gust_reject_full_data_analysis_for_github.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import scipy\n",
    "from scipy.stats import norm\n",
    "from scipy.stats import t\n",
    "import scipy.stats as stats\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "from utils import plot_test_curve\n",
    "from matplotlib.patches import Patch\n",
    "from matplotlib.lines import Line2D\n",
    "from matplotlib import rcParams\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.genmod.bayes_mixed_glm as glm\n",
    "import seaborn as sns\n",
    "\n",
    "rcParams['savefig.dpi'] = 400\n",
    "rcParams['figure.dpi'] = 400"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Combining all Data form all flight configs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "long_data_high_lift=pd.read_csv('metrics_data/high_lift_data_long.csv', index_col=0)\n",
    "long_data_high_lift['AoA'] = 10\n",
    "long_data_high_lift['gust_mag'] = [ 0.09 if c == 7.5 else 0.12 if c == 10 else 0.17 if c == 12.5 else 0 for c in long_data_high_lift['gust']]\n",
    "long_data_low_lift=pd.read_csv('metrics_data/low_lift_data_long.csv', index_col=0)\n",
    "long_data_low_lift['gust_mag'] = [ 0.14 if c == 4 else 0.23 if c == 6 else 0.32 if c == 8 else 0 for c in long_data_low_lift['gust']]\n",
    "long_data_medium_lift=pd.read_csv('metrics_data/med_lift_data_long.csv', index_col=0)\n",
    "long_data_medium_lift['gust_mag'] = [ 0.24 if c == 3 else 0.4 if c == 4.5 else 0.56 if c == 6 else 0 for c in long_data_medium_lift['gust']]\n",
    "\n",
    "frames = [long_data_high_lift, long_data_low_lift, long_data_medium_lift]\n",
    "\n",
    "total_data_df = pd.concat(frames).reset_index(drop=True)\n",
    "total_data_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_data_df['tap_cat'] = ['Ta' if c==6 else 'Tb' if c==3 else 'Tc' for c in total_data_df['taps']]\n",
    "total_data_df['IDnum'] = [str(int(c)) for c in total_data_df['IDnum']]\n",
    "total_data_df['gust_dir'] = total_data_df['gust_mag']*total_data_df['dir']\n",
    "total_data_df['gustID'] = [str(int(c)) for c in total_data_df['gustID']]\n",
    "total_data_df['contID'] = [str(int(c)) for c in total_data_df['contID']]\n",
    "total_data_df['AoA'] = [str(int(c)) for c in total_data_df['AoA']]\n",
    "total_data_df['speed'] = [str(int(c)) for c in total_data_df['speed']]\n",
    "total_data_df['e_reduct'] = [ c*100 for c in total_data_df['e_reduct']]\n",
    "total_data_df['flight_cond'] = total_data_df[\"AoA\"]+'_'+total_data_df[\"speed\"]\n",
    "\n",
    "total_data_df['ind'] = [str(int(c)) for c in total_data_df.index]\n",
    "total_data_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "no_1t_df = total_data_df.copy()\n",
    "ind_1t = no_1t_df[no_1t_df['taps'] == 1].index\n",
    "no_1t_df.drop(ind_1t , inplace=True)\n",
    "\n",
    "no_3t_df = total_data_df.copy()\n",
    "ind_3t = no_3t_df[no_3t_df['taps'] == 3].index\n",
    "no_3t_df.drop(ind_3t , inplace=True)\n",
    "\n",
    "no_6t_df = total_data_df.copy()\n",
    "ind_6t = no_6t_df[no_6t_df['taps'] == 6].index\n",
    "no_6t_df.drop(ind_6t , inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_data_df['red_mean'] = [total_data_df['e_reduct'][total_data_df.IDnum==c].mean() for c in total_data_df['IDnum']]\n",
    "total_data_df['red_dev'] = abs(total_data_df['e_reduct'] - total_data_df['red_mean'])\n",
    "total_data_df['rise_mean'] = [total_data_df['r_time'][total_data_df.IDnum==c].mean() for c in total_data_df['IDnum']]\n",
    "total_data_df['red_std'] = [total_data_df['e_reduct'][total_data_df.IDnum==c].std() for c in total_data_df['IDnum']]\n",
    "total_data_df['cont_gustID'] = total_data_df[\"contID\"]+'_'+total_data_df[\"gustID\"]\n",
    "total_data_df['r_t_mean'] = [total_data_df['r_time'][total_data_df.cont_gustID==c].mean() for c in total_data_df['cont_gustID']]\n",
    "total_data_df['rise_skew'] = [total_data_df['r_time'][total_data_df.IDnum==c].skew() for c in total_data_df['IDnum']]\n",
    "total_data_df['red_skew'] = [total_data_df['e_reduct'][total_data_df.IDnum==c].skew() for c in total_data_df['IDnum']]\n",
    "\n",
    "#total_data_df.to_csv('data/all_data_df.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(total_data_df[1800:1800+18]['rise_skew'])\n",
    "print(total_data_df[1800:1800+18]['rise_skew'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#first = 64 #dropped outliers\n",
    "first = 0 \n",
    "for_stats_high_lift=pd.read_csv('metrics_data/high_lift_metrics.csv', index_col=0)\n",
    "for_stats_low_lift=pd.read_csv('metrics_data/low_lift_metrics.csv', index_col=0)\n",
    "for_stats_med_lift=pd.read_csv('metrics_data/med_lift_metrics.csv', index_col=0)\n",
    "gusts_high_lift = [7.5, 10, 12.5]\n",
    "dirs = [1,-1]\n",
    "taps = [1,3,6]\n",
    "gusts_low_lift = [4, 6, 8]\n",
    "gusts_med_lift = [3,4.5,6]\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(2,3, figsize=(8,6))\n",
    "\n",
    "for i in range(len(gusts_high_lift)):\n",
    "    gust_ind = i%len(gusts_high_lift)\n",
    "    for ii in range(len(dirs)):\n",
    "        dir_ind = ii%len(dirs)\n",
    "    \n",
    "    \n",
    "        rel_stats_high_lift = for_stats_high_lift.loc[for_stats_high_lift.gust==gusts_high_lift[gust_ind]][for_stats_high_lift.dir==dirs[dir_ind]].reset_index()\n",
    "     \n",
    "        ax[0,0].errorbar(rel_stats_high_lift['taps'],rel_stats_high_lift['red_mean']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_high_lift['d_colors'][0],\n",
    "                     marker = rel_stats_high_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_high_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,0].errorbar(rel_stats_high_lift['taps'],rel_stats_high_lift['rise_mean'],yerr=0,\n",
    "                     color = rel_stats_high_lift['d_colors'][0],\n",
    "                     marker = rel_stats_high_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_high_lift['alphas'][0])#, capsize=5)\n",
    "        \n",
    "        rel_stats_med_lift = for_stats_med_lift.loc[for_stats_med_lift.gust==gusts_med_lift[gust_ind]][for_stats_med_lift.dir==dirs[dir_ind]].reset_index()\n",
    "        \n",
    "        print(rel_stats_med_lift['red_mean']*100)\n",
    "        \n",
    "        ax[0,1].errorbar(rel_stats_med_lift['taps'],rel_stats_med_lift['red_mean']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_med_lift['d_colors'][0],\n",
    "                     marker = rel_stats_med_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_med_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,1].errorbar(rel_stats_med_lift['taps'],rel_stats_med_lift['rise_mean'],yerr=0,\n",
    "                     color = rel_stats_med_lift['d_colors'][0],\n",
    "                     marker = rel_stats_med_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_med_lift['alphas'][0])#, capsize=5)\n",
    "        \n",
    "        rel_stats_low_lift = for_stats_low_lift.loc[for_stats_low_lift.gust==gusts_low_lift[gust_ind]][for_stats_low_lift.dir==dirs[dir_ind]].reset_index()\n",
    "\n",
    "        ax[0,2].errorbar(rel_stats_low_lift['taps'],rel_stats_low_lift['red_mean']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_low_lift['d_colors'][0],\n",
    "                     marker = rel_stats_low_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_low_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,2].errorbar(rel_stats_low_lift['taps'],rel_stats_low_lift['rise_mean'],yerr=0,\n",
    "                     color = rel_stats_low_lift['d_colors'][0],\n",
    "                     marker = rel_stats_low_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_low_lift['alphas'][0])#, capsize=5)\n",
    "        \n",
    "        \n",
    "        \n",
    "ax[0,0].plot(np.array([1,3,6]),np.array([1,1,1])*99,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[0,0].scatter(np.array([3,6]), np.array([99,99]),color='k',marker='o')\n",
    "ax[0,0].scatter(np.array([1]), np.array([99]),color='k',marker='o',\n",
    "                facecolor='w')\n",
    "\n",
    "ax[0,1].plot(np.array([1,3,6]),np.array([1,1,1])*99,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[0,1].scatter(np.array([3,6]), np.array([99,99]),color='k',marker='o')\n",
    "ax[0,1].scatter(np.array([1]), np.array([99]),color='k',marker='o',\n",
    "                facecolor='w')\n",
    "\n",
    "ax[0,2].plot(np.array([1,3,6]),np.array([1,1,1])*99,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[0,2].scatter(np.array([3,6]), np.array([99,99]),color='k',marker='o')\n",
    "ax[0,2].scatter(np.array([1]), np.array([99]),color='k',marker='o',\n",
    "                facecolor='w')\n",
    "\n",
    "ax[1,0].plot(np.array([1,3,6]),np.array([1,1,1])*0.1,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[1,0].scatter(np.array([1,3,6]), np.array([0.1,0.1,0.1]),color='k',marker='o')\n",
    "ax[1,1].plot(np.array([1,3,6]),np.array([1,1,1])*0.1,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[1,1].scatter(np.array([1,3,6]), np.array([0.1,0.1,0.1]),color='k',marker='o')\n",
    "\n",
    "ax[1,2].plot(np.array([1,3,6]),np.array([1,1,1])*0.1,ls=(0, (3,3,3,3)), color='k')\n",
    "ax[1,2].scatter(np.array([1,3,6]), np.array([0.1,0.1,0.1]),color='k',marker='o')\n",
    "\n",
    "\n",
    "ax[0,0].spines[\"top\"].set_visible(False)\n",
    "ax[0,0].spines[\"right\"].set_visible(False)\n",
    "ax[0,0].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,0].xaxis.set_visible(False)\n",
    "ax[1,0].spines[\"top\"].set_visible(False)\n",
    "ax[1,0].spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax[0,1].spines[\"top\"].set_visible(False)\n",
    "ax[0,1].spines[\"right\"].set_visible(False)\n",
    "ax[0,1].spines[\"left\"].set_visible(False)\n",
    "ax[0,1].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,1].yaxis.set_visible(False)\n",
    "ax[0,1].xaxis.set_visible(False)\n",
    "ax[1,1].spines[\"top\"].set_visible(False)\n",
    "ax[1,1].spines[\"right\"].set_visible(False)\n",
    "ax[1,1].spines[\"left\"].set_visible(False)\n",
    "\n",
    "ax[1,1].yaxis.set_visible(False)\n",
    "\n",
    "ax[0,0].set_ylabel('Mean Gust Reduction (%)', fontsize=14)\n",
    "ax[1,0].set_ylabel('Mean Rise Time (s)', fontsize=14)\n",
    "ax[1,0].set_xticks(taps)\n",
    "ax[1,1].set_xticks(taps)\n",
    "\n",
    "\n",
    "ax[0,0].set_ylim([40,101])\n",
    "ax[1,0].set_ylim([0,1.25])\n",
    "ax[0,0].set_xlim([0,7])\n",
    "ax[1,0].set_xlim([0,7])\n",
    "\n",
    "ax[0,0].set_xticks(taps)\n",
    "ax[0,0].set_yticks([40,60,80,100])\n",
    "ax[1,0].set_yticks([0.25,0.50,0.75,1,1.25])\n",
    "\n",
    "\n",
    "ax[0,1].set_ylim([40,101])\n",
    "ax[1,1].set_ylim([0,1.25])\n",
    "\n",
    "             \n",
    "ax[0,2].set_ylim([40,101])\n",
    "ax[1,2].set_ylim([0,1.25])\n",
    "ax[0,2].set_xlim([0,7])\n",
    "ax[1,2].set_xlim([0,7])\n",
    "\n",
    "ax[0,1].set_xlim([0,7])\n",
    "ax[1,1].set_xlim([0,7])\n",
    "\n",
    "ax[0,2].set_xticks(taps)\n",
    "\n",
    "\n",
    "ax[0,2].spines[\"top\"].set_visible(False)\n",
    "ax[0,2].spines[\"right\"].set_visible(False)\n",
    "ax[0,2].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,2].xaxis.set_visible(False)\n",
    "ax[1,2].spines[\"top\"].set_visible(False)\n",
    "ax[1,2].spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax[0,2].spines[\"top\"].set_visible(False)\n",
    "ax[0,2].spines[\"right\"].set_visible(False)\n",
    "ax[0,2].spines[\"left\"].set_visible(False)\n",
    "ax[0,2].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,2].yaxis.set_visible(False)\n",
    "ax[0,2].xaxis.set_visible(False)\n",
    "ax[1,2].spines[\"top\"].set_visible(False)\n",
    "ax[1,2].spines[\"right\"].set_visible(False)\n",
    "ax[1,2].spines[\"left\"].set_visible(False)\n",
    "ax[1,2].yaxis.set_visible(False)\n",
    "\n",
    "\n",
    "ax[1,1].set_xlabel('Number of taps', fontsize=14)\n",
    "ax[1,2].set_xticks(taps)\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig('plots/e_reduct_mean_and_rise')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "first = 0 \n",
    "for_stats_high_lift=pd.read_csv('metrics_data/high_lift_metrics.csv', index_col=0)\n",
    "for_stats_low_lift=pd.read_csv('metrics_data/low_lift_metrics.csv', index_col=0)\n",
    "for_stats_med_lift=pd.read_csv('metrics_data/med_lift_metrics.csv', index_col=0)\n",
    "gusts_high_lift = [7.5, 10, 12.5]\n",
    "dirs = [1,-1]\n",
    "taps = [1,3,6]\n",
    "gusts_low_lift = [4, 6, 8]\n",
    "gusts_med_lift = [3,4.5,6]\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(2,3, figsize=(8,6))\n",
    "\n",
    "for i in range(len(gusts_high_lift)):\n",
    "    gust_ind = i%len(gusts_high_lift)\n",
    "    for ii in range(len(dirs)):\n",
    "        dir_ind = ii%len(dirs)\n",
    "    \n",
    "    \n",
    "        rel_stats_high_lift = for_stats_high_lift.loc[for_stats_high_lift.gust==gusts_high_lift[gust_ind]][for_stats_high_lift.dir==dirs[dir_ind]].reset_index()\n",
    "     \n",
    "        ax[0,0].errorbar(rel_stats_high_lift['taps'],rel_stats_high_lift['red_std']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_high_lift['d_colors'][0],\n",
    "                     marker = rel_stats_high_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_high_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,0].errorbar(rel_stats_high_lift['taps'],rel_stats_high_lift['rise_std'],yerr=0,\n",
    "                     color = rel_stats_high_lift['d_colors'][0],\n",
    "                     marker = rel_stats_high_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_high_lift['alphas'][0])#, capsize=5)\n",
    "        \n",
    "        rel_stats_med_lift = for_stats_med_lift.loc[for_stats_med_lift.gust==gusts_med_lift[gust_ind]][for_stats_med_lift.dir==dirs[dir_ind]].reset_index()\n",
    "        \n",
    "        print(rel_stats_med_lift['red_mean']*100)\n",
    "        \n",
    "        ax[0,1].errorbar(rel_stats_med_lift['taps'],rel_stats_med_lift['red_std']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_med_lift['d_colors'][0],\n",
    "                     marker = rel_stats_med_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_med_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,1].errorbar(rel_stats_med_lift['taps'],rel_stats_med_lift['rise_std'],yerr=0,\n",
    "                     color = rel_stats_med_lift['d_colors'][0],\n",
    "                     marker = rel_stats_med_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_med_lift['alphas'][0])#, capsize=5)\n",
    "        \n",
    "        rel_stats_low_lift = for_stats_low_lift.loc[for_stats_low_lift.gust==gusts_low_lift[gust_ind]][for_stats_low_lift.dir==dirs[dir_ind]].reset_index()\n",
    "\n",
    "        ax[0,2].errorbar(rel_stats_low_lift['taps'],rel_stats_low_lift['red_std']*100,yerr=0,  # iqr,\n",
    "                     color = rel_stats_low_lift['d_colors'][0],\n",
    "                     marker = rel_stats_low_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_low_lift['alphas'][0]) #, capsize=3)\n",
    "        \n",
    "        ax[1,2].errorbar(rel_stats_low_lift['taps'],rel_stats_low_lift['rise_std'],yerr=0,\n",
    "                     color = rel_stats_low_lift['d_colors'][0],\n",
    "                     marker = rel_stats_low_lift['d_marks'][0],\n",
    "                     alpha = rel_stats_low_lift['alphas'][0])#, capsize=5)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "ax[0,0].spines[\"top\"].set_visible(False)\n",
    "ax[0,0].spines[\"right\"].set_visible(False)\n",
    "ax[0,0].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,0].xaxis.set_visible(False)\n",
    "ax[1,0].spines[\"top\"].set_visible(False)\n",
    "ax[1,0].spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax[0,1].spines[\"top\"].set_visible(False)\n",
    "ax[0,1].spines[\"right\"].set_visible(False)\n",
    "ax[0,1].spines[\"left\"].set_visible(False)\n",
    "ax[0,1].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,1].yaxis.set_visible(False)\n",
    "ax[0,1].xaxis.set_visible(False)\n",
    "ax[1,1].spines[\"top\"].set_visible(False)\n",
    "ax[1,1].spines[\"right\"].set_visible(False)\n",
    "ax[1,1].spines[\"left\"].set_visible(False)\n",
    "\n",
    "ax[1,1].yaxis.set_visible(False)\n",
    "\n",
    "ax[0,0].set_ylabel('Gust Reduction STD (%)', fontsize=14)\n",
    "ax[1,0].set_ylabel('Rise Time STD (s)', fontsize=14)\n",
    "ax[1,0].set_xticks(taps)\n",
    "ax[1,1].set_xticks(taps)\n",
    "\n",
    "\n",
    "ax[0,0].set_ylim([0,45])\n",
    "ax[1,0].set_ylim([0,1.6])\n",
    "ax[0,0].set_xlim([0,7])\n",
    "ax[1,0].set_xlim([0,7])\n",
    "\n",
    "ax[0,0].set_xticks(taps)\n",
    "ax[0,0].set_yticks([0,15,30,45])\n",
    "ax[1,0].set_yticks([0, 0.4,0.8,1.2,1.6])\n",
    "\n",
    "\n",
    "ax[0,1].set_ylim([0,45])\n",
    "ax[1,1].set_ylim([0,1.6])\n",
    "\n",
    "             \n",
    "ax[0,2].set_ylim([0,45])\n",
    "ax[1,2].set_ylim([0,1.6])\n",
    "ax[0,2].set_xlim([0,7])\n",
    "ax[1,2].set_xlim([0,7])\n",
    "\n",
    "ax[0,1].set_xlim([0,7])\n",
    "ax[1,1].set_xlim([0,7])\n",
    "\n",
    "ax[0,2].set_xticks(taps)\n",
    "\n",
    "\n",
    "ax[0,2].spines[\"top\"].set_visible(False)\n",
    "ax[0,2].spines[\"right\"].set_visible(False)\n",
    "ax[0,2].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,2].xaxis.set_visible(False)\n",
    "ax[1,2].spines[\"top\"].set_visible(False)\n",
    "ax[1,2].spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax[0,2].spines[\"top\"].set_visible(False)\n",
    "ax[0,2].spines[\"right\"].set_visible(False)\n",
    "ax[0,2].spines[\"left\"].set_visible(False)\n",
    "ax[0,2].spines[\"bottom\"].set_visible(False)\n",
    "ax[0,2].yaxis.set_visible(False)\n",
    "ax[0,2].xaxis.set_visible(False)\n",
    "ax[1,2].spines[\"top\"].set_visible(False)\n",
    "ax[1,2].spines[\"right\"].set_visible(False)\n",
    "ax[1,2].spines[\"left\"].set_visible(False)\n",
    "ax[1,2].yaxis.set_visible(False)\n",
    "\n",
    "\n",
    "ax[1,1].set_xlabel('Number of taps', fontsize=14)\n",
    "ax[1,2].set_xticks(taps)\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig('plots/e_reduct_std_and_rise')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "d_colors=['#C0822C','grey','#0F665E']\n",
    "d_marks = ['^','.','v']\n",
    "alphas = [1, 0.6, 0.3]\n",
    "fig, ax = plt.subplots(1,1, figsize=(6,3))\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "legend_elements = [Line2D([0], [0], color=d_colors[0], lw=2,alpha=alphas[0],marker='^', label='+Strong'),\n",
    "                   Line2D([0], [0], color=d_colors[0],lw=2,alpha=alphas[1], marker='^',label='+Moderate'),\n",
    "                   Line2D([0], [0], color=d_colors[0], lw=2,alpha=alphas[2],marker='^', label='+Mild'),\n",
    "                   Line2D([0], [0], color=d_colors[2], lw=2,alpha=alphas[2],marker='v', label='- Mild'),\n",
    "                   Line2D([0], [0], color=d_colors[2],lw=2,alpha=alphas[1], marker='v',label='- Moderate'),\n",
    "                   Line2D([0], [0], color=d_colors[2], lw=2,alpha=alphas[0],marker='v', label='- Strong')]\n",
    "\n",
    "fig.legend(handles=legend_elements, frameon=False, bbox_to_anchor=(0.25, 0.25),\n",
    "    loc=3, ncol=1, borderaxespad=0, title= '$\\\\bf{Gust Deflections}$')\n",
    "\n",
    "\n",
    "legend_elements2=[Line2D([0], [0], marker='o', color='w', markerfacecolor='none',\n",
    "                         markeredgecolor='k', label='significant',markersize=6),\n",
    "                  Line2D([0], [0], marker='o', color='w', markerfacecolor='k',\n",
    "                         markeredgecolor='k', label='nonsignificant',markersize=6)]\n",
    "\n",
    "fig.legend(handles=legend_elements2, frameon=False, bbox_to_anchor=(0.50, 0.37),\n",
    "    loc=3, ncol=1, borderaxespad=0, title= '$\\\\bf{Statistical Difference}$')\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig('plots/legend')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "d_colors=['#C0822C','grey','#0F665E']\n",
    "d_marks = ['^','.','v']\n",
    "alphas = [1, 0.6, 0.3]\n",
    "fig, ax = plt.subplots(1,1, figsize=(6,3))\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "legend_elements1 = [Line2D([0], [0], color=d_colors[0], lw=2,alpha=alphas[0], label='+Strong'),\n",
    "                   Line2D([0], [0], color=d_colors[0],lw=2,alpha=alphas[1], label='+Moderate'),\n",
    "                   Line2D([0], [0], color=d_colors[0], lw=2,alpha=alphas[2], label='+Mild'),\n",
    "                   Line2D([0], [0], color='k',ls=\"-\", lw=2,alpha=alphas[0], label='Average')]\n",
    "legend_elements2 = [Line2D([0], [0], color=d_colors[2], lw=2,alpha=alphas[0], label='- Strong'),\n",
    "                   Line2D([0], [0], color=d_colors[2],lw=2,alpha=alphas[1], label='- Moderate'),\n",
    "                   Line2D([0], [0], color=d_colors[2], lw=2,alpha=alphas[2], label='- Mild')]\n",
    "\n",
    "\n",
    "legend1 = fig.legend(handles=legend_elements1, frameon=False, bbox_to_anchor=(0.1, 0.55),\n",
    "    loc=3, ncol=4, borderaxespad=0, title= 'Gust Deflections')\n",
    "legend12 = fig.legend(handles=legend_elements2, frameon=False, bbox_to_anchor=(0.2, 0.5),\n",
    "    loc=3, ncol=3, borderaxespad=0)\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig('plots/other_legend')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d_colors=['#C0822C','tab:grey','#0F665E']\n",
    "d_marks = ['^','.','v']\n",
    "alphas = [1, 0.6, 0.3]\n",
    "fig, ax = plt.subplots(1,1, figsize=(6,3))\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "legend_elements1 = [Line2D([0], [0], color=d_colors[1], lw=2,alpha=0.5, label='Individual test'),\n",
    "                   Line2D([0], [0], color='k',ls=\"-\", lw=2,alpha=alphas[0], label='Controller average')]\n",
    "\n",
    "\n",
    "legend1 = fig.legend(handles=legend_elements1, frameon=False, bbox_to_anchor=(0.1, 0.55),\n",
    "    loc=3, ncol=2, borderaxespad=0)\n",
    "\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig('plots/gray_legend')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}