{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Try out the PhytoClassUCSC model for yourself\n",
        "\n",
        "Using this notebook, you should able to clone the model repo off of Hugging Face, grab an syringe from the Santa Cruz Wharf IFCB dataset on the CalOOS Dashboard instance, and run it through the classifier.\n",
        "\n",
        "Using the __GPU Hardware Accelerator__ will significantly increase the processing time.\n",
        "\n",
        "\n",
        "### REMOVE USERNAME AND PW before publishing"
      ],
      "metadata": {
        "id": "UQP7BLJX1281"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vvpmWDgOzzau",
        "outputId": "7306bae1-982c-425a-ed6b-5e7b38808ae8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting keras_preprocessing\n",
            "  Downloading Keras_Preprocessing-1.1.2-py2.py3-none-any.whl (42 kB)\n",
            "\u001b[?25l     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/42.6 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m42.6/42.6 kB\u001b[0m \u001b[31m1.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.9.1 in /usr/local/lib/python3.10/dist-packages (from keras_preprocessing) (1.22.4)\n",
            "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.10/dist-packages (from keras_preprocessing) (1.16.0)\n",
            "Installing collected packages: keras_preprocessing\n",
            "Successfully installed keras_preprocessing-1.1.2\n",
            "Cloning into 'phytoClassUCSC'...\n",
            "remote: Enumerating objects: 53, done.\u001b[K\n",
            "remote: Counting objects: 100% (53/53), done.\u001b[K\n",
            "remote: Compressing objects: 100% (48/48), done.\u001b[K\n",
            "remote: Total 53 (delta 12), reused 0 (delta 0), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (53/53), 557.16 KiB | 1.24 MiB/s, done.\n",
            "Collecting git+https://github.com/joefutrelle/pyifcb.git\n",
            "  Cloning https://github.com/joefutrelle/pyifcb.git to /tmp/pip-req-build-h_9ypdyg\n",
            "  Running command git clone --filter=blob:none --quiet https://github.com/joefutrelle/pyifcb.git /tmp/pip-req-build-h_9ypdyg\n",
            "  Resolved https://github.com/joefutrelle/pyifcb.git to commit e7ecbd925170ac59f0a728f48a751657a4c40307\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting scipy<1.9.2 (from pyifcb==0.0.1)\n",
            "  Downloading scipy-1.9.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (43.9 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m43.9/43.9 MB\u001b[0m \u001b[31m11.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (1.5.3)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (3.8.0)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (2.27.1)\n",
            "Requirement already satisfied: Pillow in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (9.4.0)\n",
            "Collecting rectpack (from pyifcb==0.0.1)\n",
            "  Downloading rectpack-0.2.2.tar.gz (17 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: scikit-image in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (0.19.3)\n",
            "Collecting pysmb (from pyifcb==0.0.1)\n",
            "  Downloading pysmb-1.2.9.1.zip (1.4 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.4/1.4 MB\u001b[0m \u001b[31m76.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: pyyaml in /usr/local/lib/python3.10/dist-packages (from pyifcb==0.0.1) (6.0.1)\n",
            "Requirement already satisfied: numpy<1.25.0,>=1.18.5 in /usr/local/lib/python3.10/dist-packages (from scipy<1.9.2->pyifcb==0.0.1) (1.22.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->pyifcb==0.0.1) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->pyifcb==0.0.1) (2022.7.1)\n",
            "Requirement already satisfied: pyasn1 in /usr/local/lib/python3.10/dist-packages (from pysmb->pyifcb==0.0.1) (0.5.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from pysmb->pyifcb==0.0.1) (4.65.0)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->pyifcb==0.0.1) (1.26.16)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->pyifcb==0.0.1) (2023.7.22)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->pyifcb==0.0.1) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->pyifcb==0.0.1) (3.4)\n",
            "Requirement already satisfied: networkx>=2.2 in /usr/local/lib/python3.10/dist-packages (from scikit-image->pyifcb==0.0.1) (3.1)\n",
            "Requirement already satisfied: imageio>=2.4.1 in /usr/local/lib/python3.10/dist-packages (from scikit-image->pyifcb==0.0.1) (2.25.1)\n",
            "Requirement already satisfied: tifffile>=2019.7.26 in /usr/local/lib/python3.10/dist-packages (from scikit-image->pyifcb==0.0.1) (2023.7.18)\n",
            "Requirement already satisfied: PyWavelets>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-image->pyifcb==0.0.1) (1.4.1)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from scikit-image->pyifcb==0.0.1) (23.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->pyifcb==0.0.1) (1.16.0)\n",
            "Building wheels for collected packages: pyifcb, pysmb, rectpack\n",
            "  Building wheel for pyifcb (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pyifcb: filename=pyifcb-0.0.1-py3-none-any.whl size=61518 sha256=39bbb112eb2e88ba944d3a7cc51387080e1147dc634b8cd0ed95cc76a86413f6\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-gcsarmzt/wheels/61/fa/8f/c0c33addc3ecffe7d8dc392af68e3f8eb0316b2808da6897e6\n",
            "  Building wheel for pysmb (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pysmb: filename=pysmb-1.2.9.1-py3-none-any.whl size=84802 sha256=c093dd18f06490f77d7b649d26e85125f2a6941b5e4c0d39eb7eef3f2539aa28\n",
            "  Stored in directory: /root/.cache/pip/wheels/b6/13/a6/22f752798d4429d1f973f90e1fdaf8eb782a899fc691b57f48\n",
            "  Building wheel for rectpack (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for rectpack: filename=rectpack-0.2.2-py3-none-any.whl size=19333 sha256=ac7f7862793ba19a9f3b379dd5b2c8f9f7f08fd5e1c09b1bbcab36804aa79c9f\n",
            "  Stored in directory: /root/.cache/pip/wheels/e9/ea/e9/cd0237c0ccb9cb7312bb94cc023689592c4f07e4f3b1b9dd00\n",
            "Successfully built pyifcb pysmb rectpack\n",
            "Installing collected packages: rectpack, scipy, pysmb, pyifcb\n",
            "  Attempting uninstall: scipy\n",
            "    Found existing installation: scipy 1.10.1\n",
            "    Uninstalling scipy-1.10.1:\n",
            "      Successfully uninstalled scipy-1.10.1\n",
            "Successfully installed pyifcb-0.0.1 pysmb-1.2.9.1 rectpack-0.2.2 scipy-1.9.1\n"
          ]
        }
      ],
      "source": [
        "!pip install keras_preprocessing\n",
        "!git clone https://patcdaniel:zozmir-1qempa-kenrAb@huggingface.co/patcdaniel/phytoClassUCSC\n",
        "!pip install -U git+https://github.com/joefutrelle/pyifcb.git\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "import keras_preprocessing.image as keras_img\n",
        "import numpy as np\n",
        "import ifcb\n",
        "import json, os\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import urllib.request, urllib.parse\n",
        "from PIL import Image\n",
        "import pandas as pd"
      ],
      "metadata": {
        "id": "0NXIPkqB049o"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Load the Model (phytoClassUCSC.h5)"
      ],
      "metadata": {
        "id": "doj0y_g02dyI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model  = tf.keras.saving.load_model(\"./phytoClassUCSC/phytoClassUCSC.h5\")\n",
        "with open(\"./phytoClassUCSC/class_list.json\") as json_file:\n",
        "  class_list = list(json.load(json_file))"
      ],
      "metadata": {
        "id": "nfX2zKtB0-vP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Grab an entire syringe (.roi, .hdr, .adc) from the CalOOS Dashboard.\n",
        "\n",
        "Let's download some images off of the CalOOS IFCB Dashboard ([ifcb.caloos.org](ifcb.caloos.org))\n",
        "\n"
      ],
      "metadata": {
        "id": "5fZDcm3T2l_V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "base_url = \"https://ifcb.caloos.org\"\n",
        "data_set = \"santa-cruz-municipal-wharf\"\n",
        "syringe = \"D20230719T064404_IFCB104\"\n",
        "url = \"/\".join([base_url, data_set, syringe])\n",
        "\n",
        "for base in ['.roi', '.adc','.hdr']:\n",
        "  full_url = url + base\n",
        "  save_name = full_url.split(\"/\")[-1]\n",
        "  print(\"Retrieving {} from {}\".format(save_name, full_url))\n",
        "  urllib.request.urlretrieve(full_url, filename=os.path.join(\"/content\",save_name))"
      ],
      "metadata": {
        "id": "3v8h0UnwKGMa",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ce53a507-bf5c-4f5a-c02d-8ba924217515"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Retrieving D20230719T064404_IFCB104.roi from https://ifcb.caloos.org/santa-cruz-municipal-wharf/D20230719T064404_IFCB104.roi\n",
            "Retrieving D20230719T064404_IFCB104.adc from https://ifcb.caloos.org/santa-cruz-municipal-wharf/D20230719T064404_IFCB104.adc\n",
            "Retrieving D20230719T064404_IFCB104.hdr from https://ifcb.caloos.org/santa-cruz-municipal-wharf/D20230719T064404_IFCB104.hdr\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The images need to be reshaped into a certain size and format for the model, so the code below loads the image, resizes it, changes it to a three channel R,G,B and returns the image as an array."
      ],
      "metadata": {
        "id": "Pqwu31gzBFt2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def prep_image(img_data):\n",
        "\n",
        "    \"\"\"Load and prep images for model, reshape and normalize rgb to greyscale\"\"\"\n",
        "\n",
        "    target_size=(224,224)\n",
        "    img = keras_img.img_to_array(Image.fromarray(img_data).resize(target_size))\n",
        "    img /= 255\n",
        "    img = img.reshape((1, img.shape[0], img.shape[1], img.shape[2]))\n",
        "    return img\n",
        "\n",
        "\n",
        "def build_image_stack(roi_fname):\n",
        "\n",
        "  \"\"\" Return a matric of preprocessed images from a singe syringe\"\"\"\n",
        "\n",
        "  with ifcb.open_raw(os.path.join(\"/content\",roi_fname)) as roi_data:\n",
        "      array_index = 0\n",
        "      roi_names = []\n",
        "      img_stack = np.empty(shape=(len(roi_data.images),224,224,3))\n",
        "      for roi_num, img_data in roi_data.images.items():\n",
        "          img_stack[array_index,:,:,:] = prep_image(img_data)\n",
        "          array_index += 1\n",
        "          roi_names.append(roi_num)\n",
        "\n",
        "      # Also return run and inhibit times for sample volume calculation\n",
        "      run_time = roi_data.hdr_attributes['runTime']\n",
        "      inhibit_time = roi_data.hdr_attributes['inhibitTime']\n",
        "\n",
        "  return img_stack, roi_names, inhibit_time, run_time"
      ],
      "metadata": {
        "id": "7aU9WZCQLoyG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img_stack, roi_names, inhibit_time, run_time = build_image_stack(\"D20230719T064404_IFCB104.roi\")\n"
      ],
      "metadata": {
        "id": "3Bjtay0R4LWS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use the model to predict the class"
      ],
      "metadata": {
        "id": "j6s9ytvHAa7q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "yhat = model.predict(img_stack)"
      ],
      "metadata": {
        "id": "4ZMlMC-p5oqL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9adcacd2-d35c-43f1-9b44-5327f39571a0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "86/86 [==============================] - 23s 154ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now select the top class for each prediction"
      ],
      "metadata": {
        "id": "BgCICpRbTJEH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "top_ix = np.argmax(yhat,axis=1)\n",
        "top_prob = []\n",
        "top_class = []\n",
        "for i, ix in enumerate(top_ix):\n",
        "  top_prob.append(yhat[i,ix])\n",
        "  top_class.append(class_list[ix])"
      ],
      "metadata": {
        "id": "3KjrVwSxRDu1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_full = pd.DataFrame(data= {\"roi\":roi_names, \"top_class\":top_class,\"top_prob\":top_prob})\n",
        "df_full['img_url'] = [\"/\".join([base_url,\"data\", syringe + \"_{:04d}.png\".format(r)]) for r in roi_names ]\n",
        "flowrate = 0.25; # .25 mls per minute\n",
        "volume_analyzed = round(((run_time - inhibit_time) * flowrate)/60, 3)\n",
        "print(\"Sample Volume: {} mL\".format(volume_analyzed))\n",
        "df_full"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "id": "aHyEaEGTSgTX",
        "outputId": "3278b587-e3f4-4cd7-c61f-9756a93c5067"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Sample Volume: 4.108 mL\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       roi     top_class  top_prob  \\\n",
              "0        2      Ciliates  0.329502   \n",
              "1        3       Centric  0.717917   \n",
              "2        4       Centric  0.861939   \n",
              "3        5      Eucampia  0.983304   \n",
              "4        6      Eucampia  0.999946   \n",
              "...    ...           ...       ...   \n",
              "2723  2726  NanoP_less10  0.878550   \n",
              "2724  2727   Chaetoceros  0.994962   \n",
              "2725  2728      Eucampia  0.997788   \n",
              "2726  2729  Prorocentrum  0.999849   \n",
              "2727  2730      Eucampia  0.999993   \n",
              "\n",
              "                                                img_url  \n",
              "0     https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "1     https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "2     https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "3     https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "4     https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "...                                                 ...  \n",
              "2723  https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "2724  https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "2725  https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "2726  https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "2727  https://ifcb.caloos.org/data/D20230719T064404_...  \n",
              "\n",
              "[2728 rows x 4 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-005a6e1a-e87f-40fb-aa59-f6550ec8bbe1\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>roi</th>\n",
              "      <th>top_class</th>\n",
              "      <th>top_prob</th>\n",
              "      <th>img_url</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2</td>\n",
              "      <td>Ciliates</td>\n",
              "      <td>0.329502</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.717917</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.861939</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>5</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.983304</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.999946</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2723</th>\n",
              "      <td>2726</td>\n",
              "      <td>NanoP_less10</td>\n",
              "      <td>0.878550</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2724</th>\n",
              "      <td>2727</td>\n",
              "      <td>Chaetoceros</td>\n",
              "      <td>0.994962</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2725</th>\n",
              "      <td>2728</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.997788</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2726</th>\n",
              "      <td>2729</td>\n",
              "      <td>Prorocentrum</td>\n",
              "      <td>0.999849</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2727</th>\n",
              "      <td>2730</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.999993</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2728 rows × 4 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-005a6e1a-e87f-40fb-aa59-f6550ec8bbe1')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-2c90a5ee-ffaf-4c7d-8464-7a998c7dfa71\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-2c90a5ee-ffaf-4c7d-8464-7a998c7dfa71')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-2c90a5ee-ffaf-4c7d-8464-7a998c7dfa71 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-005a6e1a-e87f-40fb-aa59-f6550ec8bbe1 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-005a6e1a-e87f-40fb-aa59-f6550ec8bbe1');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "*Now* lets convert the data output into cell concentrations.\n",
        "- Calculate the sample volume\n",
        "- \\# per mL"
      ],
      "metadata": {
        "id": "Lj1gNXIdLpbv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def load_thresholds(fname):\n",
        "    \"\"\" Load pre-determined class-specific thresholds  \"\"\"\n",
        "    with open(fname, 'r') as file:\n",
        "        thresh = json.load(file)\n",
        "        thresh_vals = np.array([thresh[k] for k in thresh.keys()])\n",
        "    return thresh_vals\n",
        "\n",
        "threshold_vals = load_thresholds(\"/content/phytoClassUCSC/class_threshold_v1.0.json\")"
      ],
      "metadata": {
        "id": "-l9WN8PLZzx8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Map the threshold values for each class to the dataframe\n",
        "thresh = pd.DataFrame(data={\"class_val\":threshold_vals})\n",
        "thresh.index = class_list[:-1]\n",
        "df_full['class_threshold'] = df_full['top_class'].map(thresh['class_val'])\n",
        "\n",
        "# Prefill the series with value \"Unclassified\" and then replace the rows where the classifier is greater than the\n",
        "df_full[\"top_class_thresh\"] = \"Unclassified\"\n",
        "greater_than_thresh = df_full['top_prob'] > df_full[\"class_threshold\"] # Boolean series, True where greater than threshold\n",
        "df_full[\"top_class_thresh\"][greater_than_thresh] = df_full[\"top_class\"][greater_than_thresh]; # replace \"Unclassifed\" values with actual class, where True\n",
        "df_full['thresh_diff'] = df_full['top_prob'] - df_full[\"class_threshold\"]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BIuj5gS6gano",
        "outputId": "7625df86-c470-4363-917d-4ecfcbd32120"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-90-922700357ba9>:9: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_full[\"top_class_thresh\"][greater_than_thresh] = df_full[\"top_class\"][greater_than_thresh]; # replace \"Unclassifed\" values with actual class, where True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_full"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 687
        },
        "id": "Pn_ytX-0h8iZ",
        "outputId": "0ebedd65-d0c0-4213-8e34-e28cc514a260"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       roi     top_class  top_prob  \\\n",
              "0        2      Ciliates  0.329502   \n",
              "1        3       Centric  0.717917   \n",
              "2        4       Centric  0.861939   \n",
              "3        5      Eucampia  0.983304   \n",
              "4        6      Eucampia  0.999946   \n",
              "...    ...           ...       ...   \n",
              "2723  2726  NanoP_less10  0.878550   \n",
              "2724  2727   Chaetoceros  0.994962   \n",
              "2725  2728      Eucampia  0.997788   \n",
              "2726  2729  Prorocentrum  0.999849   \n",
              "2727  2730      Eucampia  0.999993   \n",
              "\n",
              "                                                img_url top_class_thresh  \\\n",
              "0     https://ifcb.caloos.org/data/D20230719T064404_...     Unclassified   \n",
              "1     https://ifcb.caloos.org/data/D20230719T064404_...          Centric   \n",
              "2     https://ifcb.caloos.org/data/D20230719T064404_...          Centric   \n",
              "3     https://ifcb.caloos.org/data/D20230719T064404_...         Eucampia   \n",
              "4     https://ifcb.caloos.org/data/D20230719T064404_...         Eucampia   \n",
              "...                                                 ...              ...   \n",
              "2723  https://ifcb.caloos.org/data/D20230719T064404_...     Unclassified   \n",
              "2724  https://ifcb.caloos.org/data/D20230719T064404_...      Chaetoceros   \n",
              "2725  https://ifcb.caloos.org/data/D20230719T064404_...         Eucampia   \n",
              "2726  https://ifcb.caloos.org/data/D20230719T064404_...     Prorocentrum   \n",
              "2727  https://ifcb.caloos.org/data/D20230719T064404_...         Eucampia   \n",
              "\n",
              "      class_threshold  thresh_diff  \n",
              "0                0.49    -0.160498  \n",
              "1                0.70     0.017917  \n",
              "2                0.70     0.161939  \n",
              "3                0.88     0.103304  \n",
              "4                0.88     0.119946  \n",
              "...               ...          ...  \n",
              "2723             0.92    -0.041450  \n",
              "2724             0.89     0.104962  \n",
              "2725             0.88     0.117788  \n",
              "2726             0.92     0.079849  \n",
              "2727             0.88     0.119993  \n",
              "\n",
              "[2728 rows x 7 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-54a68fe6-eef6-4e53-a4ab-2a388a58ec3b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>roi</th>\n",
              "      <th>top_class</th>\n",
              "      <th>top_prob</th>\n",
              "      <th>img_url</th>\n",
              "      <th>top_class_thresh</th>\n",
              "      <th>class_threshold</th>\n",
              "      <th>thresh_diff</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2</td>\n",
              "      <td>Ciliates</td>\n",
              "      <td>0.329502</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Unclassified</td>\n",
              "      <td>0.49</td>\n",
              "      <td>-0.160498</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.717917</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.70</td>\n",
              "      <td>0.017917</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.861939</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Centric</td>\n",
              "      <td>0.70</td>\n",
              "      <td>0.161939</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>5</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.983304</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.88</td>\n",
              "      <td>0.103304</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.999946</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.88</td>\n",
              "      <td>0.119946</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2723</th>\n",
              "      <td>2726</td>\n",
              "      <td>NanoP_less10</td>\n",
              "      <td>0.878550</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Unclassified</td>\n",
              "      <td>0.92</td>\n",
              "      <td>-0.041450</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2724</th>\n",
              "      <td>2727</td>\n",
              "      <td>Chaetoceros</td>\n",
              "      <td>0.994962</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Chaetoceros</td>\n",
              "      <td>0.89</td>\n",
              "      <td>0.104962</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2725</th>\n",
              "      <td>2728</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.997788</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.88</td>\n",
              "      <td>0.117788</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2726</th>\n",
              "      <td>2729</td>\n",
              "      <td>Prorocentrum</td>\n",
              "      <td>0.999849</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Prorocentrum</td>\n",
              "      <td>0.92</td>\n",
              "      <td>0.079849</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2727</th>\n",
              "      <td>2730</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.999993</td>\n",
              "      <td>https://ifcb.caloos.org/data/D20230719T064404_...</td>\n",
              "      <td>Eucampia</td>\n",
              "      <td>0.88</td>\n",
              "      <td>0.119993</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2728 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-54a68fe6-eef6-4e53-a4ab-2a388a58ec3b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-4e1ef764-e1e2-4502-ad97-dabb4ebae8e7\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4e1ef764-e1e2-4502-ad97-dabb4ebae8e7')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-4e1ef764-e1e2-4502-ad97-dabb4ebae8e7 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-54a68fe6-eef6-4e53-a4ab-2a388a58ec3b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-54a68fe6-eef6-4e53-a4ab-2a388a58ec3b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Count number of images for each class ##"
      ],
      "metadata": {
        "id": "wfIWHO46mHZW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "total = df_full.groupby('top_class_thresh')['roi'].count()\n",
        "total = total.sort_values(ascending=False)\n",
        "labels = total.index\n",
        "total_val = total.values"
      ],
      "metadata": {
        "id": "NwPpJK_4mNzR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots()\n",
        "fig.set_size_inches(12,6)\n",
        "sns.barplot(x=labels,y=total_val, palette=\"Blues_d\")\n",
        "ax.tick_params(axis='x', rotation=90)\n",
        "ax.set_yscale('log')\n",
        "ax.set_ylabel(\"# per syringe\")\n",
        "for i, count in enumerate(total_val):\n",
        "    div_by = 1/len(total_val)\n",
        "    ax.text(div_by*i + .004,.03,str(int(count)),rotation='vertical', transform=ax.transAxes,c='w',weight='bold')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 669
        },
        "id": "OhXYZ3JpmN5G",
        "outputId": "76b20351-732b-42f4-ffb1-2d88478f0c92"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "HpjV5SgGo_wW"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}