MFCS_FullCP_PlotFigures_20240510.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8433112a",
   "metadata": {},
   "source": [
    "# Jupyter notebook for plotting Figure 1 in \"Conformal Validity Guarnatees Exist for Any Data Distribution\" (ICML 2024)\n",
    "\n",
    "Notebook by Drew Prinster (drew@cs.jhu.edu)\n",
    "\n",
    "Last updated June 28th, 2024"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "351ed359",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f76bf255-4730-43b1-a09f-9c48b7a0eece",
   "metadata": {},
   "source": [
    "## Full CP for multistep FCS plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f2961afa-660c-45eb-9338-8cd1855c658a",
   "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>seed</th>\n",
       "      <th>step</th>\n",
       "      <th>dataset</th>\n",
       "      <th>muh_fun</th>\n",
       "      <th>method</th>\n",
       "      <th>coverage</th>\n",
       "      <th>width</th>\n",
       "      <th>muh_test</th>\n",
       "      <th>y_test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_ex</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.266371</td>\n",
       "      <td>0.095683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_1fcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.266371</td>\n",
       "      <td>0.095683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_mfcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.266371</td>\n",
       "      <td>0.095683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_ex</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.78</td>\n",
       "      <td>0.745564</td>\n",
       "      <td>0.786640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_1fcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.22</td>\n",
       "      <td>0.745564</td>\n",
       "      <td>0.786640</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14995</th>\n",
       "      <td>999</td>\n",
       "      <td>4</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_1fcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.22</td>\n",
       "      <td>0.990007</td>\n",
       "      <td>1.088215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14996</th>\n",
       "      <td>999</td>\n",
       "      <td>4</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_mfcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.22</td>\n",
       "      <td>0.990007</td>\n",
       "      <td>1.088215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14997</th>\n",
       "      <td>999</td>\n",
       "      <td>5</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_ex</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.102536</td>\n",
       "      <td>0.147410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14998</th>\n",
       "      <td>999</td>\n",
       "      <td>5</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_1fcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.102536</td>\n",
       "      <td>0.147410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14999</th>\n",
       "      <td>999</td>\n",
       "      <td>5</td>\n",
       "      <td>red</td>\n",
       "      <td>ridge</td>\n",
       "      <td>full_mfcs</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.102536</td>\n",
       "      <td>0.147410</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15000 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       seed  step dataset muh_fun     method  coverage  width  muh_test  \\\n",
       "0         0     1     red   ridge    full_ex       1.0   0.66  0.266371   \n",
       "1         0     1     red   ridge  full_1fcs       1.0   0.98  0.266371   \n",
       "2         0     1     red   ridge  full_mfcs       1.0   0.98  0.266371   \n",
       "3         0     2     red   ridge    full_ex       1.0   0.78  0.745564   \n",
       "4         0     2     red   ridge  full_1fcs       1.0   2.22  0.745564   \n",
       "...     ...   ...     ...     ...        ...       ...    ...       ...   \n",
       "14995   999     4     red   ridge  full_1fcs       1.0   2.22  0.990007   \n",
       "14996   999     4     red   ridge  full_mfcs       1.0   2.22  0.990007   \n",
       "14997   999     5     red   ridge    full_ex       1.0   0.56  0.102536   \n",
       "14998   999     5     red   ridge  full_1fcs       1.0   0.48  0.102536   \n",
       "14999   999     5     red   ridge  full_mfcs       1.0   0.82  0.102536   \n",
       "\n",
       "         y_test  \n",
       "0      0.095683  \n",
       "1      0.095683  \n",
       "2      0.095683  \n",
       "3      0.786640  \n",
       "4      0.786640  \n",
       "...         ...  \n",
       "14995  1.088215  \n",
       "14996  1.088215  \n",
       "14997  0.147410  \n",
       "14998  0.147410  \n",
       "14999  0.147410  \n",
       "\n",
       "[15000 rows x 9 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "num seeds =  1000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/dprinster/anaconda3/envs/MachineLearning/lib/python3.9/site-packages/matplotlib/tight_bbox.py:71: RuntimeWarning: divide by zero encountered in double_scalars\n",
      "  fig.patch.set_bounds(x0 / w1, y0 / h1,\n",
      "/Users/dprinster/anaconda3/envs/MachineLearning/lib/python3.9/site-packages/matplotlib/tight_bbox.py:72: RuntimeWarning: divide by zero encountered in double_scalars\n",
      "  fig.bbox.width / w1, fig.bbox.height / h1)\n",
      "/Users/dprinster/anaconda3/envs/MachineLearning/lib/python3.9/site-packages/matplotlib/patches.py:743: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  x1 = self.convert_xunits(self._x0 + self._width)\n",
      "/Users/dprinster/anaconda3/envs/MachineLearning/lib/python3.9/site-packages/matplotlib/patches.py:744: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  y1 = self.convert_yunits(self._y0 + self._height)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 0x0 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Read in data\n",
    "flur_color = 'blue'\n",
    "n_init_str = '32'\n",
    "lmbda_str = '8'\n",
    "reg_str = '0.01'\n",
    "n_steps_str = '5'\n",
    "n_seeds_str = '1000'\n",
    "plt_legend = True\n",
    "\n",
    "data = pd.read_csv('./results/2024-01-24_FullCPMultistepDesignExpts_' + flur_color + '_ridge_ntrain_init'+ n_init_str +'_steps'+n_steps_str+'_nseed'+n_seeds_str+'_lmbda'+lmbda_str+'_reg'+reg_str+'_depth2_seedinit0_PIs_replaceTrue_v2.csv')\n",
    "display(data)\n",
    "\n",
    "\n",
    "\n",
    "plt.rc(\"axes.spines\", top=False, right=False)\n",
    "# avoid Type 3 fonts\n",
    "matplotlib.rcParams['pdf.fonttype'] = 42\n",
    "matplotlib.rcParams['ps.fonttype'] = 42\n",
    "\n",
    "\n",
    "muh_name = 'ridge'\n",
    "ntrain = 64 ## 96, 192 ## \n",
    "metric = 'coverage'\n",
    "n_seeds = len(set(data['seed']))\n",
    "n_test = len(set(data['seed']))\n",
    "print(\"num seeds = \", n_seeds)\n",
    "capsize=5\n",
    "capwidth=1.75\n",
    "elinewidth=3\n",
    "\n",
    "\n",
    "\n",
    "wid_q = 0.25\n",
    "wid_e_color = 'gray'\n",
    "\n",
    "CB_color_cycle = ['#ff7f00', '#4daf4a', '#377eb8',\n",
    "                  '#e41a1c', '#984ea3', '#f781bf',  \n",
    "                  '#999999', '#dede00', '#a65628']\n",
    "\n",
    "## Plotting params\n",
    "markersize=12.5\n",
    "linewidth=3.5\n",
    "transparency=0.75\n",
    "linestyle_wid='--'\n",
    "linestyle_fit='-.'\n",
    "y_ax_label_size=22\n",
    "x_ax_label_size=22\n",
    "n_steps = 5\n",
    "\n",
    "tick_sizes=20\n",
    "suptitle_size=26\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## Coverage\n",
    "Ex_FullCP_cov = data[data['method']=='full_ex'].groupby(['step']).mean()['coverage']\n",
    "Ex_FullCP_cov_stderr = np.array(data[data['method']=='full_ex'][['coverage', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "FCS1_FullCP_cov = data[data['method']=='full_1fcs'].groupby(['step']).mean()['coverage']\n",
    "FCS1_FullCP_cov_stderr = np.array(data[data['method']=='full_ex'][['coverage', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "MFCS_FullCP_cov = data[data['method']=='full_mfcs'].groupby(['step']).mean()['coverage']\n",
    "MFCS_FullCP_cov_stderr = np.array(data[data['method']=='full_ex'][['coverage', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## Width\n",
    "Ex_FullCP_wid = data[data['method']=='full_ex'].groupby(['step']).mean()['width']\n",
    "Ex_FullCP_wid_stderr = np.array(data[data['method']=='full_ex'][['width', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "FCS1_FullCP_wid = data[data['method']=='full_1fcs'].groupby(['step']).mean()['width']\n",
    "FCS1_FullCP_wid_stderr = np.array(data[data['method']=='full_1fcs'][['width', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "MFCS_FullCP_wid = data[data['method']=='full_mfcs'].groupby(['step']).mean()['width']\n",
    "MFCS_FullCP_wid_stderr = np.array(data[data['method']=='full_mfcs'][['width', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "\n",
    "\n",
    "## Model performance\n",
    "Ex_FullCP_fit = data[data['method']=='full_ex'].groupby(['step']).mean()['y_test'] # np.abs(data[data['method']=='full_ex'].groupby(['step'])['y_test'] - data[data['method']=='full_ex'].groupby(['step'])['muh_test']).mean()\n",
    "Ex_FullCP_fit_stderr = np.array(data[data['method']=='full_ex'][['y_test', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0] #np.array(data[data['method']=='full_ex'][['MSE', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "FCS1_FullCP_fit = data[data['method']=='full_1fcs'].groupby(['step']).mean()['y_test'] # np.abs(data[data['method']=='full_ex'].groupby(['step'])['y_test'] - data[data['method']=='full_ex'].groupby(['step'])['muh_test']).mean()\n",
    "FCS1_FullCP_fit_stderr = np.array(data[data['method']=='full_1fcs'][['y_test', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0] #np.array(data[data['method']=='full_ex'][['MSE', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "MFCS_FullCP_fit = data[data['method']=='full_mfcs'].groupby(['step']).mean()['y_test'] # np.abs(data[data['method']=='full_ex'].groupby(['step'])['y_test'] - data[data['method']=='full_ex'].groupby(['step'])['muh_test']).mean()\n",
    "MFCS_FullCP_fit_stderr = np.array(data[data['method']=='full_mfcs'][['y_test', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0] #np.array(data[data['method']=='full_ex'][['MSE', 'step']].groupby(['step']).std() / np.sqrt(n_seeds)).T[0]\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## Plot coverage\n",
    "fig1, ax1 = plt.subplots(figsize=(8, 4))\n",
    "fig1.tight_layout()\n",
    "\n",
    "(_, caps, _) = ax1.errorbar(range(1, n_steps+1), Ex_FullCP_cov[0:n_steps], marker='s', yerr = Ex_FullCP_cov_stderr[0:n_steps],  label = 'Exchangeable Full CP (Vovk et al., 2005)', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[0], ecolor = CB_color_cycle[0], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "(_, caps, _) = ax1.errorbar(range(1, n_steps+1), FCS1_FullCP_cov[0:n_steps], marker='v', yerr = FCS1_FullCP_cov_stderr[0:n_steps],  label = 'One-step FCS Full CP (Fannjiang et al., 2022)', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[1], ecolor = CB_color_cycle[1], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "(_, caps, _) = ax1.errorbar(range(1, n_steps+1), MFCS_FullCP_cov[0:n_steps], marker='o', yerr = MFCS_FullCP_cov_stderr[0:n_steps], label = 'Multistep FCS Full CP (proposed)', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[2], ecolor = CB_color_cycle[2], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "\n",
    "ax1.axhline(y = 0.9, linestyle = ':', color = 'black', linewidth=linewidth*1.5, label='Target coverage')\n",
    "\n",
    "handles, labels = ax1.get_legend_handles_labels()\n",
    "ax1.set_ylabel('Mean coverage', fontsize=y_ax_label_size)\n",
    "ax1.set_xlabel('Number of design steps $t$', fontsize=x_ax_label_size)\n",
    "ax1.set_ylim([0.75,1.000001])\n",
    "ax1.set_yticks(np.arange(0.7,1.000001, 0.1))\n",
    "ax1.tick_params(axis='both', labelsize=tick_sizes)\n",
    "ax1.xaxis.set_ticks(np.arange(1, 6, 1))\n",
    "ax1.grid()\n",
    "\n",
    "fig1.savefig('./results_figures/FullCPDesignExpts_' + flur_color + '_coverage.pdf',bbox_inches='tight', dpi=300)\n",
    "\n",
    "\n",
    "\n",
    "## Plot interval widths\n",
    "\n",
    "fig2, ax2 = plt.subplots(figsize=(8, 4))\n",
    "fig2.tight_layout()\n",
    "\n",
    "(_, caps, _) = ax2.errorbar(range(1, n_steps+1), Ex_FullCP_wid[0:n_steps], marker='s', yerr = Ex_FullCP_wid_stderr[0:n_steps],  label = 'Exchangeable Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[0], ecolor = CB_color_cycle[0], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "(_, caps, _) = ax2.errorbar(range(1, n_steps+1), FCS1_FullCP_wid[0:n_steps], marker='v', yerr = FCS1_FullCP_wid_stderr[0:n_steps],  label = 'One-step FCS Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[1], ecolor = CB_color_cycle[1], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "(_, caps, _) = ax2.errorbar(range(1, n_steps+1), MFCS_FullCP_wid[0:n_steps], marker='o', yerr = MFCS_FullCP_wid_stderr[0:n_steps], label = 'Multistep FCS Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[2], ecolor = CB_color_cycle[2], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "ax2.xaxis.set_ticks(np.arange(1, 6, 1))\n",
    "\n",
    "ax2.set_ylabel('Mean interval width', fontsize=y_ax_label_size)\n",
    "ax2.set_xlabel('Number of design steps $t$', fontsize=x_ax_label_size)\n",
    "ax2.set_ylim([0,2.25])\n",
    "\n",
    "if (flur_color == 'red'):\n",
    "    ax2.set_title('Red fluorescent protein design', y=1.025, fontsize=suptitle_size)\n",
    "elif (flur_color == 'blue'):\n",
    "    ax2.set_title('Blue fluorescent protein design', y=1.025, fontsize=suptitle_size)\n",
    "\n",
    "ax2.tick_params(axis='both', labelsize=tick_sizes)\n",
    "ax2.grid()\n",
    "\n",
    "fig2.savefig('./results_figures/FullCPDesignExpts_' + flur_color + '_width.pdf',bbox_inches='tight', dpi=300)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## Plot mean fitness\n",
    "fig3, ax3 = plt.subplots(figsize=(8, 4))\n",
    "fig3.tight_layout()\n",
    "\n",
    "(_, caps, _) = ax3.errorbar(range(1, n_steps+1), Ex_FullCP_fit[0:n_steps], marker='s', yerr = Ex_FullCP_fit_stderr[0:n_steps], label = 'Exchangeable Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[0],ecolor = CB_color_cycle[0], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "\n",
    "(_, caps, _) = ax3.errorbar(range(1, n_steps+1), FCS1_FullCP_fit[0:n_steps], marker='v', yerr = FCS1_FullCP_fit_stderr[0:n_steps],  label = 'One-step FCS Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[1],ecolor = CB_color_cycle[1], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "    \n",
    "(_, caps, _) = ax3.errorbar(range(1, n_steps+1), MFCS_FullCP_fit[0:n_steps], marker='o', yerr = MFCS_FullCP_fit_stderr[0:n_steps], label = 'Multistep FCS Full CP', linewidth=linewidth, alpha =transparency, markersize=markersize, color=CB_color_cycle[2], ecolor = CB_color_cycle[2], capsize=capsize,elinewidth=elinewidth)\n",
    "for cap in caps:\n",
    "    cap.set_markeredgewidth(capwidth)\n",
    "\n",
    "ax3.set_ylabel('Mean true fitness', fontsize=y_ax_label_size)\n",
    "ax3.set_xlabel('Number of design steps $t$', fontsize=x_ax_label_size)\n",
    "ax3.tick_params(axis='both', labelsize=tick_sizes)\n",
    "ax3.xaxis.set_ticks(np.arange(1, 6, 1))\n",
    "ax3.set_ylim([0.65,1.325])\n",
    "ax3.set_yticks(np.arange(0.8,1.3, 0.2))\n",
    "\n",
    "ax3.grid()\n",
    "\n",
    "fig3.savefig('./results_figures/FullCPDesignExpts_' + flur_color + '_fitness.pdf',bbox_inches='tight', dpi=300)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## Plot legend\n",
    "fig4, ax4 = plt.subplots(figsize=(0, 0))\n",
    "ax4.set_xticks([])\n",
    "ax4.set_yticks([])\n",
    "\n",
    "order = [0, 1, 2, 3]\n",
    "ax4.legend([handles[idx] for idx in order],[labels[idx] for idx in order], loc=[-0.05,1.225],ncol=4,fontsize=14)\n",
    "\n",
    "fig4.savefig('./results_figures/FullCPDesignExpts_legend.pdf',bbox_inches='tight', dpi=200)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd716b17-5ee7-4cf5-a59d-90c0baa1442f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}