.1 (2020-08-26 22:50:31 UTC) successfully loaded. See ?R.utils for help.\n",
            "\n",
            "\n",
            "Attaching package: ‘R.utils’\n",
            "\n",
            "\n",
            "The following object is masked from ‘package:tidyr’:\n",
            "\n",
            "    extract\n",
            "\n",
            "\n",
            "The following object is masked from ‘package:keras’:\n",
            "\n",
            "    evaluate\n",
            "\n",
            "\n",
            "The following object is masked from ‘package:utils’:\n",
            "\n",
            "    timestamp\n",
            "\n",
            "\n",
            "The following objects are masked from ‘package:base’:\n",
            "\n",
            "    cat, commandArgs, getOption, inherits, isOpen, nullfile, parse,\n",
            "    warnings\n",
            "\n",
            "\n",
            "Please point your browser to the following url: \n",
            "\n",
            "https://accounts.google.com/o/oauth2/auth?client_id=603366585132-dpeg5tt0et3go5of2374d83ifevk5086.apps.googleusercontent.com&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdrive%20https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fuserinfo.email&redirect_uri=urn%3Aietf%3Awg%3Aoauth%3A2.0%3Aoob&response_type=code\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Enter authorization code: 4/1AY0e-g5rrOGPsa-5Jo_mg9GbxaXjyTCTFUyTkTFbJQbWDfgO__fu0zu_bpY\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 128
        },
        "id": "ZIvb8Wsg4jPc",
        "outputId": "ec6836fc-65c3-4e3c-a2d2-dd1c9f93cbfc"
      },
      "source": [
        "flowersInfo <- drive_find(\"newFlowers\\\\.zip\", n_max = 1)\n",
        "flowersInfo[,1:2]\n",
        "flowersID <- flowersInfo$id"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  name           id                               \n",
              "1 newFlowers.zip 1BD_NgV6hytm-D4xtO4JPzwExPcGyh5MI"
            ],
            "text/latex": "A tibble: 1 × 2\n\\begin{tabular}{ll}\n name & id\\\\\n <chr> & <chr>\\\\\n\\hline\n\t newFlowers.zip & 1BD\\_NgV6hytm-D4xtO4JPzwExPcGyh5MI\\\\\n\\end{tabular}\n",
            "text/markdown": "\nA tibble: 1 × 2\n\n| name &lt;chr&gt; | id &lt;chr&gt; |\n|---|---|\n| newFlowers.zip | 1BD_NgV6hytm-D4xtO4JPzwExPcGyh5MI |\n\n",
            "text/html": [
              "<table class=\"dataframe\">\n",
              "<caption>A tibble: 1 × 2</caption>\n",
              "<thead>\n",
              "\t<tr><th scope=col>name</th><th scope=col>id</th></tr>\n",
              "\t<tr><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th></tr>\n",
              "</thead>\n",
              "<tbody>\n",
              "\t<tr><td>newFlowers.zip</td><td>1BD_NgV6hytm-D4xtO4JPzwExPcGyh5MI</td></tr>\n",
              "</tbody>\n",
              "</table>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "id": "gqpsbo6C5G1B",
        "outputId": "81ecd0ec-943c-4680-b512-a2ddf5781608"
      },
      "source": [
        "drive_download(file=as_id(flowersID))\n",
        "dir()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "File downloaded:\n",
            "  * newFlowers.zip\n",
            "Saved locally as:\n",
            "  * newFlowers.zip\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] \"newFlowers.zip\" \"sample_data\"   "
            ],
            "text/latex": "\\begin{enumerate*}\n\\item 'newFlowers.zip'\n\\item 'sample\\_data'\n\\end{enumerate*}\n",
            "text/markdown": "1. 'newFlowers.zip'\n2. 'sample_data'\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>'newFlowers.zip'</li><li>'sample_data'</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FGrJbh4cbj-c"
      },
      "source": [
        "detach(\"package:R.utils\")\n",
        "detach(\"package:R.oo\")"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gNADMsEXnimW"
      },
      "source": [
        "## set up data and directories"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rFYsX3OITuFV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "c96b16f9-fd97-4ee1-9864-792ffdb8065f"
      },
      "source": [
        "utils::unzip(\"newFlowers.zip\")\n",
        "flowers <- dir(path = \"newFlowers\")\n",
        "flowers"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] \"daisy\"     \"dandelion\" \"rose\"      \"sunflower\" \"tulip\"    "
            ],
            "text/latex": "\\begin{enumerate*}\n\\item 'daisy'\n\\item 'dandelion'\n\\item 'rose'\n\\item 'sunflower'\n\\item 'tulip'\n\\end{enumerate*}\n",
            "text/markdown": "1. 'daisy'\n2. 'dandelion'\n3. 'rose'\n4. 'sunflower'\n5. 'tulip'\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>'daisy'</li><li>'dandelion'</li><li>'rose'</li><li>'sunflower'</li><li>'tulip'</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GuLtCI_pYQtD"
      },
      "source": [
        "#tmp as a holder for the T/F that come out of these\n",
        "\n",
        "tmp <- if(dir.exists(\"train\")) file.remove(dir(path = \"train/\", full.names = TRUE, recursive = TRUE))\n",
        "tmp <- if(dir.exists(\"test\")) file.remove(dir(path = \"test/\", full.names = TRUE, recursive = TRUE))\n",
        "tmp <- if(dir.exists(\"validation\")) file.remove(dir(path = \"validation/\", full.names = TRUE, recursive = TRUE))\n",
        "\n",
        "tmp <- if(!dir.exists(\"train\")) dir.create(\"train/\")\n",
        "tmp <- if(!dir.exists(\"test\")) dir.create(\"test/\")\n",
        "tmp <- if(!dir.exists(\"validation\")) dir.create(\"validation/\")\n",
        "\n",
        "tmp <- map(flowers, ~ if(!dir.exists(file.path(\"train\", .x)))\n",
        "                  dir.create(file.path(\"train\", .x ))\n",
        "                  )\n",
        "    \n",
        "tmp <- map(flowers, ~ if(!dir.exists(file.path(\"test\", .x)))\n",
        "                  dir.create(file.path(\"test\", .x ))\n",
        "                  )\n",
        "    \n",
        "tmp <- map(flowers, ~ if(!dir.exists(file.path(\"validation\", .x)))\n",
        "                  dir.create(file.path(\"validation\", .x ))\n",
        "                  )\n",
        "\n"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mp0VAVHNZLPb"
      },
      "source": [
        "# populate directories\n",
        "set.seed(2023)\n",
        "sourcedir <- \"newFlowers\"\n",
        "for(flower in flowers) {\n",
        "  files <- sample(dir(file.path(sourcedir, flower))) #randomizes\n",
        "    for(i in 1:length(files)) {\n",
        "      if (i/length(files) <= 0.2 ) destination <- \"validation\"\n",
        "      if (i/length(files) > 0.2 & i/length(files) <= 0.4) destination <- \"test\"\n",
        "      if (i/length(files) > 0.4 ) destination <- \"train\"\n",
        "      file.copy(from = file.path(sourcedir, flower, files[i]), \n",
        "                to = file.path( destination, flower, files[i]))\n",
        "    } # i loop              \n",
        "} # flower loop"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "id": "sToBrTViYhij",
        "outputId": "d6891f70-17c6-4db0-c535-c7590bf4cda2"
      },
      "source": [
        "head(dir(\"train/rose\"))\n",
        "head(dir(\"test/rose\"))\n",
        "head(dir(\"validation/rose\"))\n",
        "\n",
        "length(dir(\"test/rose\"))\n",
        "length(dir(\"train/rose\"))\n",
        "length(dir(\"validation/rose\"))\n"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] \"10090824183_d02c613f10_m.jpg\" \"110472418_87b6a3aa98_m.jpg\"  \n",
              "[3] \"11102341464_508d558dfc_n.jpg\" \"11233672494_d8bf0a3dbf_n.jpg\"\n",
              "[5] \"118974357_0faa23cce9_n.jpg\"   \"11944957684_2cc806276e.jpg\"  "
            ],
            "text/latex": "\\begin{enumerate*}\n\\item '10090824183\\_d02c613f10\\_m.jpg'\n\\item '110472418\\_87b6a3aa98\\_m.jpg'\n\\item '11102341464\\_508d558dfc\\_n.jpg'\n\\item '11233672494\\_d8bf0a3dbf\\_n.jpg'\n\\item '118974357\\_0faa23cce9\\_n.jpg'\n\\item '11944957684\\_2cc806276e.jpg'\n\\end{enumerate*}\n",
            "text/markdown": "1. '10090824183_d02c613f10_m.jpg'\n2. '110472418_87b6a3aa98_m.jpg'\n3. '11102341464_508d558dfc_n.jpg'\n4. '11233672494_d8bf0a3dbf_n.jpg'\n5. '118974357_0faa23cce9_n.jpg'\n6. '11944957684_2cc806276e.jpg'\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>'10090824183_d02c613f10_m.jpg'</li><li>'110472418_87b6a3aa98_m.jpg'</li><li>'11102341464_508d558dfc_n.jpg'</li><li>'11233672494_d8bf0a3dbf_n.jpg'</li><li>'118974357_0faa23cce9_n.jpg'</li><li>'11944957684_2cc806276e.jpg'</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] \"10894627425_ec76bbc757_n.jpg\" \"12238827553_cf427bfd51_n.jpg\"\n",
              "[3] \"12395698413_c0388278f7.jpg\"   \"13264214185_d6aa79b3bd.jpg\"  \n",
              "[5] \"13929462317_96342a9a44.jpg\"   \"14019883858_e5d2a0ec10_n.jpg\""
            ],
            "text/latex": "\\begin{enumerate*}\n\\item '10894627425\\_ec76bbc757\\_n.jpg'\n\\item '12238827553\\_cf427bfd51\\_n.jpg'\n\\item '12395698413\\_c0388278f7.jpg'\n\\item '13264214185\\_d6aa79b3bd.jpg'\n\\item '13929462317\\_96342a9a44.jpg'\n\\item '14019883858\\_e5d2a0ec10\\_n.jpg'\n\\end{enumerate*}\n",
            "text/markdown": "1. '10894627425_ec76bbc757_n.jpg'\n2. '12238827553_cf427bfd51_n.jpg'\n3. '12395698413_c0388278f7.jpg'\n4. '13264214185_d6aa79b3bd.jpg'\n5. '13929462317_96342a9a44.jpg'\n6. '14019883858_e5d2a0ec10_n.jpg'\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>'10894627425_ec76bbc757_n.jpg'</li><li>'12238827553_cf427bfd51_n.jpg'</li><li>'12395698413_c0388278f7.jpg'</li><li>'13264214185_d6aa79b3bd.jpg'</li><li>'13929462317_96342a9a44.jpg'</li><li>'14019883858_e5d2a0ec10_n.jpg'</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] \"102501987_3cdb8e5394_n.jpg\"   \"10503217854_e66a804309.jpg\"  \n",
              "[3] \"12240303_80d87f77a3_n.jpg\"    \"12243068283_ee4c2683e2_n.jpg\"\n",
              "[5] \"12243069253_e512464095_n.jpg\" \"123128873_546b8b7355_n.jpg\"  "
            ],
            "text/latex": "\\begin{enumerate*}\n\\item '102501987\\_3cdb8e5394\\_n.jpg'\n\\item '10503217854\\_e66a804309.jpg'\n\\item '12240303\\_80d87f77a3\\_n.jpg'\n\\item '12243068283\\_ee4c2683e2\\_n.jpg'\n\\item '12243069253\\_e512464095\\_n.jpg'\n\\item '123128873\\_546b8b7355\\_n.jpg'\n\\end{enumerate*}\n",
            "text/markdown": "1. '102501987_3cdb8e5394_n.jpg'\n2. '10503217854_e66a804309.jpg'\n3. '12240303_80d87f77a3_n.jpg'\n4. '12243068283_ee4c2683e2_n.jpg'\n5. '12243069253_e512464095_n.jpg'\n6. '123128873_546b8b7355_n.jpg'\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>'102501987_3cdb8e5394_n.jpg'</li><li>'10503217854_e66a804309.jpg'</li><li>'12240303_80d87f77a3_n.jpg'</li><li>'12243068283_ee4c2683e2_n.jpg'</li><li>'12243069253_e512464095_n.jpg'</li><li>'123128873_546b8b7355_n.jpg'</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 146"
            ],
            "text/latex": "146",
            "text/markdown": "146",
            "text/html": [
              "146"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 437"
            ],
            "text/latex": "437",
            "text/markdown": "437",
            "text/html": [
              "437"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 145"
            ],
            "text/latex": "145",
            "text/markdown": "145",
            "text/html": [
              "145"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1lV0poTqeBzn"
      },
      "source": [
        "## Set up image generators 1\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pTcve4NUeGOu"
      },
      "source": [
        "rescale_datagen <- image_data_generator(rescale = 1/255)\n",
        "\n",
        "\n",
        "train_generator <- flow_images_from_directory(\n",
        "    \"train\",\n",
        "    rescale_datagen,\n",
        "    batch_size=16,\n",
        "    target_size=c(150,150)\n",
        ")\n",
        "\n",
        "validation_generator <- flow_images_from_directory(\n",
        "    \"validation\",\n",
        "    rescale_datagen,\n",
        "    batch_size=16,\n",
        "    target_size=c(150,150)\n",
        ")\n",
        "\n",
        "test_generator <- flow_images_from_directory(\n",
        "    \"test\",\n",
        "    rescale_datagen,\n",
        "    batch_size=16,\n",
        "    target_size=c(150,150)\n",
        "\n",
        ")"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wspRbAr4eUTj"
      },
      "source": [
        "## Set up Model 1"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "02W4tooZeS27",
        "outputId": "dd69a02b-ed98-4602-e907-ad6965995f4a"
      },
      "source": [
        "model1 <- keras_model_sequential() %>%\n",
        "    layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = \"relu\", input_shape = c(150,150,3)) %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_flatten() %>%\n",
        "    layer_dense(units = 512, activation = \"relu\") %>%\n",
        "    layer_dense(units = length(flowers), activation = \"softmax\")\n",
        "\n",
        "# compile it\n",
        "model1 %>% compile(\n",
        "    loss = \"categorical_crossentropy\",\n",
        "    optimizer = \"rmsprop\",\n",
        "    metrics = \"acc\"\n",
        "    )\n",
        "\n",
        "summary(model1)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "________________________________________________________________________________\n",
            "Layer (type)                        Output Shape                    Param #     \n",
            "================================================================================\n",
            "conv2d_3 (Conv2D)                   (None, 148, 148, 32)            896         \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2D)      (None, 74, 74, 32)              0           \n",
            "________________________________________________________________________________\n",
            "conv2d_2 (Conv2D)                   (None, 72, 72, 64)              18496       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2D)      (None, 36, 36, 64)              0           \n",
            "________________________________________________________________________________\n",
            "conv2d_1 (Conv2D)                   (None, 34, 34, 128)             73856       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2D)      (None, 17, 17, 128)             0           \n",
            "________________________________________________________________________________\n",
            "conv2d (Conv2D)                     (None, 15, 15, 128)             147584      \n",
            "________________________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D)        (None, 7, 7, 128)               0           \n",
            "________________________________________________________________________________\n",
            "flatten (Flatten)                   (None, 6272)                    0           \n",
            "________________________________________________________________________________\n",
            "dense_1 (Dense)                     (None, 512)                     3211776     \n",
            "________________________________________________________________________________\n",
            "dense (Dense)                       (None, 5)                       2565        \n",
            "================================================================================\n",
            "Total params: 3,455,173\n",
            "Trainable params: 3,455,173\n",
            "Non-trainable params: 0\n",
            "________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "id": "oaUJtpEQeeHI",
        "outputId": "4be59ac1-148b-4a05-c83e-b4611101f77f"
      },
      "source": [
        "# first: how many images do we have?\n",
        "length(dir(\"train\", recursive = TRUE))\n",
        "\n",
        "# 2482, and we are using a batch size of 32, so we should use\n",
        "2482/16 #155 steps per epoch\n",
        "\n",
        "# how many validation images?\n",
        "length(dir(\"validation\", recursive = TRUE)) #825\n",
        "825/16 #51 validation steps"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 2482"
            ],
            "text/latex": "2482",
            "text/markdown": "2482",
            "text/html": [
              "2482"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 155.125"
            ],
            "text/latex": "155.125",
            "text/markdown": "155.125",
            "text/html": [
              "155.125"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 825"
            ],
            "text/latex": "825",
            "text/markdown": "825",
            "text/html": [
              "825"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 51.5625"
            ],
            "text/latex": "51.5625",
            "text/markdown": "51.5625",
            "text/html": [
              "51.5625"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        },
        "id": "DUQSOED9e6_t",
        "outputId": "bbd47239-b765-4e7b-fbd8-e29c989e6183"
      },
      "source": [
        "system.time(history <- model1 %>% fit(\n",
        "        train_generator,\n",
        "        steps_per_epoch = 155,\n",
        "        epochs = 30,\n",
        "        validation_data = validation_generator,\n",
        "        validation_steps = 51))\n",
        "\n",
        "save_model_hdf5(model1, \"flowers_model1_ep30.hdf5\")\n",
        "drive_put(\"flowers_model1_ep30.hdf5\", \"DeepLearning/Flowers/flowers_model1_ep30.hdf5\")\n",
        "file.remove(\"flowers_model1_ep30.hdf5\")\n",
        "\n",
        "history"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Warning message in fit_generator(., train_generator, steps_per_epoch = 155, epochs = 30, :\n",
            "“`fit_generator` is deprecated. Use `fit` instead, it now accept generators.”\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "   user  system elapsed \n",
              "393.554  50.074 346.866 "
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\n",
              "Final epoch (plot to see history):\n",
              "    loss: 0.08353\n",
              "     acc: 0.985\n",
              "val_loss: 5.117\n",
              " val_acc: 0.6569 "
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 471
        },
        "id": "gwtggdoRfgFS",
        "outputId": "8fa37e85-50f5-4d70-92fc-bbeb89592c6f"
      },
      "source": [
        "plot(history)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "`geom_smooth()` using formula 'y ~ x'\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAIAAAByhViMAAAACXBIWXMAABJ0AAASdAHeZh94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\ngElQ7AAAAEyCYgcAAGASFDsAAACT+H9c5IAD6Zg8pAAAAABJRU5ErkJggg==",
            "text/plain": [
              "plot without title"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 420,
              "height": 420
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HhRT0Uo1hw6h"
      },
      "source": [
        "OK maybe 7 epochs is about where validation maxes out.  So refit for 7 epochs and check test.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        },
        "id": "8HWLiFj0hvLg",
        "outputId": "6e5835c6-2131-4e6d-c214-7969b8005791"
      },
      "source": [
        "model1 <- keras_model_sequential() %>%\n",
        "    layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = \"relu\", input_shape = c(150,150,3)) %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_flatten() %>%\n",
        "    layer_dense(units = 512, activation = \"relu\") %>%\n",
        "    layer_dense(units = length(flowers), activation = \"softmax\")\n",
        "\n",
        "# compile it\n",
        "model1 %>% compile(\n",
        "    loss = \"categorical_crossentropy\",\n",
        "    optimizer = \"rmsprop\",\n",
        "    metrics = \"acc\"\n",
        "    )\n",
        "\n",
        "system.time(history <- model1 %>% fit_generator(\n",
        "        train_generator,\n",
        "        steps_per_epoch = 155,\n",
        "        epochs = 7,\n",
        "        validation_data = validation_generator,\n",
        "        validation_steps = 51))\n",
        "\n",
        "save_model_hdf5(model1, \"flowers_model1_ep7.hdf5\")\n",
        "drive_put(\"flowers_model1_ep7.hdf5\", \"DeepLearning/Flowers/flowers_model1_ep7.hdf5\")\n",
        "file.remove(\"flowers_model1_ep7.hdf5\")\n",
        "\n",
        "history"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Warning message in fit_generator(., train_generator, steps_per_epoch = 155, epochs = 7, :\n",
            "“`fit_generator` is deprecated. Use `fit` instead, it now accept generators.”\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "   user  system elapsed \n",
              " 74.133   7.003  62.047 "
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\n",
              "Final epoch (plot to see history):\n",
              "    loss: 0.4267\n",
              "     acc: 0.8483\n",
              "val_loss: 1.077\n",
              " val_acc: 0.6961 "
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "id": "sh8C4pirihdO",
        "outputId": "64fce50f-760d-4ae8-f347-25bd38e9623d"
      },
      "source": [
        "results1 <- model1 %>% evaluate(test_generator)\n",
        "results1 %>% round(3)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              " loss   acc \n",
              "1.003 0.680 "
            ],
            "text/latex": "\\begin{description*}\n\\item[loss] 1.003\n\\item[acc] 0.68\n\\end{description*}\n",
            "text/markdown": "loss\n:   1.003acc\n:   0.68\n\n",
            "text/html": [
              "<style>\n",
              ".dl-inline {width: auto; margin:0; padding: 0}\n",
              ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
              ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
              ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
              "</style><dl class=dl-inline><dt>loss</dt><dd>1.003</dd><dt>acc</dt><dd>0.68</dd></dl>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RwQ8PZ5ikPYb"
      },
      "source": [
        "## Set up image generators 2\n",
        "\n",
        "Here use larger images, larger batch size, and an additional model layer\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "emeHJHuCkPYt"
      },
      "source": [
        "rescale_datagen <- image_data_generator(rescale = 1/255)\n",
        "\n",
        "\n",
        "train_generator2 <- flow_images_from_directory(\n",
        "    \"train\",\n",
        "    rescale_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        ")\n",
        "\n",
        "validation_generator2 <- flow_images_from_directory(\n",
        "    \"validation\",\n",
        "    rescale_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        ")\n",
        "\n",
        "test_generator2 <- flow_images_from_directory(\n",
        "    \"test\",\n",
        "    rescale_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        "\n",
        ")"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "grVLvvFXkPYu"
      },
      "source": [
        "## Set up Model 2"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ZXt8QmgkPYu",
        "outputId": "8d9c053b-06ec-4050-d404-d723ec99f401"
      },
      "source": [
        "model2 <- keras_model_sequential() %>%\n",
        "    layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = \"relu\", input_shape = c(256,256,3)) %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_flatten() %>%\n",
        "    layer_dense(units = 512, activation = \"relu\") %>%\n",
        "    layer_dense(units = length(flowers), activation = \"softmax\")\n",
        "\n",
        "# compile it\n",
        "model2 %>% compile(\n",
        "    loss = \"categorical_crossentropy\",\n",
        "    optimizer = \"rmsprop\",\n",
        "    metrics = \"acc\"\n",
        "    )\n",
        "\n",
        "summary(model2)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_5\"\n",
            "________________________________________________________________________________\n",
            "Layer (type)                        Output Shape                    Param #     \n",
            "================================================================================\n",
            "conv2d_24 (Conv2D)                  (None, 254, 254, 32)            896         \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_24 (MaxPooling2D)     (None, 127, 127, 32)            0           \n",
            "________________________________________________________________________________\n",
            "conv2d_23 (Conv2D)                  (None, 125, 125, 64)            18496       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_23 (MaxPooling2D)     (None, 62, 62, 64)              0           \n",
            "________________________________________________________________________________\n",
            "conv2d_22 (Conv2D)                  (None, 60, 60, 128)             73856       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_22 (MaxPooling2D)     (None, 30, 30, 128)             0           \n",
            "________________________________________________________________________________\n",
            "conv2d_21 (Conv2D)                  (None, 28, 28, 128)             147584      \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_21 (MaxPooling2D)     (None, 14, 14, 128)             0           \n",
            "________________________________________________________________________________\n",
            "conv2d_20 (Conv2D)                  (None, 12, 12, 128)             147584      \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_20 (MaxPooling2D)     (None, 6, 6, 128)               0           \n",
            "________________________________________________________________________________\n",
            "flatten_5 (Flatten)                 (None, 4608)                    0           \n",
            "________________________________________________________________________________\n",
            "dense_11 (Dense)                    (None, 512)                     2359808     \n",
            "________________________________________________________________________________\n",
            "dense_10 (Dense)                    (None, 5)                       2565        \n",
            "================================================================================\n",
            "Total params: 2,750,789\n",
            "Trainable params: 2,750,789\n",
            "Non-trainable params: 0\n",
            "________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "id": "HR5BVh54kPYv",
        "outputId": "026476e5-4453-45c0-f750-a4214d3fd429"
      },
      "source": [
        "# first: how many images do we have?\n",
        "length(dir(\"train\", recursive = TRUE))\n",
        "\n",
        "# 2482, and we are using a batch size of 32, so we should use\n",
        "2482/32 #78 steps per epoch\n",
        "\n",
        "# how many validation images?\n",
        "length(dir(\"validation\", recursive = TRUE)) #825\n",
        "825/32 #26 validation steps"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 2482"
            ],
            "text/latex": "2482",
            "text/markdown": "2482",
            "text/html": [
              "2482"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 77.5625"
            ],
            "text/latex": "77.5625",
            "text/markdown": "77.5625",
            "text/html": [
              "77.5625"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 825"
            ],
            "text/latex": "825",
            "text/markdown": "825",
            "text/html": [
              "825"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 25.78125"
            ],
            "text/latex": "25.78125",
            "text/markdown": "25.78125",
            "text/html": [
              "25.78125"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "id": "OcN4qKFtkPYx",
        "outputId": "6879f895-5957-455e-9d2c-d81830e912e3"
      },
      "source": [
        "system.time(history2 <- model2 %>% fit(\n",
        "        train_generator2,\n",
        "        steps_per_epoch = 78,\n",
        "        epochs = 30,\n",
        "        validation_data = validation_generator2,\n",
        "        validation_steps = 26))\n",
        "\n",
        "history2"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "   user  system elapsed \n",
              "396.006  45.980 336.075 "
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\n",
              "Final epoch (plot to see history):\n",
              "    loss: 0.05112\n",
              "     acc: 0.9855\n",
              "val_loss: 5.232\n",
              " val_acc: 0.6473 "
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 471
        },
        "id": "7sgwY5adkPYy",
        "outputId": "27d69bef-5c77-49f8-8a0f-8832be89de8a"
      },
      "source": [
        "plot(history2)"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "`geom_smooth()` using formula 'y ~ x'\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAIAAAByhViMAAAACXBIWXMAABJ0AAASdAHeZh94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\nrkJggg==",
            "text/plain": [
              "plot without title"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 420,
              "height": 420
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vQgugQbbndGG"
      },
      "source": [
        "Refit at 5 epochs"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "id": "CuPV4nrNnfMJ",
        "outputId": "72c98732-6b80-47a9-fef2-c6c76cb52b69"
      },
      "source": [
        "model2 <- keras_model_sequential() %>%\n",
        "    layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = \"relu\", input_shape = c(256,256,3)) %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_flatten() %>%\n",
        "    layer_dense(units = 512, activation = \"relu\") %>%\n",
        "    layer_dense(units = length(flowers), activation = \"softmax\")\n",
        "\n",
        "# compile it\n",
        "model2 %>% compile(\n",
        "    loss = \"categorical_crossentropy\",\n",
        "    optimizer = \"rmsprop\",\n",
        "    metrics = \"acc\"\n",
        "    )\n",
        "\n",
        "system.time(history2 <- model2 %>% fit(\n",
        "        train_generator2,\n",
        "        steps_per_epoch = 78,\n",
        "        epochs = 5,\n",
        "        validation_data = validation_generator2,\n",
        "        validation_steps = 26))\n",
        "\n",
        "save_model_hdf5(model2, \"flowers_model2_ep5.hdf5\")\n",
        "drive_put(\"flowers_model2_ep5.hdf5\", \"DeepLearning/Flowers/flowers_model2_ep5.hdf5\")\n",
        "file.remove(\"flowers_model2_ep5.hdf5\")\n",
        "\n",
        "history2\n"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "   user  system elapsed \n",
              " 66.272   7.353  55.806 "
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\n",
              "Final epoch (plot to see history):\n",
              "    loss: 0.7979\n",
              "     acc: 0.6906\n",
              "val_loss: 0.8985\n",
              " val_acc: 0.6691 "
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "id": "hTlViW1gnpsc",
        "outputId": "cd549f06-620b-4bab-afe9-4c60bfb5c122"
      },
      "source": [
        "results2 <- model2 %>% evaluate(test_generator2)\n",
        "results2 %>% round(3)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              " loss   acc \n",
              "0.867 0.665 "
            ],
            "text/latex": "\\begin{description*}\n\\item[loss] 0.867\n\\item[acc] 0.665\n\\end{description*}\n",
            "text/markdown": "loss\n:   0.867acc\n:   0.665\n\n",
            "text/html": [
              "<style>\n",
              ".dl-inline {width: auto; margin:0; padding: 0}\n",
              ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
              ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
              ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
              "</style><dl class=dl-inline><dt>loss</dt><dd>0.867</dd><dt>acc</dt><dd>0.665</dd></dl>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s8X3RSCZkPYz"
      },
      "source": [
        ""
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yPbeUSfb7q3Y"
      },
      "source": [
        "## Set up image generators 3: augmentation\n",
        "\n",
        "Like model 2, but with augmentation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5Fp2KrE87q3k"
      },
      "source": [
        "rescale_datagen <- image_data_generator(rescale = 1/255)\n",
        "\n",
        "augment_datagen <- image_data_generator(rescale = 1/255,\n",
        "  rotation_range=90,\n",
        "  width_shift_range=.2,\n",
        "  height_shift_range=.2,\n",
        "  shear_range=.2,\n",
        "  zoom_range=.2,\n",
        "  horizontal_flip=TRUE)\n",
        "\n",
        "train_generator3 <- flow_images_from_directory(\n",
        "    \"train\",\n",
        "    augment_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        ")\n",
        "\n",
        "validation_generator3 <- flow_images_from_directory(\n",
        "    \"validation\",\n",
        "    rescale_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        ")\n",
        "\n",
        "test_generator3 <- flow_images_from_directory(\n",
        "    \"test\",\n",
        "    rescale_datagen,\n",
        "    batch_size=32,\n",
        "    target_size=c(256,256)\n",
        "\n",
        ")"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P3drqbY57q3l"
      },
      "source": [
        "## Set up Model 3"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k5vizl4D7q3l",
        "outputId": "38bcebf0-4d12-42d7-d714-a18cb5b7221c"
      },
      "source": [
        "model3 <- keras_model_sequential() %>%\n",
        "    layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = \"relu\", input_shape = c(256,256,3)) %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_conv_2d(filters = 128, kernel_size = c(3,3), activation = \"relu\") %>%\n",
        "    layer_max_pooling_2d(pool_size=c(2,2)) %>%\n",
        "    layer_flatten() %>%\n",
        "    layer_dense(units = 512, activation = \"relu\") %>%\n",
        "    layer_dense(units = length(flowers), activation = \"softmax\")\n",
        "\n",
        "# compile it\n",
        "model3 %>% compile(\n",
        "    loss = \"categorical_crossentropy\",\n",
        "    optimizer = \"rmsprop\",\n",
        "    metrics = \"acc\"\n",
        "    )\n",
        "\n",
        "summary(model3)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "________________________________________________________________________________\n",
            "Layer (type)                        Output Shape                    Param #     \n",
            "================================================================================\n",
            "conv2d_4 (Conv2D)                   (None, 254, 254, 32)            896         \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2D)      (None, 127, 127, 32)            0           \n",
            "________________________________________________________________________________\n",
            "conv2d_3 (Conv2D)                   (None, 125, 125, 64)            18496       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2D)      (None, 62, 62, 64)              0           \n",
            "________________________________________________________________________________\n",
            "conv2d_2 (Conv2D)                   (None, 60, 60, 128)             73856       \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2D)      (None, 30, 30, 128)             0           \n",
            "________________________________________________________________________________\n",
            "conv2d_1 (Conv2D)                   (None, 28, 28, 128)             147584      \n",
            "________________________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2D)      (None, 14, 14, 128)             0           \n",
            "________________________________________________________________________________\n",
            "conv2d (Conv2D)                     (None, 12, 12, 128)             147584      \n",
            "________________________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D)        (None, 6, 6, 128)               0           \n",
            "________________________________________________________________________________\n",
            "flatten (Flatten)                   (None, 4608)                    0           \n",
            "________________________________________________________________________________\n",
            "dense_1 (Dense)                     (None, 512)                     2359808     \n",
            "________________________________________________________________________________\n",
            "dense (Dense)                       (None, 5)                       2565        \n",
            "================================================================================\n",
            "Total params: 2,750,789\n",
            "Trainable params: 2,750,789\n",
            "Non-trainable params: 0\n",
            "________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XPuSrKVy5e6j"
      },
      "source": [
        "### Train the model.  Here I am going to save it out every 10 epochs.\n",
        "\n",
        "Previous runs showed that it was still improving at epoch 50, so will run 150 and save every 10 so that I don't need to refit."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        },
        "id": "IdBpWwov7q3m",
        "outputId": "7d90ed06-9e0f-46ea-cdf1-7416fef15afa"
      },
      "source": [
        "total_epochs <- 150 \n",
        "save_interval <- 10\n",
        "#i <- 1\n",
        "save_seq <- seq(from=save_interval, to=total_epochs, by=save_interval)\n",
        "\n",
        "history3_list <- list()\n",
        "\n",
        "system.time(for(i in save_seq) {\n",
        "  history3_tmp <- model3 %>% fit(\n",
        "        train_generator3,\n",
        "        steps_per_epoch = 78,\n",
        "        epochs = save_interval,\n",
        "        validation_data = validation_generator3,\n",
        "        validation_steps = 26)\n",
        "\n",
        "  history3_list[[i/save_interval]] <- history3_tmp$metrics # all we need is metrics\n",
        "\n",
        "  file_name <- str_c(\"flowers_model3_ep\", i, \".hdf5\")\n",
        "  save_model_hdf5(model3, file_name)\n",
        "  drive_put(file_name, str_c(\"DeepLearning/Flowers/\", file_name), verbose=FALSE)\n",
        "  file.remove(file_name)\n",
        "  save(history3_list, file=\"history3_list.Rdata\")\n",
        "  drive_put(\"history3_list.Rdata\", \"DeepLearning/Flowers/history3_list.Rdata\", verbose=FALSE)\n",
        "  print(str_c(\"completed epoch: \", i))\n",
        "})\n"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1] \"completed epoch: 10\"\n",
            "[1] \"completed epoch: 20\"\n",
            "[1] \"completed epoch: 30\"\n",
            "[1] \"completed epoch: 40\"\n",
            "[1] \"completed epoch: 50\"\n",
            "[1] \"completed epoch: 60\"\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Auto-refreshing stale OAuth token.\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[1] \"completed epoch: 70\"\n",
            "[1] \"completed epoch: 80\"\n",
            "[1] \"completed epoch: 90\"\n",
            "[1] \"completed epoch: 100\"\n",
            "[1] \"completed epoch: 110\"\n",
            "[1] \"completed epoch: 120\"\n",
            "[1] \"completed epoch: 130\"\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Auto-refreshing stale OAuth token.\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[1] \"completed epoch: 140\"\n",
            "[1] \"completed epoch: 150\"\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "    user   system  elapsed \n",
              "6691.726  151.911 8192.831 "
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "elOpFhDM7cm7",
        "outputId": "a948dba5-15e9-4942-dc63-e3b99cc0c50a"
      },
      "source": [
        "if(!file.exists(\"history3_list.Rdata\")) {\n",
        "  drive_download(\"history3_list.Rdata\")\n",
        "}\n",
        "str(history3_list)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "List of 15\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 1.583 1.245 1.099 1.011 0.964 ...\n",
            "  ..$ acc     : num [1:10] 0.323 0.479 0.558 0.604 0.63 ...\n",
            "  ..$ val_loss: num [1:10] 1.251 1.238 0.939 1.046 0.907 ...\n",
            "  ..$ val_acc : num [1:10] 0.49 0.455 0.628 0.605 0.635 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.805 0.78 0.772 0.731 0.763 ...\n",
            "  ..$ acc     : num [1:10] 0.687 0.702 0.708 0.724 0.707 ...\n",
            "  ..$ val_loss: num [1:10] 0.778 0.807 1 0.737 0.806 ...\n",
            "  ..$ val_acc : num [1:10] 0.714 0.719 0.679 0.737 0.71 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.67 0.672 0.658 0.668 0.601 ...\n",
            "  ..$ acc     : num [1:10] 0.747 0.752 0.759 0.772 0.768 ...\n",
            "  ..$ val_loss: num [1:10] 0.713 0.647 0.818 0.63 0.754 ...\n",
            "  ..$ val_acc : num [1:10] 0.755 0.773 0.739 0.792 0.756 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.59 0.659 0.584 0.58 0.587 ...\n",
            "  ..$ acc     : num [1:10] 0.786 0.769 0.788 0.781 0.786 ...\n",
            "  ..$ val_loss: num [1:10] 0.84 0.73 0.75 0.602 1.424 ...\n",
            "  ..$ val_acc : num [1:10] 0.73 0.737 0.765 0.775 0.648 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.639 0.628 0.641 0.615 0.57 ...\n",
            "  ..$ acc     : num [1:10] 0.793 0.779 0.782 0.784 0.794 ...\n",
            "  ..$ val_loss: num [1:10] 0.596 0.637 0.697 0.687 0.847 ...\n",
            "  ..$ val_acc : num [1:10] 0.799 0.79 0.77 0.752 0.735 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.596 0.621 0.687 0.611 0.602 ...\n",
            "  ..$ acc     : num [1:10] 0.785 0.785 0.778 0.79 0.777 ...\n",
            "  ..$ val_loss: num [1:10] 0.77 0.66 0.815 1.155 0.703 ...\n",
            "  ..$ val_acc : num [1:10] 0.794 0.789 0.747 0.68 0.804 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.619 0.631 0.611 0.643 0.633 ...\n",
            "  ..$ acc     : num [1:10] 0.788 0.783 0.782 0.768 0.769 ...\n",
            "  ..$ val_loss: num [1:10] 0.893 0.718 0.674 0.744 0.669 ...\n",
            "  ..$ val_acc : num [1:10] 0.747 0.755 0.812 0.782 0.782 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.645 0.67 0.63 0.622 0.758 ...\n",
            "  ..$ acc     : num [1:10] 0.779 0.769 0.784 0.787 0.77 ...\n",
            "  ..$ val_loss: num [1:10] 0.965 0.668 0.861 0.727 0.806 ...\n",
            "  ..$ val_acc : num [1:10] 0.744 0.785 0.805 0.736 0.761 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.635 0.629 0.64 0.601 0.738 ...\n",
            "  ..$ acc     : num [1:10] 0.787 0.782 0.775 0.793 0.767 ...\n",
            "  ..$ val_loss: num [1:10] 0.825 1.281 1.029 0.778 0.721 ...\n",
            "  ..$ val_acc : num [1:10] 0.741 0.781 0.731 0.778 0.756 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.655 0.645 0.683 0.692 0.641 ...\n",
            "  ..$ acc     : num [1:10] 0.776 0.782 0.776 0.773 0.775 ...\n",
            "  ..$ val_loss: num [1:10] 0.895 0.861 0.751 0.566 0.708 ...\n",
            "  ..$ val_acc : num [1:10] 0.792 0.748 0.783 0.815 0.781 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.912 0.628 0.666 0.837 0.644 ...\n",
            "  ..$ acc     : num [1:10] 0.757 0.779 0.772 0.768 0.79 ...\n",
            "  ..$ val_loss: num [1:10] 0.88 0.656 0.803 0.776 0.847 ...\n",
            "  ..$ val_acc : num [1:10] 0.719 0.8 0.776 0.802 0.795 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.756 0.63 0.649 0.657 0.64 ...\n",
            "  ..$ acc     : num [1:10] 0.78 0.781 0.769 0.776 0.777 ...\n",
            "  ..$ val_loss: num [1:10] 0.724 1.05 1.047 0.944 0.862 ...\n",
            "  ..$ val_acc : num [1:10] 0.794 0.76 0.743 0.777 0.733 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.74 0.651 0.915 0.655 0.659 ...\n",
            "  ..$ acc     : num [1:10] 0.762 0.779 0.759 0.778 0.782 ...\n",
            "  ..$ val_loss: num [1:10] 1.126 0.833 0.867 0.904 1.98 ...\n",
            "  ..$ val_acc : num [1:10] 0.531 0.788 0.778 0.762 0.668 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.613 0.713 0.627 0.794 0.642 ...\n",
            "  ..$ acc     : num [1:10] 0.786 0.78 0.782 0.764 0.776 ...\n",
            "  ..$ val_loss: num [1:10] 1.129 0.865 4.098 1.446 1.238 ...\n",
            "  ..$ val_acc : num [1:10] 0.762 0.778 0.75 0.801 0.732 ...\n",
            " $ :List of 4\n",
            "  ..$ loss    : num [1:10] 0.791 0.663 0.749 0.781 2.216 ...\n",
            "  ..$ acc     : num [1:10] 0.751 0.774 0.747 0.755 0.724 ...\n",
            "  ..$ val_loss: num [1:10] 2.39 1.13 1.4 1.19 0.75 ...\n",
            "  ..$ val_acc : num [1:10] 0.616 0.605 0.758 0.768 0.765 ...\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "ATrlLoT67q3n",
        "outputId": "f181736d-e450-4dcb-a097-5803d84774dd"
      },
      "source": [
        "history3 <- bind_rows(history3_list) %>%\n",
        "  mutate(epoch=1:nrow(.))\n",
        "dim(history3)\n",
        "head(history3)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "[1] 150   5"
            ],
            "text/latex": "\\begin{enumerate*}\n\\item 150\n\\item 5\n\\end{enumerate*}\n",
            "text/markdown": "1. 150\n2. 5\n\n\n",
            "text/html": [
              "<style>\n",
              ".list-inline {list-style: none; margin:0; padding: 0}\n",
              ".list-inline>li {display: inline-block}\n",
              ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
              "</style>\n",
              "<ol class=list-inline><li>150</li><li>5</li></ol>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  loss      acc       val_loss  val_acc   epoch\n",
              "1 1.5831740 0.3227236 1.2506963 0.4896970 1    \n",
              "2 1.2448081 0.4794520 1.2377491 0.4545455 2    \n",
              "3 1.0988530 0.5584207 0.9389179 0.6278788 3    \n",
              "4 1.0111643 0.6035455 1.0459580 0.6048485 4    \n",
              "5 0.9641103 0.6301370 0.9074947 0.6351515 5    \n",
              "6 0.9292060 0.6345689 0.8676798 0.6484848 6    "
            ],
            "text/latex": "A tibble: 6 × 5\n\\begin{tabular}{lllll}\n loss & acc & val\\_loss & val\\_acc & epoch\\\\\n <dbl> & <dbl> & <dbl> & <dbl> & <int>\\\\\n\\hline\n\t 1.5831740 & 0.3227236 & 1.2506963 & 0.4896970 & 1\\\\\n\t 1.2448081 & 0.4794520 & 1.2377491 & 0.4545455 & 2\\\\\n\t 1.0988530 & 0.5584207 & 0.9389179 & 0.6278788 & 3\\\\\n\t 1.0111643 & 0.6035455 & 1.0459580 & 0.6048485 & 4\\\\\n\t 0.9641103 & 0.6301370 & 0.9074947 & 0.6351515 & 5\\\\\n\t 0.9292060 & 0.6345689 & 0.8676798 & 0.6484848 & 6\\\\\n\\end{tabular}\n",
            "text/markdown": "\nA tibble: 6 × 5\n\n| loss &lt;dbl&gt; | acc &lt;dbl&gt; | val_loss &lt;dbl&gt; | val_acc &lt;dbl&gt; | epoch &lt;int&gt; |\n|---|---|---|---|---|\n| 1.5831740 | 0.3227236 | 1.2506963 | 0.4896970 | 1 |\n| 1.2448081 | 0.4794520 | 1.2377491 | 0.4545455 | 2 |\n| 1.0988530 | 0.5584207 | 0.9389179 | 0.6278788 | 3 |\n| 1.0111643 | 0.6035455 | 1.0459580 | 0.6048485 | 4 |\n| 0.9641103 | 0.6301370 | 0.9074947 | 0.6351515 | 5 |\n| 0.9292060 | 0.6345689 | 0.8676798 | 0.6484848 | 6 |\n\n",
            "text/html": [
              "<table class=\"dataframe\">\n",
              "<caption>A tibble: 6 × 5</caption>\n",
              "<thead>\n",
              "\t<tr><th scope=col>loss</th><th scope=col>acc</th><th scope=col>val_loss</th><th scope=col>val_acc</th><th scope=col>epoch</th></tr>\n",
              "\t<tr><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;int&gt;</th></tr>\n",
              "</thead>\n",
              "<tbody>\n",
              "\t<tr><td>1.5831740</td><td>0.3227236</td><td>1.2506963</td><td>0.4896970</td><td>1</td></tr>\n",
              "\t<tr><td>1.2448081</td><td>0.4794520</td><td>1.2377491</td><td>0.4545455</td><td>2</td></tr>\n",
              "\t<tr><td>1.0988530</td><td>0.5584207</td><td>0.9389179</td><td>0.6278788</td><td>3</td></tr>\n",
              "\t<tr><td>1.0111643</td><td>0.6035455</td><td>1.0459580</td><td>0.6048485</td><td>4</td></tr>\n",
              "\t<tr><td>0.9641103</td><td>0.6301370</td><td>0.9074947</td><td>0.6351515</td><td>5</td></tr>\n",
              "\t<tr><td>0.9292060</td><td>0.6345689</td><td>0.8676798</td><td>0.6484848</td><td>6</td></tr>\n",
              "</tbody>\n",
              "</table>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 471
        },
        "id": "d0wjp7Uuof8A",
        "outputId": "d0ffb70d-5d89-490f-cb6f-fa637dfcb526"
      },
      "source": [
        "history3 %>%\n",
        "    pivot_longer(cols=-epoch) %>%\n",
        "    mutate(dataset = ifelse(str_detect(name, \"val\"), \"validation\", \"training\"),\n",
        "           type=ifelse(str_detect(name, \"acc\"), \"accuracy\", \"loss\")) %>%\n",
        "    filter(value < 3) %>% # remove a few outliers\n",
        "    ggplot(aes(x=epoch, y=value, color=dataset)) +\n",
        "        geom_smooth(se=FALSE) +\n",
        "        geom_point() +\n",
        "        facet_wrap(~type, scales = \"free_y\", ncol=1) \n",
        "\n",
        "    "
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "`geom_smooth()` using method = 'loess' and formula 'y ~ x'\n",
            "\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAIAAAByhViMAAAACXBIWXMAABJ0AAASdAHeZh94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",
            "text/plain": [
              "plot without title"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 420,
              "height": 420
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "apc4_Tdsx1AC"
      },
      "source": [
        "Looks like we max out accuracy on the validation set at about 50 epochs.  \n",
        "\n",
        "Let's load that model and see how it does"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L2uS7M8BxNsL",
        "outputId": "fe7950b6-e5dd-4aaa-ed06-994f78abf7d1"
      },
      "source": [
        "drive_download(\"DeepLearning/Flowers/flowers_model3_ep50.hdf5\")\n",
        "model3 <- load_model_hdf5(\"flowers_model3_ep50.hdf5\")"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Auto-refreshing stale OAuth token.\n",
            "\n",
            "File downloaded:\n",
            "  * flowers_model3_ep50.hdf5\n",
            "Saved locally as:\n",
            "  * flowers_model3_ep50.hdf5\n",
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "id": "SEY92nFFxWFV",
        "outputId": "efe29c73-9866-4325-f38e-a3d70fc2935c"
      },
      "source": [
        "results <- model3 %>% evaluate(test_generator3)\n",
        "results %>% round(3)\n"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              " loss   acc \n",
              "0.655 0.812 "
            ],
            "text/latex": "\\begin{description*}\n\\item[loss] 0.655\n\\item[acc] 0.812\n\\end{description*}\n",
            "text/markdown": "loss\n:   0.655acc\n:   0.812\n\n",
            "text/html": [
              "<style>\n",
              ".dl-inline {width: auto; margin:0; padding: 0}\n",
              ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
              ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
              ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
              "</style><dl class=dl-inline><dt>loss</dt><dd>0.655</dd><dt>acc</dt><dd>0.812</dd></dl>\n"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XbQS8z-oxvis"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K0RhQYrRotcZ"
      },
      "source": [
        "## Next steps:\n",
        "\n",
        "dropout layers and/or regularization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fwBWcJaOotBT"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}