{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0904d215",
   "metadata": {},
   "source": [
    "# Notebook 05: Simple ML Regression\n",
    "\n",
    "### Goal: Basic training a ML using a single feature/predictor/input and a single ML model\n",
    "\n",
    "#### Reminder of Problem Statement\n",
    "\n",
    "Before we jump into the ML, I want to remind you of the ML task we want to accomplish in the paper. \n",
    "\n",
    "1. Does this image contain a thunderstorm? <-- Classification\n",
    "2. How many lightning flashes are in this image? <-- Regression\n",
    "\n",
    "#### Background\n",
    "\n",
    "For the training of regression problems is basically the same as the classification. We will still use the same steps as the previous notebook, just with a small changed to the labels ```y```. \n",
    "\n",
    "#### Step 1 & 2: Import packages and load data for Regression  \n",
    "We only want 1 feature again to make things simple, which is feature 0. We also will need to change ```class_labels``` to false."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "831307b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#needed packages \n",
    "import xarray as xr\n",
    "import matplotlib.pyplot as plt \n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "#plot parameters that I personally like, feel free to make these your own.\n",
    "import matplotlib\n",
    "matplotlib.rcParams['axes.facecolor'] = [0.9,0.9,0.9] #makes a grey background to the axis face\n",
    "matplotlib.rcParams['axes.labelsize'] = 14 #fontsize in pts\n",
    "matplotlib.rcParams['axes.titlesize'] = 14 \n",
    "matplotlib.rcParams['xtick.labelsize'] = 12 \n",
    "matplotlib.rcParams['ytick.labelsize'] = 12 \n",
    "matplotlib.rcParams['legend.fontsize'] = 12 \n",
    "matplotlib.rcParams['legend.facecolor'] = 'w' \n",
    "matplotlib.rcParams['savefig.transparent'] = False\n",
    "\n",
    "#make default resolution of figures much higher (i.e., High definition)\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "#import some helper functions for our other directory.\n",
    "import sys\n",
    "sys.path.insert(1, '../scripts/')\n",
    "from aux_functions import load_n_combine_df\n",
    "(X_train,y_train),(X_validate,y_validate),(X_test,y_test) = load_n_combine_df(path_to_data='../datasets/sevir/',features_to_keep=np.arange(0,1,1),class_labels=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "763030a8",
   "metadata": {},
   "source": [
    "Let's check to make sure the labels are indeed decimal numbers instead of 0's and 1's. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8c5fa1db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'number of flahses')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 270,
       "width": 401
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_train,bins=100)\n",
    "plt.xlabel('number of flahses')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd7e3009",
   "metadata": {},
   "source": [
    "Great, it is indeed more than just 0's and 1's, but something you'll notice right here, there are ALOT of no flash images. You will see if we plot the number of flashes as a function of the minimum brightness temperature it might be very difficult to fit a linear method (i.e., Linear regression) to the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f9943cc1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of flashes')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 325,
       "width": 348
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#this is something to help make the ticks show up where I want them \n",
    "from matplotlib.ticker import (MultipleLocator, FormatStrFormatter,\n",
    "                               AutoMinorLocator)\n",
    "\n",
    "#color I like. The order of ratios is [Red,Green,Blue]\n",
    "r = [255/255,127/255,127/255]\n",
    "\n",
    "#make figure  \n",
    "fig = plt.figure(figsize=(5,5))\n",
    "#set background color to white so we can copy paste out of the notebook if we want \n",
    "fig.set_facecolor('w')\n",
    "\n",
    "#get axis for drawing\n",
    "ax = plt.gca()\n",
    "\n",
    "#plot data \n",
    "ax.scatter(X_train[:,0],y_train,color=r,s=1,marker='+')\n",
    "\n",
    "#set limits \n",
    "ax.set_xlim([-95,30])\n",
    "#label axes\n",
    "ax.set_xlabel('Minimum Brightness Temperature, [$\\degree$C]')\n",
    "ax.set_ylabel('Number of flashes')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b3d575fd",
   "metadata": {},
   "source": [
    "For now, lets try it anyway \n",
    "\n",
    "#### Step 3: Initialize model\n",
    "\n",
    "Same as with classification, we can use the ```()``` after the model name to initialize a ML model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "aefd44a9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearRegression()\n"
     ]
    }
   ],
   "source": [
    "#load model from sklearn\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "#initialize\n",
    "model = LinearRegression()\n",
    "\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2507a32d",
   "metadata": {},
   "source": [
    "#### Step 4: Train your ML model! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4340ae46",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = model.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a8c08c9",
   "metadata": {},
   "source": [
    "#### Step 5: Evaluate your ML model\n",
    "\n",
    "As a sanity check, we will first look at the *one-to-one* plot where the x-axis is the predicted number of flashes, and the y-axis is the true number of flashes. A perfect prediction will be directly along the diagonal. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bd387cce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'GLM measurement, [$number of flashes$]')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 324,
       "width": 330
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get predictions \n",
    "yhat = model.predict(X_validate)\n",
    "\n",
    "#make figure  \n",
    "fig = plt.figure(figsize=(5,5))\n",
    "#set background color to white so we can copy paste out of the notebook if we want \n",
    "fig.set_facecolor('w')\n",
    "\n",
    "#get axis for drawing\n",
    "ax = plt.gca()\n",
    "\n",
    "#plot data \n",
    "ax.scatter(yhat,y_validate,color=r,s=1,marker='+')\n",
    "ax.plot([0,3500],[0,3500],'-k')\n",
    "ax.set_xlabel('ML Prediction, [$number of flashes$]')\n",
    "ax.set_xlabel('GLM measurement, [$number of flashes$]')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bef1b90e",
   "metadata": {},
   "source": [
    "As you can see, there is not a great correspondence between the ML model predicted flashes and the true number of flashes. One work around of this issue is to train on instances where there is already more than 1 lightning flash. While this might seem like cheating, based on our > 90% accurate classification model, we could use the two ML models in tandem. In other words, we could use the regression model only on images classified to contain flashes from the classification model. So let's drop the zeros out of the dataset and give it a try."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "de7dd49f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 325,
       "width": 348
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(X_train,y_train),(X_validate,y_validate),(X_test,y_test) = load_n_combine_df(path_to_data='../datasets/sevir/',features_to_keep=np.arange(0,1,1),class_labels=False,dropzeros=True)\n",
    "\n",
    "#remake scatter plot from Step 2\n",
    "#make figure  \n",
    "fig = plt.figure(figsize=(5,5))\n",
    "#set background color to white so we can copy paste out of the notebook if we want \n",
    "fig.set_facecolor('w')\n",
    "\n",
    "#get axis for drawing\n",
    "ax = plt.gca()\n",
    "\n",
    "#plot data \n",
    "ax.scatter(X_train[:,0],y_train,color=r,s=1,marker='+')\n",
    "\n",
    "#set limits \n",
    "ax.set_xlim([-95,30])\n",
    "#label axes\n",
    "ax.set_xlabel('Minimum Brightness Temperature, [$\\degree$C]')\n",
    "ax.set_ylabel('Number of flashes')\n",
    "\n",
    "\n",
    "print(np.min(y_train))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64d80579",
   "metadata": {},
   "source": [
    "Good, there are no 0s in the label vector (```y```). Now let's re-train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9e354516",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = model.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a9e412dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'GLM measurement, [$number of flashes$]')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 324,
       "width": 330
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get predictions \n",
    "yhat = model.predict(X_validate)\n",
    "\n",
    "#make figure  \n",
    "fig = plt.figure(figsize=(5,5))\n",
    "#set background color to white so we can copy paste out of the notebook if we want \n",
    "fig.set_facecolor('w')\n",
    "\n",
    "#get axis for drawing\n",
    "ax = plt.gca()\n",
    "\n",
    "#plot data \n",
    "ax.scatter(yhat,y_validate,color=r,s=1,marker='+')\n",
    "ax.plot([0,3500],[0,3500],'-k')\n",
    "ax.set_xlabel('ML Prediction, [$number of flashes$]')\n",
    "ax.set_xlabel('GLM measurement, [$number of flashes$]')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52e8e150",
   "metadata": {},
   "source": [
    "It's better. but still doesn’t look great. But given the relatively non-linear relationship between the minimum brightness temperature and the number of flashes, this is probably to be expected. Let's calculate some metrics to quantitatively evaluate the trained ML model.\n",
    "\n",
    "The metrics for regression are a bit different than for classification. Common metrics are the mean bias, mean absolute error (MAE), Root Mean Square Error (RMSE) and the coefficient of determination (R^2). Mathematically, these metrics are defined: \n",
    "\n",
    "$$ \\mathrm{Bias} =  \\frac{1}{N} \\sum_{j=1}^{N} (y_j - \\hat{y}_j) $$\n",
    "\n",
    "$$ \\mathrm{MAE} = \\frac{1}{N} \\sum_{j=1}^{N} |y_j - \\hat{y}_j| $$\n",
    "\n",
    "$$ \\mathrm{RMSE} = \\sqrt{\\frac{1}{N}  \\sum_{j=1}^{N} (y_j - \\hat{y}_j)^{2}} $$ \n",
    "\n",
    "$$ \\mathrm{R^{2}} = 1 - \\frac{\\sum_{j=1}^{N} (y_j - \\hat{y}_j)^{2}}{\\sum_{j=1}^{N} (y_j - \\bar{y})^{2}} $$\n",
    "\n",
    "We have included all of these metrics again in the ```gewitter_functions.py``` script."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c8a0071d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE:131.15 flashes, RMSE:202.35 flashes, Bias:6.49 flashes, Rsquared:0.19\n"
     ]
    }
   ],
   "source": [
    "from gewitter_functions import get_mae,get_rmse,get_bias,get_r2\n",
    "\n",
    "yhat = model.predict(X_validate)\n",
    "mae = get_mae(y_validate,yhat)\n",
    "rmse = get_rmse(y_validate,yhat)\n",
    "bias = get_bias(y_validate,yhat)\n",
    "r2 = get_r2(y_validate,yhat)\n",
    "\n",
    "#print them out so we can see them \n",
    "print('MAE:{} flashes, RMSE:{} flashes, Bias:{} flashes, Rsquared:{}'.format(np.round(mae,2),np.round(rmse,2),np.round(bias,2),np.round(r2,2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1324e94",
   "metadata": {},
   "source": [
    "There we go! We have a simple linear regression predicting the number of flashes in an image. While these results don’t look great (missing on average by 30 flashes) we will show in the more advanced ML model (using more features) that we can get a more accurate model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14bb77ce",
   "metadata": {},
   "source": [
    "#### Step 6: Save your trained model\n",
    "\n",
    "Sometimes loading the training dataset and re-training the model each time can be cumbersome. There is a way to save the trained models. We will use the python ```pickle``` package, I know kinda an odd name, to do this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a70fbfad",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "name = 'LinearRegression.pkl'\n",
    "start_path = '../datasets/sklearnmodels/regression/onefeature/'\n",
    "savefile = open(start_path + name,'wb')\n",
    "pickle.dump(model,savefile)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "633eeb38",
   "metadata": {},
   "source": [
    "#### Step 7: Load a saved model\n",
    "\n",
    "Now that you have it saved, if you need to load it do the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9409e50b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearRegression()\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "name = 'LinearRegression.pkl'\n",
    "start_path = '../datasets/sklearnmodels/regression/onefeature/'\n",
    "#notice the change from wb to rb \n",
    "savefile = open(start_path + name,'rb')\n",
    "#notice the change from dump to load \n",
    "model = pickle.load(savefile)\n",
    "\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8444c07",
   "metadata": {},
   "source": [
    "In the next notebook we will look at training a ML model with all 36 predictors. Notebook 6 is for classification, while Notebook 7 is for regression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "353291d8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "atmos6010",
   "language": "python",
   "name": "u0035056"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}