train_random_forest.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### This notebook is used to train a scikit-learn random forest regression model to predict diatom carbon biomass from environmental and optical parameters. The trained model is saved using pickle, so that it can be loaded and deployed later in deploy_trained_model.ipynb notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import HistGradientBoostingRegressor\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# library for save and load scikit-learn models\n",
    "import pickle\n",
    "\n",
    "# some example code from:\n",
    "# https://towardsdatascience.com/random-forest-in-python-24d0893d51c0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>enviro_grid_index</th>\n",
       "      <th>cruise</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>t_std</th>\n",
       "      <th>s_avg</th>\n",
       "      <th>s_std</th>\n",
       "      <th>chla_avg</th>\n",
       "      <th>chla_std</th>\n",
       "      <th>chlb_avg</th>\n",
       "      <th>chlb_std</th>\n",
       "      <th>...</th>\n",
       "      <th>chlb_log</th>\n",
       "      <th>chlc_log</th>\n",
       "      <th>ppc_log</th>\n",
       "      <th>psc_log</th>\n",
       "      <th>chlbb_log</th>\n",
       "      <th>bbp532_log</th>\n",
       "      <th>chlb2c_log</th>\n",
       "      <th>chlbb</th>\n",
       "      <th>chlb2c</th>\n",
       "      <th>diatCarb_log10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>174</td>\n",
       "      <td>1</td>\n",
       "      <td>15.379565</td>\n",
       "      <td>0.015037</td>\n",
       "      <td>33.361638</td>\n",
       "      <td>0.009658</td>\n",
       "      <td>3.036134</td>\n",
       "      <td>0.073602</td>\n",
       "      <td>0.382791</td>\n",
       "      <td>0.016337</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.417038</td>\n",
       "      <td>-0.372212</td>\n",
       "      <td>-0.254414</td>\n",
       "      <td>0.169451</td>\n",
       "      <td>2.890307</td>\n",
       "      <td>-2.429431</td>\n",
       "      <td>-0.044826</td>\n",
       "      <td>776.796044</td>\n",
       "      <td>0.901933</td>\n",
       "      <td>0.135997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>181</td>\n",
       "      <td>1</td>\n",
       "      <td>15.227767</td>\n",
       "      <td>0.035458</td>\n",
       "      <td>33.185741</td>\n",
       "      <td>0.014888</td>\n",
       "      <td>4.716638</td>\n",
       "      <td>0.037522</td>\n",
       "      <td>0.600443</td>\n",
       "      <td>0.008299</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.221528</td>\n",
       "      <td>-0.136159</td>\n",
       "      <td>-0.062828</td>\n",
       "      <td>0.348987</td>\n",
       "      <td>3.014347</td>\n",
       "      <td>-2.359437</td>\n",
       "      <td>-0.085369</td>\n",
       "      <td>1033.587328</td>\n",
       "      <td>0.821543</td>\n",
       "      <td>1.073820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196</td>\n",
       "      <td>1</td>\n",
       "      <td>15.040281</td>\n",
       "      <td>0.007821</td>\n",
       "      <td>33.254836</td>\n",
       "      <td>0.000617</td>\n",
       "      <td>2.800934</td>\n",
       "      <td>0.046619</td>\n",
       "      <td>0.319163</td>\n",
       "      <td>0.001315</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.495988</td>\n",
       "      <td>-0.426783</td>\n",
       "      <td>-0.236449</td>\n",
       "      <td>0.150757</td>\n",
       "      <td>2.770050</td>\n",
       "      <td>-2.346608</td>\n",
       "      <td>-0.069205</td>\n",
       "      <td>588.910931</td>\n",
       "      <td>0.852698</td>\n",
       "      <td>0.680539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>203</td>\n",
       "      <td>1</td>\n",
       "      <td>14.741559</td>\n",
       "      <td>0.029090</td>\n",
       "      <td>32.983054</td>\n",
       "      <td>0.061902</td>\n",
       "      <td>2.165557</td>\n",
       "      <td>0.095339</td>\n",
       "      <td>0.277416</td>\n",
       "      <td>0.022504</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.556868</td>\n",
       "      <td>-0.500465</td>\n",
       "      <td>-0.397578</td>\n",
       "      <td>0.029491</td>\n",
       "      <td>2.742604</td>\n",
       "      <td>-2.441786</td>\n",
       "      <td>-0.056403</td>\n",
       "      <td>552.845289</td>\n",
       "      <td>0.878208</td>\n",
       "      <td>0.441662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>211</td>\n",
       "      <td>1</td>\n",
       "      <td>14.567237</td>\n",
       "      <td>0.017287</td>\n",
       "      <td>32.982505</td>\n",
       "      <td>0.035529</td>\n",
       "      <td>2.637895</td>\n",
       "      <td>0.040919</td>\n",
       "      <td>0.308879</td>\n",
       "      <td>0.006489</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.510212</td>\n",
       "      <td>-0.412711</td>\n",
       "      <td>-0.290288</td>\n",
       "      <td>0.113209</td>\n",
       "      <td>2.773040</td>\n",
       "      <td>-2.378860</td>\n",
       "      <td>-0.097501</td>\n",
       "      <td>592.979368</td>\n",
       "      <td>0.798912</td>\n",
       "      <td>0.425136</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   enviro_grid_index  cruise      t_avg     t_std      s_avg     s_std  \\\n",
       "0                174       1  15.379565  0.015037  33.361638  0.009658   \n",
       "1                181       1  15.227767  0.035458  33.185741  0.014888   \n",
       "2                196       1  15.040281  0.007821  33.254836  0.000617   \n",
       "3                203       1  14.741559  0.029090  32.983054  0.061902   \n",
       "4                211       1  14.567237  0.017287  32.982505  0.035529   \n",
       "\n",
       "   chla_avg  chla_std  chlb_avg  chlb_std  ...  chlb_log  chlc_log   ppc_log  \\\n",
       "0  3.036134  0.073602  0.382791  0.016337  ... -0.417038 -0.372212 -0.254414   \n",
       "1  4.716638  0.037522  0.600443  0.008299  ... -0.221528 -0.136159 -0.062828   \n",
       "2  2.800934  0.046619  0.319163  0.001315  ... -0.495988 -0.426783 -0.236449   \n",
       "3  2.165557  0.095339  0.277416  0.022504  ... -0.556868 -0.500465 -0.397578   \n",
       "4  2.637895  0.040919  0.308879  0.006489  ... -0.510212 -0.412711 -0.290288   \n",
       "\n",
       "    psc_log  chlbb_log  bbp532_log  chlb2c_log        chlbb    chlb2c  \\\n",
       "0  0.169451   2.890307   -2.429431   -0.044826   776.796044  0.901933   \n",
       "1  0.348987   3.014347   -2.359437   -0.085369  1033.587328  0.821543   \n",
       "2  0.150757   2.770050   -2.346608   -0.069205   588.910931  0.852698   \n",
       "3  0.029491   2.742604   -2.441786   -0.056403   552.845289  0.878208   \n",
       "4  0.113209   2.773040   -2.378860   -0.097501   592.979368  0.798912   \n",
       "\n",
       "   diatCarb_log10  \n",
       "0        0.135997  \n",
       "1        1.073820  \n",
       "2        0.680539  \n",
       "3        0.441662  \n",
       "4        0.425136  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read in data as pandas dataframe and display first 5 rows\n",
    "data = pd.read_csv('../model_training_data/ifcb_enviro_26Jul2023.csv')\n",
    "data.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>enviro_grid_index</th>\n",
       "      <th>cruise</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>t_std</th>\n",
       "      <th>s_avg</th>\n",
       "      <th>s_std</th>\n",
       "      <th>chla_avg</th>\n",
       "      <th>chla_std</th>\n",
       "      <th>chlb_avg</th>\n",
       "      <th>chlb_std</th>\n",
       "      <th>...</th>\n",
       "      <th>chlb_log</th>\n",
       "      <th>chlc_log</th>\n",
       "      <th>ppc_log</th>\n",
       "      <th>psc_log</th>\n",
       "      <th>chlbb_log</th>\n",
       "      <th>bbp532_log</th>\n",
       "      <th>chlb2c_log</th>\n",
       "      <th>chlbb</th>\n",
       "      <th>chlb2c</th>\n",
       "      <th>diatCarb_log10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>174</td>\n",
       "      <td>1</td>\n",
       "      <td>15.379565</td>\n",
       "      <td>0.015037</td>\n",
       "      <td>33.361638</td>\n",
       "      <td>0.009658</td>\n",
       "      <td>3.036134</td>\n",
       "      <td>0.073602</td>\n",
       "      <td>0.382791</td>\n",
       "      <td>0.016337</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.417038</td>\n",
       "      <td>-0.372212</td>\n",
       "      <td>-0.254414</td>\n",
       "      <td>0.169451</td>\n",
       "      <td>2.890307</td>\n",
       "      <td>-2.429431</td>\n",
       "      <td>-0.044826</td>\n",
       "      <td>776.796044</td>\n",
       "      <td>0.901933</td>\n",
       "      <td>0.135997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>181</td>\n",
       "      <td>1</td>\n",
       "      <td>15.227767</td>\n",
       "      <td>0.035458</td>\n",
       "      <td>33.185741</td>\n",
       "      <td>0.014888</td>\n",
       "      <td>4.716638</td>\n",
       "      <td>0.037522</td>\n",
       "      <td>0.600443</td>\n",
       "      <td>0.008299</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.221528</td>\n",
       "      <td>-0.136159</td>\n",
       "      <td>-0.062828</td>\n",
       "      <td>0.348987</td>\n",
       "      <td>3.014347</td>\n",
       "      <td>-2.359437</td>\n",
       "      <td>-0.085369</td>\n",
       "      <td>1033.587328</td>\n",
       "      <td>0.821543</td>\n",
       "      <td>1.073820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196</td>\n",
       "      <td>1</td>\n",
       "      <td>15.040281</td>\n",
       "      <td>0.007821</td>\n",
       "      <td>33.254836</td>\n",
       "      <td>0.000617</td>\n",
       "      <td>2.800934</td>\n",
       "      <td>0.046619</td>\n",
       "      <td>0.319163</td>\n",
       "      <td>0.001315</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.495988</td>\n",
       "      <td>-0.426783</td>\n",
       "      <td>-0.236449</td>\n",
       "      <td>0.150757</td>\n",
       "      <td>2.770050</td>\n",
       "      <td>-2.346608</td>\n",
       "      <td>-0.069205</td>\n",
       "      <td>588.910931</td>\n",
       "      <td>0.852698</td>\n",
       "      <td>0.680539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>203</td>\n",
       "      <td>1</td>\n",
       "      <td>14.741559</td>\n",
       "      <td>0.029090</td>\n",
       "      <td>32.983054</td>\n",
       "      <td>0.061902</td>\n",
       "      <td>2.165557</td>\n",
       "      <td>0.095339</td>\n",
       "      <td>0.277416</td>\n",
       "      <td>0.022504</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.556868</td>\n",
       "      <td>-0.500465</td>\n",
       "      <td>-0.397578</td>\n",
       "      <td>0.029491</td>\n",
       "      <td>2.742604</td>\n",
       "      <td>-2.441786</td>\n",
       "      <td>-0.056403</td>\n",
       "      <td>552.845289</td>\n",
       "      <td>0.878208</td>\n",
       "      <td>0.441662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>211</td>\n",
       "      <td>1</td>\n",
       "      <td>14.567237</td>\n",
       "      <td>0.017287</td>\n",
       "      <td>32.982505</td>\n",
       "      <td>0.035529</td>\n",
       "      <td>2.637895</td>\n",
       "      <td>0.040919</td>\n",
       "      <td>0.308879</td>\n",
       "      <td>0.006489</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.510212</td>\n",
       "      <td>-0.412711</td>\n",
       "      <td>-0.290288</td>\n",
       "      <td>0.113209</td>\n",
       "      <td>2.773040</td>\n",
       "      <td>-2.378860</td>\n",
       "      <td>-0.097501</td>\n",
       "      <td>592.979368</td>\n",
       "      <td>0.798912</td>\n",
       "      <td>0.425136</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>6550</th>\n",
       "      <td>101294</td>\n",
       "      <td>8</td>\n",
       "      <td>14.007938</td>\n",
       "      <td>0.014773</td>\n",
       "      <td>34.859329</td>\n",
       "      <td>0.001180</td>\n",
       "      <td>2.613734</td>\n",
       "      <td>0.022299</td>\n",
       "      <td>0.246030</td>\n",
       "      <td>0.002445</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.609011</td>\n",
       "      <td>-0.455332</td>\n",
       "      <td>-0.305308</td>\n",
       "      <td>-0.034151</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.153679</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.701973</td>\n",
       "      <td>0.022158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6551</th>\n",
       "      <td>101299</td>\n",
       "      <td>8</td>\n",
       "      <td>14.109504</td>\n",
       "      <td>0.049641</td>\n",
       "      <td>34.855933</td>\n",
       "      <td>0.002464</td>\n",
       "      <td>6.078749</td>\n",
       "      <td>1.465459</td>\n",
       "      <td>0.776812</td>\n",
       "      <td>0.225881</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.109684</td>\n",
       "      <td>-0.042954</td>\n",
       "      <td>0.064734</td>\n",
       "      <td>0.443202</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.066730</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.857571</td>\n",
       "      <td>0.113777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6552</th>\n",
       "      <td>101305</td>\n",
       "      <td>8</td>\n",
       "      <td>13.797833</td>\n",
       "      <td>0.082510</td>\n",
       "      <td>34.820696</td>\n",
       "      <td>0.003199</td>\n",
       "      <td>25.287404</td>\n",
       "      <td>1.178184</td>\n",
       "      <td>4.354826</td>\n",
       "      <td>0.272359</td>\n",
       "      <td>...</td>\n",
       "      <td>0.638971</td>\n",
       "      <td>0.635266</td>\n",
       "      <td>0.666525</td>\n",
       "      <td>1.134371</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.003705</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.008568</td>\n",
       "      <td>0.101796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6553</th>\n",
       "      <td>101316</td>\n",
       "      <td>8</td>\n",
       "      <td>13.826412</td>\n",
       "      <td>0.068178</td>\n",
       "      <td>34.799279</td>\n",
       "      <td>0.000766</td>\n",
       "      <td>19.998346</td>\n",
       "      <td>0.333792</td>\n",
       "      <td>2.864524</td>\n",
       "      <td>0.081697</td>\n",
       "      <td>...</td>\n",
       "      <td>0.457052</td>\n",
       "      <td>0.529280</td>\n",
       "      <td>0.551496</td>\n",
       "      <td>0.983307</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.072228</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.846783</td>\n",
       "      <td>1.654914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6554</th>\n",
       "      <td>101320</td>\n",
       "      <td>8</td>\n",
       "      <td>14.149879</td>\n",
       "      <td>0.071456</td>\n",
       "      <td>34.795725</td>\n",
       "      <td>0.000669</td>\n",
       "      <td>28.980305</td>\n",
       "      <td>0.916038</td>\n",
       "      <td>4.874042</td>\n",
       "      <td>0.212312</td>\n",
       "      <td>...</td>\n",
       "      <td>0.687889</td>\n",
       "      <td>0.700119</td>\n",
       "      <td>0.752727</td>\n",
       "      <td>1.194858</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.012229</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.972234</td>\n",
       "      <td>1.471701</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6555 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      enviro_grid_index  cruise      t_avg     t_std      s_avg     s_std  \\\n",
       "0                   174       1  15.379565  0.015037  33.361638  0.009658   \n",
       "1                   181       1  15.227767  0.035458  33.185741  0.014888   \n",
       "2                   196       1  15.040281  0.007821  33.254836  0.000617   \n",
       "3                   203       1  14.741559  0.029090  32.983054  0.061902   \n",
       "4                   211       1  14.567237  0.017287  32.982505  0.035529   \n",
       "...                 ...     ...        ...       ...        ...       ...   \n",
       "6550             101294       8  14.007938  0.014773  34.859329  0.001180   \n",
       "6551             101299       8  14.109504  0.049641  34.855933  0.002464   \n",
       "6552             101305       8  13.797833  0.082510  34.820696  0.003199   \n",
       "6553             101316       8  13.826412  0.068178  34.799279  0.000766   \n",
       "6554             101320       8  14.149879  0.071456  34.795725  0.000669   \n",
       "\n",
       "       chla_avg  chla_std  chlb_avg  chlb_std  ...  chlb_log  chlc_log  \\\n",
       "0      3.036134  0.073602  0.382791  0.016337  ... -0.417038 -0.372212   \n",
       "1      4.716638  0.037522  0.600443  0.008299  ... -0.221528 -0.136159   \n",
       "2      2.800934  0.046619  0.319163  0.001315  ... -0.495988 -0.426783   \n",
       "3      2.165557  0.095339  0.277416  0.022504  ... -0.556868 -0.500465   \n",
       "4      2.637895  0.040919  0.308879  0.006489  ... -0.510212 -0.412711   \n",
       "...         ...       ...       ...       ...  ...       ...       ...   \n",
       "6550   2.613734  0.022299  0.246030  0.002445  ... -0.609011 -0.455332   \n",
       "6551   6.078749  1.465459  0.776812  0.225881  ... -0.109684 -0.042954   \n",
       "6552  25.287404  1.178184  4.354826  0.272359  ...  0.638971  0.635266   \n",
       "6553  19.998346  0.333792  2.864524  0.081697  ...  0.457052  0.529280   \n",
       "6554  28.980305  0.916038  4.874042  0.212312  ...  0.687889  0.700119   \n",
       "\n",
       "       ppc_log   psc_log  chlbb_log  bbp532_log  chlb2c_log        chlbb  \\\n",
       "0    -0.254414  0.169451   2.890307   -2.429431   -0.044826   776.796044   \n",
       "1    -0.062828  0.348987   3.014347   -2.359437   -0.085369  1033.587328   \n",
       "2    -0.236449  0.150757   2.770050   -2.346608   -0.069205   588.910931   \n",
       "3    -0.397578  0.029491   2.742604   -2.441786   -0.056403   552.845289   \n",
       "4    -0.290288  0.113209   2.773040   -2.378860   -0.097501   592.979368   \n",
       "...        ...       ...        ...         ...         ...          ...   \n",
       "6550 -0.305308 -0.034151        NaN         NaN   -0.153679          NaN   \n",
       "6551  0.064734  0.443202        NaN         NaN   -0.066730          NaN   \n",
       "6552  0.666525  1.134371        NaN         NaN    0.003705          NaN   \n",
       "6553  0.551496  0.983307        NaN         NaN   -0.072228          NaN   \n",
       "6554  0.752727  1.194858        NaN         NaN   -0.012229          NaN   \n",
       "\n",
       "        chlb2c  diatCarb_log10  \n",
       "0     0.901933        0.135997  \n",
       "1     0.821543        1.073820  \n",
       "2     0.852698        0.680539  \n",
       "3     0.878208        0.441662  \n",
       "4     0.798912        0.425136  \n",
       "...        ...             ...  \n",
       "6550  0.701973        0.022158  \n",
       "6551  0.857571        0.113777  \n",
       "6552  1.008568        0.101796  \n",
       "6553  0.846783        1.654914  \n",
       "6554  0.972234        1.471701  \n",
       "\n",
       "[6555 rows x 41 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = data[['t_avg','chla_avg','diatCarb']] #, 'bbp532_avg','lat','s_avg','chlb_avg','chlc_avg','ppc_avg'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/g_/tv6zcf195p1dfbn61x_n_mq80000gq/T/ipykernel_25429/3761598809.py:4: 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.replace([np.inf, -np.inf], np.nan, inplace=True)\n",
      "/var/folders/g_/tv6zcf195p1dfbn61x_n_mq80000gq/T/ipykernel_25429/3761598809.py:7: 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.dropna(inplace=True)\n"
     ]
    }
   ],
   "source": [
    "# Remove rows with any NaN or Inf values\n",
    "\n",
    "# Rename inf or -inf as NaN\n",
    "df.replace([np.inf, -np.inf], np.nan, inplace=True)\n",
    "\n",
    "# Drop rows with NaN\n",
    "df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>t_avg</th>\n",
       "      <th>chla_avg</th>\n",
       "      <th>diatCarb</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6510.000000</td>\n",
       "      <td>6510.000000</td>\n",
       "      <td>6510.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>12.312465</td>\n",
       "      <td>0.719951</td>\n",
       "      <td>4.065167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5.900007</td>\n",
       "      <td>1.061565</td>\n",
       "      <td>10.403997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-0.522733</td>\n",
       "      <td>0.000277</td>\n",
       "      <td>0.000488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>8.571136</td>\n",
       "      <td>0.246198</td>\n",
       "      <td>0.319799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.952555</td>\n",
       "      <td>0.434503</td>\n",
       "      <td>0.893130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16.389986</td>\n",
       "      <td>0.786179</td>\n",
       "      <td>3.338545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>27.961023</td>\n",
       "      <td>28.980305</td>\n",
       "      <td>236.312534</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             t_avg     chla_avg     diatCarb\n",
       "count  6510.000000  6510.000000  6510.000000\n",
       "mean     12.312465     0.719951     4.065167\n",
       "std       5.900007     1.061565    10.403997\n",
       "min      -0.522733     0.000277     0.000488\n",
       "25%       8.571136     0.246198     0.319799\n",
       "50%      12.952555     0.434503     0.893130\n",
       "75%      16.389986     0.786179     3.338545\n",
       "max      27.961023    28.980305   236.312534"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Descriptive statistics for each column\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df['lat'] = df['lat'].abs()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Diatom carbon are the target of the algorithm (i.e., the values we want to predict)\n",
    "targets = np.array(df['diatCarb'])\n",
    "\n",
    "# Remove the labels from the features\n",
    "# axis 1 refers to the columns\n",
    "df = df.drop('diatCarb', axis = 1)\n",
    "\n",
    "# Saving feature names for later use\n",
    "feature_list = list(df.columns)\n",
    "\n",
    "# Convert to numpy array\n",
    "inputs = np.array(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using Skicit-learn to split data into training and testing sets\n",
    "\n",
    "# Split the data into training and testing sets\n",
    "x_train, x_test, y_train, y_test = train_test_split(df, targets, test_size = 0.25, random_state = 42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Features Shape: (4882, 2)\n",
      "Training Labels Shape: (4882,)\n",
      "Testing Features Shape: (1628, 2)\n",
      "Testing Labels Shape: (1628,)\n"
     ]
    }
   ],
   "source": [
    "print('Training Features Shape:', x_train.shape)\n",
    "print('Training Labels Shape:', y_train.shape)\n",
    "print('Testing Features Shape:', x_test.shape)\n",
    "print('Testing Labels Shape:', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The baseline predictions derived from a linear (log space) relationship between Chl and diatoms\n",
    "baseline_preds = 1.5*x_test['chla_avg']**1.9\n",
    "\n",
    "# # Baseline errors, and display average baseline error\n",
    "# baseline_errors = abs(baseline_preds - test_labels)\n",
    "# print('Average baseline error: ', round(np.mean(baseline_errors), 2))\n",
    "# Average baseline error:  5.06 degrees."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Baseline errors, and display average baseline error\n",
    "baseline_errors = (abs(baseline_preds - y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average baseline error:  3.55  mg m^-3\n",
      "Median baseline error:  0.72  mg m^-3\n"
     ]
    }
   ],
   "source": [
    "print('Average baseline error: ', round(np.mean(baseline_errors), 2),' mg m^-3')\n",
    "print('Median baseline error: ', round(np.median(baseline_errors), 2),' mg m^-3')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# standardize the data\n",
    "# scaler = StandardScaler().fit(x_train)\n",
    "# X_train_scaled = pd.DataFrame(scaler.transform(x_train), index=x_train.index.values, columns=x_train.columns.values)\n",
    "# X_test_scaled = pd.DataFrame(scaler.transform(x_test), index=x_test.index.values, columns=x_test.columns.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate model with 1000 decision trees\n",
    "rf = RandomForestRegressor(n_estimators = 1000, oob_score=True, random_state = 42)\n",
    "\n",
    "#rf = HistGradientBoostingRegressor(random_state = 42)\n",
    "\n",
    "# Train the model on training data\n",
    "rf.fit(x_train, y_train);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Out-of-bag R-2 score estimate: 0.444\n",
      "Test data R-2 score: 0.458\n",
      "Test data Spearman correlation: 0.729\n",
      "Test data Pearson correlation: 0.685\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "from scipy.stats import spearmanr, pearsonr\n",
    "\n",
    "predicted_train = rf.predict(x_train)\n",
    "predicted_test = rf.predict(x_test)\n",
    "\n",
    "test_score = r2_score(y_test, predicted_test)\n",
    "spearman = spearmanr(y_test, predicted_test)\n",
    "pearson = pearsonr(y_test, predicted_test)\n",
    "\n",
    "print(f'Out-of-bag R-2 score estimate: {rf.oob_score_:>5.3}')\n",
    "print(f'Test data R-2 score: {test_score:>5.3}')\n",
    "print(f'Test data Spearman correlation: {spearman[0]:.3}')\n",
    "print(f'Test data Pearson correlation: {pearson[0]:.3}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared: 0.926\n"
     ]
    }
   ],
   "source": [
    "score = rf.score(x_train, y_train)\n",
    "print(\"R-squared:\", round(score, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Absolute Error: 2.69 mg m^-3\n",
      "Median Absolute Error: 0.63 mg m^-3\n"
     ]
    }
   ],
   "source": [
    "# Use the forest's predict method on the test data\n",
    "predictions = rf.predict(x_test)\n",
    "\n",
    "# Calculate the absolute errors\n",
    "errors = abs(y_test - predictions)\n",
    "\n",
    "# Print out the mean and median absolute errors \n",
    "print('Mean Absolute Error:', round(np.mean(errors), 2), 'mg m^-3')\n",
    "print('Median Absolute Error:', round(np.median(errors), 2), 'mg m^-3')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 99.66 %\n"
     ]
    }
   ],
   "source": [
    "# Calculate mean absolute percentage error (MAPE)\n",
    "mape = 100 * (errors / abs(y_test)) * (1/1541)\n",
    "\n",
    "# Calculate and display accuracy\n",
    "accuracy = 100 - np.mean(mape)\n",
    "print('Accuracy:', round(accuracy, 2), '%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "ax = plt.gca()\n",
    "ax.scatter(y_test,predictions, c='blue', alpha=0.5, edgecolors='none')\n",
    "ax.set_yscale('log')\n",
    "ax.set_xscale('log')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variable: t_avg                Importance: 0.58\n",
      "Variable: chla_avg             Importance: 0.42\n"
     ]
    }
   ],
   "source": [
    "# Get numerical feature importances\n",
    "importances = list(rf.feature_importances_)\n",
    "\n",
    "# List of tuples with variable and importance\n",
    "feature_importances = [(feature, round(importance, 2)) for feature, importance in zip(feature_list, importances)]\n",
    "\n",
    "# Sort the feature importances by most important first\n",
    "feature_importances = sorted(feature_importances, key = lambda x: x[1], reverse = True)\n",
    "\n",
    "# Print out the feature and importances \n",
    "[print('Variable: {:20} Importance: {}'.format(*pair)) for pair in feature_importances];\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save model\n",
    "\n",
    "# file name, can use *.pickle as a file extension\n",
    "filename = \"trained_models/random_forest.pickle_chl_sst\"\n",
    "\n",
    "# save model\n",
    "pickle.dump(rf, open(filename, \"wb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### model can now be deployed in deploy_trained_model.ipynb"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "diatom algorithm",
   "language": "python",
   "name": "diatoms"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}