Synthetic-dataset-varied-pseudocounts.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6a5b101e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from skbio.stats.composition import clr\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from debiasm.torch_functions import rescale\n",
    "from scipy.stats import pointbiserialr, mannwhitneyu\n",
    "from matplotlib.colors import LogNorm, Normalize\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set(rc={'axes.facecolor':'white', \n",
    "            'figure.facecolor':'white', \n",
    "            'axes.edgecolor':'black', \n",
    "            'grid.color': 'black'\n",
    "            }, \n",
    "       font_scale=2)\n",
    "np.random.seed(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90c43603",
   "metadata": {},
   "source": [
    "### description of experiment\n",
    "- we generate a set of 50 positive and 50 negative samples\n",
    "- samples w/ a positive `y` have 30 features drawn from a uniform distribution, and one that is empty\n",
    "- samples w/ a negative `y` have 20 features drawn from a uniform distribution, and 11 empty features. \n",
    "- By construction, the 31st feature is empty for all samples; question: will a clr transform introduce an association with `y` for that empty sample?\n",
    "- We give all features an added pseudocount of $10^{-6}$, before which we rescale all samples to represent a relative abundance space\n",
    "- Next, we run clr transfom using scipy.stats.composition's implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9e750fb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_samples=100\n",
    "frac_pos=.5\n",
    "n_features_both = 20\n",
    "n_features_pos = 10\n",
    "n_features_empty = 1\n",
    "\n",
    "n_pos = int( n_samples*frac_pos )\n",
    "n_neg = n_samples-n_pos\n",
    "# X = np.zeros((n_samples,\n",
    "#              n_features_both + \\\n",
    "#              n_features_pos + \\\n",
    "#              n_features_empty\n",
    "#              ))\n",
    "\n",
    "X = np.hstack((\n",
    "    np.vstack( ( np.random.uniform(size=( n_pos, n_features_both + n_features_pos )), \n",
    "            \n",
    "                np.hstack(( np.random.uniform(size = ( n_neg, n_features_both )), \n",
    "                            np.zeros((n_neg, n_features_pos))\n",
    "                          ) ))), \n",
    "    np.zeros((n_samples, n_features_empty)) \n",
    "        )) \n",
    "\n",
    "\n",
    "y = np.array([1]*n_pos + [0]*n_neg)\n",
    "\n",
    "rsc_df = rescale(X)\n",
    "pseudocount = 10**np.floor(np.log10(rsc_df[rsc_df != 0].min()))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0a78ada",
   "metadata": {},
   "source": [
    "p-value of transformed empty feature and outcome:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4334a7b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# X_clr = clr( rescale( pseudocount + rescale(X) ) ) ## the second `rescale` is different"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "acaf012e",
   "metadata": {},
   "outputs": [],
   "source": [
    "pseudo_list = np.logspace(-8, -1, 8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5edd4648",
   "metadata": {},
   "outputs": [],
   "source": [
    "rsc_df = rescale(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "06af5a6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "varied_pseudo_efeats = np.vstack( [ clr( rescale( pseudo_ + rsc_df ) )[:, -1] \n",
    "                                    for pseudo_ in pseudo_list ] ).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fff58021",
   "metadata": {},
   "outputs": [],
   "source": [
    "vary_pseudo= pd.DataFrame( {pseudo_ : clr( rescale( pseudo_ + rsc_df ) )[:, -1] \n",
    "                                    for pseudo_ in pseudo_list } ).T\\\n",
    "                    .reset_index()\n",
    "\n",
    "vary_pseudo.columns = ['pseudocount'] + \\\n",
    "                    ['Positive' if a else 'Negative' for a in y]#list(y)\n",
    "\n",
    "vary_pseudo = vary_pseudo.melt(id_vars='pseudocount')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "584a5aa4",
   "metadata": {},
   "outputs": [],
   "source": [
    "vary_pseudo['pseudocount']=vary_pseudo['pseudocount'].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b5fb15d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# vary_pseudo['Simulated label'] = vary_pseudo['variable']\n",
    "# plt.figure(figsize=(14,8))\n",
    "# ax=sns.barplot(\n",
    "# # sns.boxplot(\n",
    "#             x='pseudocount', \n",
    "#             y='value', \n",
    "#             hue='Simulated label', \n",
    "#             data=vary_pseudo, \n",
    "#             palette={'Negative':'#3b4ea2',\n",
    "#                      'Positive':'#b11e23'}, \n",
    "#             ci=100\n",
    "# #             fliersize=0\n",
    "#            )\n",
    "# # handles, labels = ax.get_legend_handles_labels()\n",
    "# # sns.scatterplot(x='pseudocount', \n",
    "# #                 y='value', \n",
    "# #                 hue='Simulated label', \n",
    "# #                 data=vary_pseudo, \n",
    "# #                 color='black', \n",
    "# #                 s=10\n",
    "# #                 )\n",
    "\n",
    "\n",
    "# pval_thresh=0.01\n",
    "# xaxis_name='pseudocount'\n",
    "\n",
    "# sig_df = pd.DataFrame({pseudo_ :\\\n",
    "#             mannwhitneyu(\n",
    "#                      vary_pseudo.loc[\n",
    "#                          (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "#                                   .value_counts().index.values[0])&\\\n",
    "#                         (vary_pseudo['pseudocount']==pseudo_)].value, \n",
    "#                      vary_pseudo.loc[\n",
    "#                          (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "#                                   .value_counts().index.values[1])&\\\n",
    "#                         (vary_pseudo['pseudocount']==pseudo_)].value\n",
    "#                         ).pvalue\n",
    "#                 for pseudo_ in pseudo_list\n",
    "#                   }, index=[0])\n",
    "\n",
    "\n",
    "# sig_df=sig_df.T.reset_index()\n",
    "# sig_df.columns=['pseudocount', 'pvalue']\n",
    "# sig_df['is_sig'] = sig_df.pvalue < pval_thresh\n",
    "# sig_df['y_val'] = vary_pseudo.value.min() * 1.3\n",
    "# sig_df['pseudocount'] = sig_df['pseudocount'].astype(float)\n",
    "\n",
    "\n",
    "# print(sig_df)\n",
    "\n",
    "# if sum(sig_df.is_sig)>0:\n",
    "#     sns.swarmplot(x='pseudocount',\n",
    "#                   y='y_val', \n",
    "#                   data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "#                   marker='+',\n",
    "#                   size= 25/2, \n",
    "#                   order=sig_df.pseudocount.unique(),\n",
    "#                   ax=ax,\n",
    "#                   color='red',\n",
    "#                   linewidth=3.5/2\n",
    "#                   )\n",
    "\n",
    "#     sns.swarmplot(x='pseudocount', \n",
    "#                   y='y_val', \n",
    "#                   data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "#                   marker='x',\n",
    "#                   size= 18.5/2,\n",
    "#                   order=sig_df.pseudocount.unique(),\n",
    "#                   ax=ax,\n",
    "#                   color='red',\n",
    "#                   linewidth=3.5/2,\n",
    "#                   )\n",
    "\n",
    "\n",
    "# plt.xticks(ticks=range(len(pseudo_list)), \n",
    "#            labels=pseudo_list, \n",
    "#            rotation=90)\n",
    "# # # plt.yscale('symlog')\n",
    "\n",
    "# # plt.yticks([])\n",
    "# plt.ylabel('Mean CLR of an empty feature')\n",
    "# plt.xlabel('Pseudocount (added in relative abundance space)')\n",
    "# # ax.legend(handles[:2], labels[:2], loc='lower right')\n",
    "# plt.title('Compositional transforms intrduce differences in empty features\\n'+\\\n",
    "#           'across all pseudocount values in silico')\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5c94ae92",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 18.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 24.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 28.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 52.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 58.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 70.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/scipy/stats/stats.py:7028: RuntimeWarning: divide by zero encountered in double_scalars\n",
      "  z = (bigu - meanrank) / sd\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    pseudocount  pvalue  is_sig      y_val\n",
      "0  1.000000e-08     0.0    True -18.781905\n",
      "1  1.000000e-07     0.0    True -18.781905\n",
      "2  1.000000e-06     0.0    True -18.781905\n",
      "3  1.000000e-05     0.0    True -18.781905\n",
      "4  1.000000e-04     0.0    True -18.781905\n",
      "5  1.000000e-03     0.0    True -18.781905\n",
      "6  1.000000e-02     0.0    True -18.781905\n",
      "7  1.000000e-01     0.0    True -18.781905\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vary_pseudo['Simulated label'] = vary_pseudo['variable']\n",
    "plt.figure(figsize=(14,8))\n",
    "# ax=sns.barplot(\n",
    "ax=sns.boxplot(\n",
    "            x='pseudocount', \n",
    "            y='value', \n",
    "            hue='Simulated label', \n",
    "            data=vary_pseudo, \n",
    "            palette={'Negative':'#3b4ea2',\n",
    "                     'Positive':'#b11e23'}, \n",
    "#             ci=100\n",
    "            fliersize=0\n",
    "           )\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "sns.swarmplot(x='pseudocount', \n",
    "                y='value', \n",
    "                hue='Simulated label', \n",
    "                data=vary_pseudo, \n",
    "              palette={'Negative':'#3b4ea2',\n",
    "                     'Positive':'#b11e23'},\n",
    "#                 color='black', \n",
    "                s=2.5, \n",
    "                dodge=True\n",
    "                )\n",
    "\n",
    "\n",
    "pval_thresh=0.01\n",
    "xaxis_name='pseudocount'\n",
    "\n",
    "sig_df = pd.DataFrame({pseudo_ :\\\n",
    "            mannwhitneyu(\n",
    "                     vary_pseudo.loc[\n",
    "                         (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "                                  .value_counts().index.values[0])&\\\n",
    "                        (vary_pseudo['pseudocount']==pseudo_)].value, \n",
    "                     vary_pseudo.loc[\n",
    "                         (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "                                  .value_counts().index.values[1])&\\\n",
    "                        (vary_pseudo['pseudocount']==pseudo_)].value\n",
    "                        ).pvalue\n",
    "                for pseudo_ in pseudo_list\n",
    "                  }, index=[0])\n",
    "\n",
    "\n",
    "sig_df=sig_df.T.reset_index()\n",
    "sig_df.columns=['pseudocount', 'pvalue']\n",
    "sig_df['is_sig'] = sig_df.pvalue < pval_thresh\n",
    "sig_df['y_val'] = vary_pseudo.value.min() * 1.3\n",
    "sig_df['pseudocount'] = sig_df['pseudocount'].astype(float)\n",
    "\n",
    "\n",
    "print(sig_df)\n",
    "\n",
    "if sum(sig_df.is_sig)>0:\n",
    "    sns.swarmplot(x='pseudocount',\n",
    "                  y='y_val', \n",
    "                  data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "                  marker='+',\n",
    "                  size= 25/2, \n",
    "                  order=sig_df.pseudocount.unique(),\n",
    "                  ax=ax,\n",
    "                  color='red',\n",
    "                  linewidth=3.5/2\n",
    "                  )\n",
    "\n",
    "    sns.swarmplot(x='pseudocount', \n",
    "                  y='y_val', \n",
    "                  data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "                  marker='x',\n",
    "                  size= 18.5/2,\n",
    "                  order=sig_df.pseudocount.unique(),\n",
    "                  ax=ax,\n",
    "                  color='red',\n",
    "                  linewidth=3.5/2,\n",
    "                  )\n",
    "\n",
    "\n",
    "plt.xticks(ticks=range(len(pseudo_list)), \n",
    "           labels=pseudo_list, \n",
    "           rotation=90)\n",
    "# # plt.yscale('symlog')\n",
    "\n",
    "# plt.yticks([])\n",
    "plt.ylabel('Mean CLR of an empty feature')\n",
    "plt.xlabel('Pseudocount (added in relative abundance space)')\n",
    "# ax.legend(handles[:2], labels[:2], loc='lower right')\n",
    "plt.title('Compositional transforms intrduce differences in empty features\\n'+\\\n",
    "          'across all pseudocount values in silico')\n",
    "plt.savefig('supp-plots/Fig-S2a.pdf', \n",
    "            dpi=900, \n",
    "            bbox_inches='tight', \n",
    "            format='pdf'\n",
    "            )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "20821158",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 16.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 18.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 24.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 30.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 54.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 58.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/seaborn/categorical.py:1296: UserWarning: 70.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/Users/george/opt/anaconda3/envs/pl2/lib/python3.6/site-packages/scipy/stats/stats.py:7028: RuntimeWarning: divide by zero encountered in double_scalars\n",
      "  z = (bigu - meanrank) / sd\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    pseudocount  pvalue  is_sig      y_val\n",
      "0  1.000000e-08     0.0    True -18.781905\n",
      "1  1.000000e-07     0.0    True -18.781905\n",
      "2  1.000000e-06     0.0    True -18.781905\n",
      "3  1.000000e-05     0.0    True -18.781905\n",
      "4  1.000000e-04     0.0    True -18.781905\n",
      "5  1.000000e-03     0.0    True -18.781905\n",
      "6  1.000000e-02     0.0    True -18.781905\n",
      "7  1.000000e-01     0.0    True -18.781905\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vary_pseudo['Simulated label'] = vary_pseudo['variable']\n",
    "plt.figure(figsize=(14,8))\n",
    "# ax=sns.barplot(\n",
    "ax=sns.boxplot(\n",
    "            x='pseudocount', \n",
    "            y='value', \n",
    "            hue='Simulated label', \n",
    "            data=vary_pseudo, \n",
    "            palette={'Negative':'#3b4ea2',\n",
    "                     'Positive':'#b11e23'}, \n",
    "#             ci=100\n",
    "            fliersize=0\n",
    "           )\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "sns.swarmplot(x='pseudocount', \n",
    "                y='value', \n",
    "                hue='Simulated label', \n",
    "                data=vary_pseudo, \n",
    "              palette={'Negative':'#3b4ea2',\n",
    "                     'Positive':'#b11e23'},\n",
    "              edgecolor='black',\n",
    "              linewidth=.05,\n",
    "#                 color='black', \n",
    "                s=2.5, \n",
    "                dodge=True\n",
    "                )\n",
    "\n",
    "\n",
    "pval_thresh=0.01\n",
    "xaxis_name='pseudocount'\n",
    "\n",
    "sig_df = pd.DataFrame({pseudo_ :\\\n",
    "            mannwhitneyu(\n",
    "                     vary_pseudo.loc[\n",
    "                         (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "                                  .value_counts().index.values[0])&\\\n",
    "                        (vary_pseudo['pseudocount']==pseudo_)].value, \n",
    "                     vary_pseudo.loc[\n",
    "                         (vary_pseudo[xaxis_name]==vary_pseudo[xaxis_name]\\\n",
    "                                  .value_counts().index.values[1])&\\\n",
    "                        (vary_pseudo['pseudocount']==pseudo_)].value\n",
    "                        ).pvalue\n",
    "                for pseudo_ in pseudo_list\n",
    "                  }, index=[0])\n",
    "\n",
    "\n",
    "sig_df=sig_df.T.reset_index()\n",
    "sig_df.columns=['pseudocount', 'pvalue']\n",
    "sig_df['is_sig'] = sig_df.pvalue < pval_thresh\n",
    "sig_df['y_val'] = vary_pseudo.value.min() * 1.3\n",
    "sig_df['pseudocount'] = sig_df['pseudocount'].astype(float)\n",
    "\n",
    "\n",
    "print(sig_df)\n",
    "\n",
    "if sum(sig_df.is_sig)>0:\n",
    "    sns.swarmplot(x='pseudocount',\n",
    "                  y='y_val', \n",
    "                  data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "                  marker='+',\n",
    "                  size= 25/2, \n",
    "                  order=sig_df.pseudocount.unique(),\n",
    "                  ax=ax,\n",
    "                  color='red',\n",
    "                  linewidth=3.5/2\n",
    "                  )\n",
    "\n",
    "    sns.swarmplot(x='pseudocount', \n",
    "                  y='y_val', \n",
    "                  data = sig_df.loc[sig_df.is_sig].reset_index(), \n",
    "                  marker='x',\n",
    "                  size= 18.5/2,\n",
    "                  order=sig_df.pseudocount.unique(),\n",
    "                  ax=ax,\n",
    "                  color='red',\n",
    "                  linewidth=3.5/2,\n",
    "                  )\n",
    "\n",
    "\n",
    "plt.xticks(ticks=range(len(pseudo_list)), \n",
    "           labels=pseudo_list, \n",
    "           rotation=90)\n",
    "# # plt.yscale('symlog')\n",
    "\n",
    "# plt.yticks([])\n",
    "plt.ylabel('Mean CLR of an empty feature')\n",
    "plt.xlabel('Pseudocount (added in relative abundance space)')\n",
    "# ax.legend(handles[:2], labels[:2], loc='lower right')\n",
    "plt.title('Compositional transforms intrduce differences in empty features\\n'+\\\n",
    "          'across all pseudocount values in silico')\n",
    "ax.get_legend().remove()\n",
    "ax.set(\n",
    "       yticklabels=[], \n",
    "       xticklabels=[])\n",
    "plt.xlabel(None)\n",
    "plt.ylabel(None)\n",
    "plt.title(None)\n",
    "\n",
    "plt.savefig('supp-plots/Fig-S2a-no-axes.pdf', \n",
    "            dpi=900, \n",
    "            bbox_inches='tight', \n",
    "            format='pdf'\n",
    "            )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca181847",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pl2",
   "language": "python",
   "name": "pl2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}