stackedbar.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from data import constants\n",
    "from mathutil import comp, percentage, analyze, electrons\n",
    "from plotutil import stackedbar\n",
    "from mathutil.analyze import groupBy\n",
    "import numpy as np\n",
    "from plotutil import periodictable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_excel(r\"C:\\Users\\jenso\\PowerFolders\\Forschung\\CSD Data Mining\\Results.xlsx\")\n",
    "df.fillna(\"\", inplace=True)\n",
    "cr_bro = pd.read_excel(r\"C:\\Users\\jenso\\PowerFolders\\Forschung\\CSD Data Mining - Chromyl\\Results.xlsx\")\n",
    "df = pd.concat([df, cr_bro])\n",
    "df[\"Group\"] = df[\"Group\"].astype(str)\n",
    "# dataset = df.loc[(df[\"Type\"] != \"Porphyrin\")]\n",
    "dataset = df.loc[(df[\"Type\"] == \"Corrole\") & (df[\"Class\"] == \"Corrole\")]\n",
    "len(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset[\"AxialLigand\"] = dataset.apply(\n",
    "    lambda x:\n",
    "    \"Py2\" if x[\"AxialLigand\"] == \"(Py)2\"\n",
    "    else x[\"AxialLigand\"]\n",
    "    ,axis = 1\n",
    ")\n",
    "dataset = electrons.createElectronData(dataset)\n",
    "# dataset = df.loc[(df[\"Type\"] !=  \"Porphyrin\")]\n",
    "dataset.reset_index(drop=True, inplace=True)\n",
    "dataset = percentage.createPercData(dataset)\n",
    "dataset = comp.createCompData(dataset)\n",
    "dataset = percentage.createCompPercData(dataset)\n",
    "dataset = comp.createSumData(dataset)\n",
    "dataset = percentage.createSumPercData(dataset)\n",
    "dataset[\"dominant\"] = dataset[constants.compColumns].idxmax(axis=1)\n",
    "len(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "group = dataset.loc[dataset[\"D\"] != \"-\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "from plotutil.misc import cm_to_inch\n",
    "import matplotlib\n",
    "! pip install SciencePlots\n",
    "! pip install natsort\n",
    "from natsort import index_natsorted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.style.use(['science', 'nature', 'no-latex'])\n",
    "plt.rcParams[\"figure.figsize\"] = (cm_to_inch(16), cm_to_inch(13))\n",
    "plt.rcParams[\"figure.dpi\"] = 1200\n",
    "plt.rcParams[\"axes.labelsize\"] = 11\n",
    "plt.rcParams[\"axes.titlesize\"] = 9\n",
    "plt.rcParams[\"xtick.labelsize\"] = 9\n",
    "plt.rcParams[\"ytick.labelsize\"] = 9\n",
    "plt.rcParams[\"font.family\"] = \"Arial\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "group = dataset.groupby(\"dominant\")\n",
    "group.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset[\"CoordNo_x\"] = dataset.apply(\n",
    "    lambda x:\n",
    "        \"freie Basen\" if x[\"CoordNo\"] == 0 \n",
    "        else \"4\" if x[\"CoordNo\"] == 4\n",
    "        else x[\"CoordNo\"] if x[\"CoordNo\"] > 4 and x[\"CoordNo\"] <=6\n",
    "        else \"> 6\",\n",
    "        axis=1\n",
    ")\n",
    "group = groupBy(dataset, constants.percCompColumns, \"CoordNo_x\")\n",
    "group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "stackedbar.plot(group, \"Koordinationszahl\", constants.compColumns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset[\"Metal_x\"] = dataset.apply(lambda x:\n",
    "                x[\"Metal\"] if x[\"Metal\"] in [\"Cr\", \"P\"]\n",
    "                else \"andere\", \n",
    "                axis=1)\n",
    "group = groupBy(dataset, constants.percCompColumns, \"Metal_x\")\n",
    "stackedbar.plot(group, \"\", constants.compColumns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "group = groupBy(dataset, constants.percCompColumns, \"d-electrons\")\n",
    "group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "stackedbar.plot(group, \"d-Elektronen\", constants.compColumns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "group = dataset.loc[(dataset[\"Metal\"] == \"Cu\")]\n",
    "group[\"Class\"] = group.apply(lambda x:\n",
    "                    \"Corrphycene\" if \"corrphycene\" in x[\"Class\"]\n",
    "                    else \"Hetero- & Isocorrole\" if x[\"Class\"] in [\"Isocorrole\", \"Heterocorrole\"]\n",
    "                    else x[\"Class\"],\n",
    "                    axis=1)\n",
    "by = \"Class\"\n",
    "group = groupBy(group, constants.percColumns, by)\n",
    "group.sort_values(\"DoopExp\", inplace=True)\n",
    "stackedbar.plot(group, \"\", constants.analysisColumns, tickRotation=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = dataset.loc[(dataset[\"Metal\"] == \"Ni\") & (dataset[\"Class\"] == \"Norcorrole\")]# & (dataset[\"Metal\"] == \"Ni\")]\n",
    "dataset[\"Metal\"] = dataset.apply(lambda x:\n",
    "    \"$H_4$\" if x[\"Ligand\"].endswith(\"(2+)\")\n",
    "    else x[\"Metal\"]\n",
    ",axis=1)\n",
    "by = \"SubstNo\"\n",
    "tm = [\"3\",\"4\",\"5\",\"6\",\"7\",\"8\",\"9\",\"10\",\"11\",\"12\"]\n",
    "group = groupBy(dataset, constants.percCompColumns, by)\n",
    "# group = groupBy(dataset, constants.percColumns, by)\n",
    "# group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "# group = group.loc[(group.index >= 4) & (group.index <=6)]\n",
    "# group.sort_values(\"DoopExp\", inplace=True)\n",
    "stackedbar.plot(group, \"Anzahl peripherer Substituenten\", constants.compColumns, print_legend=True, ncol=2)#, tickRotation=-15)\n",
    "#stackedbar.plot(group, constants.x_axis_labels[by], constants.analysisColumns, print_legend=True, ncol=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "by = \"Metal\"\n",
    "metals = [ \"Al\", \"Ga\", \"In\", \"Tl\", \"Si\", \"Ge\", \"Sn\", \"Pb\", \"P\", \"As\", \"Sb\", \"Bi\"]\n",
    "metals = [  \"Ga\", \"In\", \"Tl\", \"Si\", \"Ge\", \"Sn\", \"Pb\", \"P\", \"As\", \"Sb\", ]\n",
    "maingroups = dataset.loc[dataset[\"Metal\"].isin(metals)]\n",
    "group = groupBy(maingroups, constants.percCompColumns, by)\n",
    "# group = groupBy(dataset, constants.percColumns, by)\n",
    "group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "# group = group.loc[(group.index >= 4) & (group.index <=6)]\n",
    "stackedbar.plot(group, constants.x_axis_labels[by], constants.compColumns, print_legend=True, ncol=2)\n",
    "#stackedbar.plot(group, constants.x_axis_labels[by], constants.analysisColumns, print_legend=True, ncol=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "by = \"CoordNo_\"\n",
    "m3d = periodictable.m3d.copy()\n",
    "m4d = periodictable.m4d.copy()\n",
    "m5d = periodictable.m5d.copy()\n",
    "tm = dataset.loc[dataset[\"Metal\"].isin(m3d+m4d+m5d)]\n",
    "# tm = tm.loc[tm[\"Group\"] != \"11\"]\n",
    "tm[\"D\"] = tm.apply(\n",
    "    lambda x:\n",
    "        \"3d\" if x[\"Metal\"] in m3d\n",
    "        else \"4d\" if x[\"Metal\"] in m4d\n",
    "        else \"5d\" if x[\"Metal\"] in m5d\n",
    "        else \"\",\n",
    "    axis=1\n",
    ")\n",
    "tm[\"CoordNo_\"] = tm.apply(\n",
    "    lambda x:\n",
    "        \"4\" if x[\"CoordNo\"] == 4 and x[\"Group\"] != \"11\"\n",
    "        else \"4 (Gruppe 11)\" if x[\"CoordNo\"] == 4\n",
    "        else x[\"CoordNo\"] if x[\"CoordNo\"] > 4 and x[\"CoordNo\"] <=6\n",
    "        else \"sonstige\",\n",
    "        axis=1\n",
    ")\n",
    "\n",
    "group = groupBy(tm, constants.percCompColumns, by)\n",
    "# group = groupBy(dataset, constants.percColumns, by)\n",
    "group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "# group = group.loc[(group.index >= 4) & (group.index <=6)]\n",
    "stackedbar.plot(group, constants.x_axis_labels[by], constants.compColumns, print_legend=False, ncol=2)\n",
    "#stackedbar.plot(group, constants.x_axis_labels[by], constants.analysisColumns, print_legend=True, ncol=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "by=\"CCDC\"\n",
    "le_group = [\"Zn\"]#, \"Ag\", \"Au\"]#[\"Ni\"]#, \"Pd\", \"Pt\"]#[\"Co\"]#, \"Rh\", \"Ir\"]#[\"Fe\", \"Ru\", \"Os\"]\n",
    "data_group = dataset.loc[dataset[\"Metal\"].isin(le_group)]\n",
    "# data_group = data_group.loc[data_group[\"SubstNo\"] == 11]\n",
    "data_group[\"Rad\"] = data_group.apply(lambda x:\n",
    "    \"At-Komplex\" if x[\"Ligand\"].endswith(\"-\") or \"+\" in x[\"CoSolv\"]\n",
    "    # else \"Kation\" if x[\"Ligand\"].endswith(\"+\") or \"-\" in x[\"CoSolv\"]    else \"Dimer\" if x[\"Ligand\"].endswith(\"Dimer\")\n",
    "    else \"β-annelliert\" if \"Benzo\" in x[\"Ligand\"] or \"Pyrazino\" in x[\"Ligand\"] \n",
    "    else \"Oktaalkyl\" if x[\"Ligand\"].startswith(\"TE\") or x[\"Ligand\"].startswith(\"OE\") or x[\"Ligand\"].startswith(\"HM\") or x[\"Ligand\"].startswith(\"HE\")\n",
    "    # else \"Verkn. Cβ-meso-Aryl\" if \"Naphto\" in x[\"Ligand\"]\n",
    "    else \"Octabromo\" if x[\"Ligand\"].startswith(\"Br8\")\n",
    "    else \"Octaiodo\" if x[\"Ligand\"].startswith(\"I8\")\n",
    "    else \"Octafluoro\" if x[\"Ligand\"].startswith(\"F8\")\n",
    "    else \"Octa-$CF_3$\" if x[\"Ligand\"].startswith(\"(CF3)8\")\n",
    "    else \"Octaaryl\" if x[\"Ligand\"].startswith(\"Ph8\") or x[\"Ligand\"].startswith(\"Tol8\") or x[\"Ligand\"].startswith(\"(OMePh)\")or x[\"Ligand\"] or x[\"Ligand\"].startswith(\"(CF3Ph)8\") or x[\"Ligand\"].startswith(\"(CF3-Ph)8\")\n",
    "    else \"vollsubst.\" if x[\"SubstNo\"] == 11\n",
    "    else \"andere\", axis=1)\n",
    "# data_group[\"AxialLigand\"] = data_group.apply(lambda x:\n",
    "#     \"$(NH_3)_2$\" if x[\"AxialLigand\"] == \"(NH3)2\"\n",
    "#     else \"$Py_2$\" if x[\"AxialLigand\"] == \"Py2\"\n",
    "#     else \"$OMePy_2$\" if x[\"AxialLigand\"] == \"OMePy2\"\n",
    "#     else \"$PPh_3$\" if x[\"AxialLigand\"] == \"PPh3\"\n",
    "#     else \"$\" + x[\"AxialLigand\"] + \"$\", \n",
    "#     axis=1)\n",
    "# iv = [\"Cl\", \"Me\", \"Ph\", \"OH\", \"F\", \"Triflate\",\"μ-O\"]\n",
    "# iii = [\"H2O\", \"Py\", \"Py2\", \"(Py)2\", \"MeCN\"]\n",
    "# data_group[\"Ox\"] = data_group.apply(\n",
    "#     lambda x: \n",
    "#     \"+III\" if x[\"AxialLigand\"] in iii\n",
    "#     else \"+IV\" if x[\"AxialLigand\"] in iv \n",
    "#     else \"NO\" if x[\"AxialLigand\"] == \"NO\"\n",
    "#     else \"+II\" \n",
    "#     ,axis=1)\n",
    "# data_group = data_group.loc[data_group[\"Ox\"] == \"+IV\"]\n",
    "group = groupBy(data_group, constants.percCompColumns, by)\n",
    "# group = groupBy(data_group.loc[data_group[\"Ox\"].isin([\"+IV\"])], constants.percCompColumns, by)\n",
    "group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "# group.sort_index(key=lambda x: np.argsort(index_natsorted(group.index)), inplace=True)\n",
    "stackedbar.plot(group, \"CCDC Code\", constants.compColumns, print_legend=False, ncol=2)#, tickRotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "by=\"Ox\"\n",
    "le_group = [\"Mn\"]#, \"Tc\", \"Re\"]\n",
    "# le_group = [\"Zn\", \"Hg\", \"Cd\"]\n",
    "# le_group = [\"Cu\", \"Ag\", \"Au\"]\n",
    "data_group = dataset.loc[dataset[\"Metal\"].isin(le_group)]\n",
    "# data_group = dataset.loc[dataset[\"Group\"] == \"Ln\"]\n",
    "# group = groupBy(data_group, constants.percColumns, by)\n",
    "iv = [\"Cl\", \"OH\", \"I\", \"Br\", \"Ph\", \"Br-Ph\"]\n",
    "v = [\"O\", \"N\", \"NMes\"]\n",
    "data_group[\"Ox\"] = data_group.apply(\n",
    "    lambda x: \n",
    "    \"Cl\" if x[\"AxialLigand\"] in []#\"Cl\", \"I\", \"Br\", \"Br-Ph\"]\n",
    "    else \"+V\" if x[\"AxialLigand\"] in v\n",
    "    else \"+IV\" if x[\"AxialLigand\"] in iv \n",
    "    else \"+III\"\n",
    "    ,axis=1)\n",
    "ox = [\"+III\", \"+IV\",\"+V\", \"Cl\"]\n",
    "group = groupBy(data_group, constants.percCompColumns, by) #data_group.loc[data_group[\"Ox\"].isin([\"+IV\"])],\n",
    "# group.sort_index(key=lambda x: np.argsort(le_group), inplace=True)\n",
    "# group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "stackedbar.plot(group, \"Formale Oxidationsstufe\", constants.compColumns, print_legend=False, ncol=2)#, tickRotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "by=\"AxialLigand\"\n",
    "le_group = [\"Fe\"]\n",
    "data_group = dataset.loc[dataset[\"Metal\"].isin(le_group)]\n",
    "iv = [\"Cl\", \"Me\", \"Ph\", \"OH\", \"F\", \"Triflate\",\"μ-O\"]\n",
    "iii = [\"H2O\", \"Py\", \"Py2\", \"(Py)2\", \"MeCN\"]\n",
    "data_group[\"Ox\"] = data_group.apply(\n",
    "    lambda x: \n",
    "    \"+III\" if x[\"AxialLigand\"] in iii\n",
    "    else \"+IV\" if x[\"AxialLigand\"] in iv \n",
    "    else \"NO\" if x[\"AxialLigand\"] == \"NO\"\n",
    "    else \"+II\" \n",
    "    ,axis=1)\n",
    "ox = [\"+III\", \"+IV\",\"+V\", \"Cl\"]\n",
    "group = groupBy(data_group, constants.percColumns, by) #data_group.loc[data_group[\"Ox\"].isin([\"+IV\"])],\n",
    "# group.sort_index(key=lambda x: np.argsort(le_group), inplace=True)\n",
    "# group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "stackedbar.plot(group, \"Formale Oxidationsstufe\", constants.analysisColumns, print_legend=False, ncol=2)#, tickRotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "III_ls = [\"Py2\", \"(Py)2\"]\n",
    "III_is = [\"Py\", \"MeCN\", \"H2O\"]\n",
    "III_is_non_innocent = [\"Cl\", \"F\", \"μ-O\", \"Triflate\"] # aufpassen  Fe[Me8TPC]Cl ist ferrom. coupled S=2 vs. S=1 bei sonst\n",
    "IV_ls = [\"Ph\", \"Me\", \"OH\"] # S=1\n",
    "muO = []\n",
    "\n",
    "by=\"Ox\"\n",
    "le_group = [\"Fe\"]\n",
    "data_group = dataset.loc[dataset[\"Metal\"].isin(le_group)]\n",
    "data_group[\"Ox\"] = data_group.apply(\n",
    "    lambda x: \n",
    "    \"+III l.s.\" if x[\"AxialLigand\"] in III_ls\n",
    "    else \"+III i.s.\" if x[\"AxialLigand\"] in III_is\n",
    "    # else \"+III i.s. non-innocent f.\" if x[\"CCDC\"] == \"CAYCAH\"\n",
    "    else \"+III i.s. [Cor]*\" if x[\"AxialLigand\"] in III_is_non_innocent and x[\"CCDC\"] != \"CAYCAH\"\n",
    "    else \"+IV\" if x[\"AxialLigand\"] in IV_ls \n",
    "    else \"NO\" if x[\"AxialLigand\"] == \"NO\"\n",
    "    else \"mu\" if x[\"AxialLigand\"] in muO\n",
    "    else \"+II\" if x[\"CCDC\"] != \"CAYCAH\"\n",
    "    else \"\"\n",
    "    ,axis=1)\n",
    "data_group = data_group.loc[data_group[\"CCDC\"] != \"CAYCAH\"]\n",
    "# ox = [\"+III\", \"+IV\",\"+V\", \"Cl\"]\n",
    "group = groupBy(data_group, constants.percCompColumns, by) #data_group.loc[data_group[\"Ox\"].isin([\"+IV\"])],\n",
    "# group.sort_index(key=lambda x: np.argsort(ox), inplace=True)\n",
    "# group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "stackedbar.plot(group, \"Oxidationsstufe\", constants.compColumns, print_legend=False, ncol=2)#, tickRotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from data import datautil\n",
    "data_hic = datautil.applyHICSubclass(dataset)\n",
    "data_hic[\"Error\"] = (data_hic[\"DoopExp\"] - data_hic[\"DoopSim\"])/data_hic[\"DoopExp\"] * 100\n",
    "data_hic = datautil.getHeteroAtom(data_hic)\n",
    "data_hic = datautil.getIsoSubstituent(data_hic)\n",
    "by = \"Iso\"\n",
    "data_hic = data_hic.loc[data_hic[\"Subclass\"] == \"10-Isocorrole\"]\n",
    "# data_hic[\"HeteroM\"] = data_hic[\"Metal\"]  + \"/\" + data_hic[\"Heteroatom\"]\n",
    "# data_hic = data_hic.loc[~data_hic[\"Subclass\"].isin([\"Nope\"])]#, \"sonstige Isocorrole\"])\n",
    "group = groupBy(data_hic, constants.percCompColumns, by)\n",
    "group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "# group.sort_index(key=lambda x: np.argsort([\"N-Heterocorrole\", \"10-Heterocorrole\", \"Corrolazine\", \"5-Isocorrole\", \"10-Isocorrole\", \"sonstige Isocorrole\"]), inplace=True)\n",
    "\n",
    "fig, ax = stackedbar.plot(group, \"\", constants.compColumns, print_legend=True, ncol=2, tickRotation=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cn = dataset.loc[dataset[\"Class\"] == \"Corrolazine\"]\n",
    "group = groupBy(cn, constants.percCompColumns, \"Metal\")\n",
    "stackedbar.plot(group, \"\", constants.compColumns, print_legend=True, ncol=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_n = dataset.loc[dataset[\"Class\"].isin([\"NR-Corrole\"])]#[\"NConfused-Corrole\"])]#\"NR-Corrole\", \n",
    "data_n = datautil.getConfusedSubclass(data_n)\n",
    "data_n = datautil.getNSubst(data_n)\n",
    "data_n[\"Special\"] = data_n[\"Confusion\"] + data_n[\"NSubst\"]\n",
    "data_n[\"Special\"] = data_n.apply(\n",
    "    lambda x:\n",
    "    \"$N_A$,$N_B$-verbrückt\" if (x[\"Special\"] == \"$N_A$-CO-$N_B$\" or x[\"Special\"] == \"$N_A$-$CH_2$-$N_B$\")\n",
    "    else \"$N_A$,$N_B$-disubst.\" if x[\"Special\"] == \"$N_B$-Me & $N_A$-Me\"\n",
    "    else \"$N_A$ substituiert\" if (x[\"Special\"].startswith(\"$N_A\"))\n",
    "    else \"$N_B$ substituiert\" if x[\"Special\"].startswith(\"$N_B\")\n",
    "    else x[\"Special\"]\n",
    "    ,axis=1)\n",
    "group = groupBy(data_n, constants.percCompColumns, \"Metal\")\n",
    "\n",
    "group.sort_values(\"DoopExp\", inplace=True, ascending=True)\n",
    "stackedbar.plot(group, \"\", constants.compColumns, print_legend=False, ncol=2)#, tickRotation=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "all = datautil.applyAllSubclasses(dataset)\n",
    "all_group = groupBy(all, constants.percCompColumns, \"Subclass\")\n",
    "stackedbar.plot(all_group, \"\", constants.compColumns, tickRotation=90)"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}