diff --git "a/notebooks/plot_experimental_results.ipynb" "b/notebooks/plot_experimental_results.ipynb"
new file mode 100644--- /dev/null
+++ "b/notebooks/plot_experimental_results.ipynb"
@@ -0,0 +1,1845 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import sys\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "import pandas as pd\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "palette = ['#83B8FE', '#FFA54C', '#94ED67', '#FF7FFF']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "cv_train: (15, 52)\n"
+     ]
+    },
+    {
+     "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>model_type</th>\n",
+       "      <th>fold</th>\n",
+       "      <th>train_len</th>\n",
+       "      <th>val_len</th>\n",
+       "      <th>train_perc</th>\n",
+       "      <th>val_perc</th>\n",
+       "      <th>train_active_perc</th>\n",
+       "      <th>train_inactive_perc</th>\n",
+       "      <th>train_avg_tanimoto_dist</th>\n",
+       "      <th>val_active_perc</th>\n",
+       "      <th>...</th>\n",
+       "      <th>val_roc_auc</th>\n",
+       "      <th>test_loss</th>\n",
+       "      <th>test_acc</th>\n",
+       "      <th>test_f1_score</th>\n",
+       "      <th>test_precision</th>\n",
+       "      <th>test_recall</th>\n",
+       "      <th>test_roc_auc</th>\n",
+       "      <th>split_type</th>\n",
+       "      <th>train_unique_groups</th>\n",
+       "      <th>val_unique_groups</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>0</td>\n",
+       "      <td>616</td>\n",
+       "      <td>155</td>\n",
+       "      <td>0.798962</td>\n",
+       "      <td>0.201038</td>\n",
+       "      <td>0.514610</td>\n",
+       "      <td>0.485390</td>\n",
+       "      <td>0.376475</td>\n",
+       "      <td>0.516129</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.904333</td>\n",
+       "      <td>0.661766</td>\n",
+       "      <td>0.802326</td>\n",
+       "      <td>0.817204</td>\n",
+       "      <td>0.716981</td>\n",
+       "      <td>0.95</td>\n",
+       "      <td>0.860326</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>1</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.513776</td>\n",
+       "      <td>0.486224</td>\n",
+       "      <td>0.377372</td>\n",
+       "      <td>0.519481</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.924493</td>\n",
+       "      <td>0.658051</td>\n",
+       "      <td>0.755814</td>\n",
+       "      <td>0.764045</td>\n",
+       "      <td>0.693878</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>0.858152</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>2</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.376572</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.931139</td>\n",
+       "      <td>0.611611</td>\n",
+       "      <td>0.779070</td>\n",
+       "      <td>0.781609</td>\n",
+       "      <td>0.723404</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>0.863587</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>3</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.376068</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.924557</td>\n",
+       "      <td>0.693243</td>\n",
+       "      <td>0.732558</td>\n",
+       "      <td>0.747253</td>\n",
+       "      <td>0.666667</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>0.847826</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>4</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.377543</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.927257</td>\n",
+       "      <td>0.561546</td>\n",
+       "      <td>0.779070</td>\n",
+       "      <td>0.781609</td>\n",
+       "      <td>0.723404</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>0.878804</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 52 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  model_type  fold  train_len  val_len  train_perc  val_perc  \\\n",
+       "0    Pytorch     0        616      155    0.798962  0.201038   \n",
+       "1    Pytorch     1        617      154    0.800259  0.199741   \n",
+       "2    Pytorch     2        617      154    0.800259  0.199741   \n",
+       "3    Pytorch     3        617      154    0.800259  0.199741   \n",
+       "4    Pytorch     4        617      154    0.800259  0.199741   \n",
+       "\n",
+       "   train_active_perc  train_inactive_perc  train_avg_tanimoto_dist  \\\n",
+       "0           0.514610             0.485390                 0.376475   \n",
+       "1           0.513776             0.486224                 0.377372   \n",
+       "2           0.515397             0.484603                 0.376572   \n",
+       "3           0.515397             0.484603                 0.376068   \n",
+       "4           0.515397             0.484603                 0.377543   \n",
+       "\n",
+       "   val_active_perc  ...  val_roc_auc  test_loss  test_acc  test_f1_score  \\\n",
+       "0         0.516129  ...     0.904333   0.661766  0.802326       0.817204   \n",
+       "1         0.519481  ...     0.924493   0.658051  0.755814       0.764045   \n",
+       "2         0.512987  ...     0.931139   0.611611  0.779070       0.781609   \n",
+       "3         0.512987  ...     0.924557   0.693243  0.732558       0.747253   \n",
+       "4         0.512987  ...     0.927257   0.561546  0.779070       0.781609   \n",
+       "\n",
+       "   test_precision  test_recall  test_roc_auc  split_type  train_unique_groups  \\\n",
+       "0        0.716981         0.95      0.860326      random                  NaN   \n",
+       "1        0.693878         0.85      0.858152      random                  NaN   \n",
+       "2        0.723404         0.85      0.863587      random                  NaN   \n",
+       "3        0.666667         0.85      0.847826      random                  NaN   \n",
+       "4        0.723404         0.85      0.878804      random                  NaN   \n",
+       "\n",
+       "   val_unique_groups  \n",
+       "0                NaN  \n",
+       "1                NaN  \n",
+       "2                NaN  \n",
+       "3                NaN  \n",
+       "4                NaN  \n",
+       "\n",
+       "[5 rows x 52 columns]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "test: (9, 34)\n"
+     ]
+    },
+    {
+     "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>train_loss</th>\n",
+       "      <th>train_loss_step</th>\n",
+       "      <th>train_loss_epoch</th>\n",
+       "      <th>train_acc</th>\n",
+       "      <th>train_acc_epoch</th>\n",
+       "      <th>train_f1_score</th>\n",
+       "      <th>train_f1_score_epoch</th>\n",
+       "      <th>train_precision</th>\n",
+       "      <th>train_precision_epoch</th>\n",
+       "      <th>train_recall</th>\n",
+       "      <th>...</th>\n",
+       "      <th>train_avg_tanimoto_dist</th>\n",
+       "      <th>test_len</th>\n",
+       "      <th>test_active_perc</th>\n",
+       "      <th>test_inactive_perc</th>\n",
+       "      <th>test_avg_tanimoto_dist</th>\n",
+       "      <th>num_leaking_uniprot_train_test</th>\n",
+       "      <th>num_leaking_smiles_train_test</th>\n",
+       "      <th>perc_leaking_uniprot_train_test</th>\n",
+       "      <th>perc_leaking_smiles_train_test</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.190532</td>\n",
+       "      <td>0.307893</td>\n",
+       "      <td>0.190532</td>\n",
+       "      <td>0.939547</td>\n",
+       "      <td>0.939547</td>\n",
+       "      <td>0.940150</td>\n",
+       "      <td>0.940150</td>\n",
+       "      <td>0.930864</td>\n",
+       "      <td>0.930864</td>\n",
+       "      <td>0.949622</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.188614</td>\n",
+       "      <td>0.163636</td>\n",
+       "      <td>0.188614</td>\n",
+       "      <td>0.940806</td>\n",
+       "      <td>0.940806</td>\n",
+       "      <td>0.941760</td>\n",
+       "      <td>0.941760</td>\n",
+       "      <td>0.926829</td>\n",
+       "      <td>0.926829</td>\n",
+       "      <td>0.957179</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0.189793</td>\n",
+       "      <td>0.201865</td>\n",
+       "      <td>0.189793</td>\n",
+       "      <td>0.942065</td>\n",
+       "      <td>0.942065</td>\n",
+       "      <td>0.942786</td>\n",
+       "      <td>0.942786</td>\n",
+       "      <td>0.931204</td>\n",
+       "      <td>0.931204</td>\n",
+       "      <td>0.954660</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.543100</td>\n",
+       "      <td>0.583436</td>\n",
+       "      <td>0.543100</td>\n",
+       "      <td>0.809463</td>\n",
+       "      <td>0.809463</td>\n",
+       "      <td>0.820265</td>\n",
+       "      <td>0.820265</td>\n",
+       "      <td>0.776256</td>\n",
+       "      <td>0.776256</td>\n",
+       "      <td>0.869565</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.375305</td>\n",
+       "      <td>85</td>\n",
+       "      <td>0.541176</td>\n",
+       "      <td>0.458824</td>\n",
+       "      <td>0.394830</td>\n",
+       "      <td>0</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.011658</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.649270</td>\n",
+       "      <td>0.631336</td>\n",
+       "      <td>0.649270</td>\n",
+       "      <td>0.767263</td>\n",
+       "      <td>0.767263</td>\n",
+       "      <td>0.795964</td>\n",
+       "      <td>0.795964</td>\n",
+       "      <td>0.708583</td>\n",
+       "      <td>0.708583</td>\n",
+       "      <td>0.907928</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.375305</td>\n",
+       "      <td>85</td>\n",
+       "      <td>0.541176</td>\n",
+       "      <td>0.458824</td>\n",
+       "      <td>0.394830</td>\n",
+       "      <td>0</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.011658</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 34 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   train_loss  train_loss_step  train_loss_epoch  train_acc  train_acc_epoch  \\\n",
+       "0    0.190532         0.307893          0.190532   0.939547         0.939547   \n",
+       "1    0.188614         0.163636          0.188614   0.940806         0.940806   \n",
+       "2    0.189793         0.201865          0.189793   0.942065         0.942065   \n",
+       "0    0.543100         0.583436          0.543100   0.809463         0.809463   \n",
+       "1    0.649270         0.631336          0.649270   0.767263         0.767263   \n",
+       "\n",
+       "   train_f1_score  train_f1_score_epoch  train_precision  \\\n",
+       "0        0.940150              0.940150         0.930864   \n",
+       "1        0.941760              0.941760         0.926829   \n",
+       "2        0.942786              0.942786         0.931204   \n",
+       "0        0.820265              0.820265         0.776256   \n",
+       "1        0.795964              0.795964         0.708583   \n",
+       "\n",
+       "   train_precision_epoch  train_recall  ...  train_avg_tanimoto_dist  \\\n",
+       "0               0.930864      0.949622  ...                 0.376806   \n",
+       "1               0.926829      0.957179  ...                 0.376806   \n",
+       "2               0.931204      0.954660  ...                 0.376806   \n",
+       "0               0.776256      0.869565  ...                 0.375305   \n",
+       "1               0.708583      0.907928  ...                 0.375305   \n",
+       "\n",
+       "   test_len  test_active_perc  test_inactive_perc  test_avg_tanimoto_dist  \\\n",
+       "0        86          0.465116            0.534884                0.381147   \n",
+       "1        86          0.465116            0.534884                0.381147   \n",
+       "2        86          0.465116            0.534884                0.381147   \n",
+       "0        85          0.541176            0.458824                0.394830   \n",
+       "1        85          0.541176            0.458824                0.394830   \n",
+       "\n",
+       "   num_leaking_uniprot_train_test  num_leaking_smiles_train_test  \\\n",
+       "0                              34                             44   \n",
+       "1                              34                             44   \n",
+       "2                              34                             44   \n",
+       "0                               0                              6   \n",
+       "1                               0                              6   \n",
+       "\n",
+       "   perc_leaking_uniprot_train_test  perc_leaking_smiles_train_test split_type  \n",
+       "0                         0.832685                        0.102464     random  \n",
+       "1                         0.832685                        0.102464     random  \n",
+       "2                         0.832685                        0.102464     random  \n",
+       "0                         0.000000                        0.011658    uniprot  \n",
+       "1                         0.000000                        0.011658    uniprot  \n",
+       "\n",
+       "[5 rows x 34 columns]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "hparam: (3, 7)\n"
+     ]
+    },
+    {
+     "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>hidden_dim</th>\n",
+       "      <th>learning_rate</th>\n",
+       "      <th>smote_k_neighbors</th>\n",
+       "      <th>use_smote</th>\n",
+       "      <th>dropout</th>\n",
+       "      <th>use_batch_norm</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>128</td>\n",
+       "      <td>0.000925</td>\n",
+       "      <td>10</td>\n",
+       "      <td>True</td>\n",
+       "      <td>0.216350</td>\n",
+       "      <td>True</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>32</td>\n",
+       "      <td>0.000649</td>\n",
+       "      <td>4</td>\n",
+       "      <td>True</td>\n",
+       "      <td>0.092638</td>\n",
+       "      <td>False</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>64</td>\n",
+       "      <td>0.000771</td>\n",
+       "      <td>13</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.304535</td>\n",
+       "      <td>True</td>\n",
+       "      <td>tanimoto</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   hidden_dim  learning_rate  smote_k_neighbors  use_smote   dropout  \\\n",
+       "0         128       0.000925                 10       True  0.216350   \n",
+       "0          32       0.000649                  4       True  0.092638   \n",
+       "0          64       0.000771                 13      False  0.304535   \n",
+       "\n",
+       "   use_batch_norm split_type  \n",
+       "0            True     random  \n",
+       "0           False    uniprot  \n",
+       "0            True   tanimoto  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "majority_vote: (6, 28)\n"
+     ]
+    },
+    {
+     "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>test_acc</th>\n",
+       "      <th>test_roc_auc</th>\n",
+       "      <th>test_precision</th>\n",
+       "      <th>test_recall</th>\n",
+       "      <th>test_f1</th>\n",
+       "      <th>train_len</th>\n",
+       "      <th>train_active_perc</th>\n",
+       "      <th>train_inactive_perc</th>\n",
+       "      <th>train_avg_tanimoto_dist</th>\n",
+       "      <th>test_len</th>\n",
+       "      <th>...</th>\n",
+       "      <th>val_active_perc</th>\n",
+       "      <th>val_inactive_perc</th>\n",
+       "      <th>val_avg_tanimoto_dist</th>\n",
+       "      <th>num_leaking_uniprot_train_val</th>\n",
+       "      <th>num_leaking_smiles_train_val</th>\n",
+       "      <th>perc_leaking_uniprot_train_val</th>\n",
+       "      <th>perc_leaking_smiles_train_val</th>\n",
+       "      <th>cv_models</th>\n",
+       "      <th>model_type</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.825581</td>\n",
+       "      <td>0.847826</td>\n",
+       "      <td>0.777778</td>\n",
+       "      <td>0.875000</td>\n",
+       "      <td>0.823529</td>\n",
+       "      <td>771</td>\n",
+       "      <td>0.514916</td>\n",
+       "      <td>0.485084</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.813953</td>\n",
+       "      <td>0.868478</td>\n",
+       "      <td>0.815789</td>\n",
+       "      <td>0.775000</td>\n",
+       "      <td>0.794872</td>\n",
+       "      <td>617</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.377543</td>\n",
+       "      <td>86</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>0.487013</td>\n",
+       "      <td>0.373853</td>\n",
+       "      <td>40.0</td>\n",
+       "      <td>70.0</td>\n",
+       "      <td>0.910859</td>\n",
+       "      <td>0.176661</td>\n",
+       "      <td>True</td>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.611765</td>\n",
+       "      <td>0.614827</td>\n",
+       "      <td>0.675676</td>\n",
+       "      <td>0.543478</td>\n",
+       "      <td>0.602410</td>\n",
+       "      <td>772</td>\n",
+       "      <td>0.506477</td>\n",
+       "      <td>0.493523</td>\n",
+       "      <td>0.375305</td>\n",
+       "      <td>85</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.411765</td>\n",
+       "      <td>0.549610</td>\n",
+       "      <td>0.400000</td>\n",
+       "      <td>0.173913</td>\n",
+       "      <td>0.242424</td>\n",
+       "      <td>693</td>\n",
+       "      <td>0.484848</td>\n",
+       "      <td>0.515152</td>\n",
+       "      <td>0.377092</td>\n",
+       "      <td>85</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.696203</td>\n",
+       "      <td>0.303797</td>\n",
+       "      <td>0.359625</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.014430</td>\n",
+       "      <td>True</td>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.823761</td>\n",
+       "      <td>0.772727</td>\n",
+       "      <td>0.459459</td>\n",
+       "      <td>0.576271</td>\n",
+       "      <td>772</td>\n",
+       "      <td>0.518135</td>\n",
+       "      <td>0.481865</td>\n",
+       "      <td>0.372540</td>\n",
+       "      <td>85</td>\n",
+       "      <td>...</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Pytorch</td>\n",
+       "      <td>tanimoto</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 28 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   test_acc  test_roc_auc  test_precision  test_recall   test_f1  train_len  \\\n",
+       "0  0.825581      0.847826        0.777778     0.875000  0.823529        771   \n",
+       "1  0.813953      0.868478        0.815789     0.775000  0.794872        617   \n",
+       "0  0.611765      0.614827        0.675676     0.543478  0.602410        772   \n",
+       "1  0.411765      0.549610        0.400000     0.173913  0.242424        693   \n",
+       "0  0.705882      0.823761        0.772727     0.459459  0.576271        772   \n",
+       "\n",
+       "   train_active_perc  train_inactive_perc  train_avg_tanimoto_dist  test_len  \\\n",
+       "0           0.514916             0.485084                 0.376806        86   \n",
+       "1           0.515397             0.484603                 0.377543        86   \n",
+       "0           0.506477             0.493523                 0.375305        85   \n",
+       "1           0.484848             0.515152                 0.377092        85   \n",
+       "0           0.518135             0.481865                 0.372540        85   \n",
+       "\n",
+       "   ...  val_active_perc  val_inactive_perc  val_avg_tanimoto_dist  \\\n",
+       "0  ...              NaN                NaN                    NaN   \n",
+       "1  ...         0.512987           0.487013               0.373853   \n",
+       "0  ...              NaN                NaN                    NaN   \n",
+       "1  ...         0.696203           0.303797               0.359625   \n",
+       "0  ...              NaN                NaN                    NaN   \n",
+       "\n",
+       "   num_leaking_uniprot_train_val  num_leaking_smiles_train_val  \\\n",
+       "0                            NaN                           NaN   \n",
+       "1                           40.0                          70.0   \n",
+       "0                            NaN                           NaN   \n",
+       "1                            0.0                           8.0   \n",
+       "0                            NaN                           NaN   \n",
+       "\n",
+       "   perc_leaking_uniprot_train_val  perc_leaking_smiles_train_val  cv_models  \\\n",
+       "0                             NaN                            NaN        NaN   \n",
+       "1                        0.910859                       0.176661       True   \n",
+       "0                             NaN                            NaN        NaN   \n",
+       "1                        0.000000                       0.014430       True   \n",
+       "0                             NaN                            NaN        NaN   \n",
+       "\n",
+       "   model_type  split_type  \n",
+       "0     Pytorch      random  \n",
+       "1     Pytorch      random  \n",
+       "0     Pytorch     uniprot  \n",
+       "1     Pytorch     uniprot  \n",
+       "0     Pytorch    tanimoto  \n",
+       "\n",
+       "[5 rows x 28 columns]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "xgboost_cv_train: (15, 24)\n"
+     ]
+    },
+    {
+     "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>model_type</th>\n",
+       "      <th>fold</th>\n",
+       "      <th>train_len</th>\n",
+       "      <th>val_len</th>\n",
+       "      <th>train_perc</th>\n",
+       "      <th>val_perc</th>\n",
+       "      <th>train_active_perc</th>\n",
+       "      <th>train_inactive_perc</th>\n",
+       "      <th>train_avg_tanimoto_dist</th>\n",
+       "      <th>val_active_perc</th>\n",
+       "      <th>...</th>\n",
+       "      <th>perc_leaking_uniprot_train_val</th>\n",
+       "      <th>perc_leaking_smiles_train_val</th>\n",
+       "      <th>val_acc</th>\n",
+       "      <th>val_roc_auc</th>\n",
+       "      <th>val_precision</th>\n",
+       "      <th>val_recall</th>\n",
+       "      <th>val_f1_score</th>\n",
+       "      <th>split_type</th>\n",
+       "      <th>train_unique_groups</th>\n",
+       "      <th>val_unique_groups</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>0</td>\n",
+       "      <td>616</td>\n",
+       "      <td>155</td>\n",
+       "      <td>0.798962</td>\n",
+       "      <td>0.201038</td>\n",
+       "      <td>0.514610</td>\n",
+       "      <td>0.485390</td>\n",
+       "      <td>0.376475</td>\n",
+       "      <td>0.516129</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.886364</td>\n",
+       "      <td>0.146104</td>\n",
+       "      <td>0.838710</td>\n",
+       "      <td>0.917167</td>\n",
+       "      <td>0.816092</td>\n",
+       "      <td>0.887500</td>\n",
+       "      <td>0.850299</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>1</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.513776</td>\n",
+       "      <td>0.486224</td>\n",
+       "      <td>0.377372</td>\n",
+       "      <td>0.519481</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.920583</td>\n",
+       "      <td>0.188006</td>\n",
+       "      <td>0.883117</td>\n",
+       "      <td>0.959291</td>\n",
+       "      <td>0.860465</td>\n",
+       "      <td>0.925000</td>\n",
+       "      <td>0.891566</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>2</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.376572</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.914100</td>\n",
+       "      <td>0.144246</td>\n",
+       "      <td>0.876623</td>\n",
+       "      <td>0.917975</td>\n",
+       "      <td>0.894737</td>\n",
+       "      <td>0.860759</td>\n",
+       "      <td>0.877419</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>3</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.376068</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.876823</td>\n",
+       "      <td>0.162075</td>\n",
+       "      <td>0.883117</td>\n",
+       "      <td>0.937384</td>\n",
+       "      <td>0.858824</td>\n",
+       "      <td>0.924051</td>\n",
+       "      <td>0.890244</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>4</td>\n",
+       "      <td>617</td>\n",
+       "      <td>154</td>\n",
+       "      <td>0.800259</td>\n",
+       "      <td>0.199741</td>\n",
+       "      <td>0.515397</td>\n",
+       "      <td>0.484603</td>\n",
+       "      <td>0.377543</td>\n",
+       "      <td>0.512987</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.910859</td>\n",
+       "      <td>0.176661</td>\n",
+       "      <td>0.883117</td>\n",
+       "      <td>0.939241</td>\n",
+       "      <td>0.858824</td>\n",
+       "      <td>0.924051</td>\n",
+       "      <td>0.890244</td>\n",
+       "      <td>random</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 24 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  model_type  fold  train_len  val_len  train_perc  val_perc  \\\n",
+       "0    XGBoost     0        616      155    0.798962  0.201038   \n",
+       "1    XGBoost     1        617      154    0.800259  0.199741   \n",
+       "2    XGBoost     2        617      154    0.800259  0.199741   \n",
+       "3    XGBoost     3        617      154    0.800259  0.199741   \n",
+       "4    XGBoost     4        617      154    0.800259  0.199741   \n",
+       "\n",
+       "   train_active_perc  train_inactive_perc  train_avg_tanimoto_dist  \\\n",
+       "0           0.514610             0.485390                 0.376475   \n",
+       "1           0.513776             0.486224                 0.377372   \n",
+       "2           0.515397             0.484603                 0.376572   \n",
+       "3           0.515397             0.484603                 0.376068   \n",
+       "4           0.515397             0.484603                 0.377543   \n",
+       "\n",
+       "   val_active_perc  ...  perc_leaking_uniprot_train_val  \\\n",
+       "0         0.516129  ...                        0.886364   \n",
+       "1         0.519481  ...                        0.920583   \n",
+       "2         0.512987  ...                        0.914100   \n",
+       "3         0.512987  ...                        0.876823   \n",
+       "4         0.512987  ...                        0.910859   \n",
+       "\n",
+       "   perc_leaking_smiles_train_val   val_acc  val_roc_auc  val_precision  \\\n",
+       "0                       0.146104  0.838710     0.917167       0.816092   \n",
+       "1                       0.188006  0.883117     0.959291       0.860465   \n",
+       "2                       0.144246  0.876623     0.917975       0.894737   \n",
+       "3                       0.162075  0.883117     0.937384       0.858824   \n",
+       "4                       0.176661  0.883117     0.939241       0.858824   \n",
+       "\n",
+       "   val_recall  val_f1_score  split_type  train_unique_groups  \\\n",
+       "0    0.887500      0.850299      random                  NaN   \n",
+       "1    0.925000      0.891566      random                  NaN   \n",
+       "2    0.860759      0.877419      random                  NaN   \n",
+       "3    0.924051      0.890244      random                  NaN   \n",
+       "4    0.924051      0.890244      random                  NaN   \n",
+       "\n",
+       "   val_unique_groups  \n",
+       "0                NaN  \n",
+       "1                NaN  \n",
+       "2                NaN  \n",
+       "3                NaN  \n",
+       "4                NaN  \n",
+       "\n",
+       "[5 rows x 24 columns]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "xgboost_test: (9, 20)\n"
+     ]
+    },
+    {
+     "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>test_acc</th>\n",
+       "      <th>test_roc_auc</th>\n",
+       "      <th>test_precision</th>\n",
+       "      <th>test_recall</th>\n",
+       "      <th>test_f1_score</th>\n",
+       "      <th>model_type</th>\n",
+       "      <th>test_model_id</th>\n",
+       "      <th>train_len</th>\n",
+       "      <th>train_active_perc</th>\n",
+       "      <th>train_inactive_perc</th>\n",
+       "      <th>train_avg_tanimoto_dist</th>\n",
+       "      <th>test_len</th>\n",
+       "      <th>test_active_perc</th>\n",
+       "      <th>test_inactive_perc</th>\n",
+       "      <th>test_avg_tanimoto_dist</th>\n",
+       "      <th>num_leaking_uniprot_train_test</th>\n",
+       "      <th>num_leaking_smiles_train_test</th>\n",
+       "      <th>perc_leaking_uniprot_train_test</th>\n",
+       "      <th>perc_leaking_smiles_train_test</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.767442</td>\n",
+       "      <td>0.879348</td>\n",
+       "      <td>0.700000</td>\n",
+       "      <td>0.875000</td>\n",
+       "      <td>0.777778</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>0</td>\n",
+       "      <td>771</td>\n",
+       "      <td>0.514916</td>\n",
+       "      <td>0.485084</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.790698</td>\n",
+       "      <td>0.884239</td>\n",
+       "      <td>0.711538</td>\n",
+       "      <td>0.925000</td>\n",
+       "      <td>0.804348</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>1</td>\n",
+       "      <td>771</td>\n",
+       "      <td>0.514916</td>\n",
+       "      <td>0.485084</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0.767442</td>\n",
+       "      <td>0.882065</td>\n",
+       "      <td>0.692308</td>\n",
+       "      <td>0.900000</td>\n",
+       "      <td>0.782609</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>2</td>\n",
+       "      <td>771</td>\n",
+       "      <td>0.514916</td>\n",
+       "      <td>0.485084</td>\n",
+       "      <td>0.376806</td>\n",
+       "      <td>86</td>\n",
+       "      <td>0.465116</td>\n",
+       "      <td>0.534884</td>\n",
+       "      <td>0.381147</td>\n",
+       "      <td>34</td>\n",
+       "      <td>44</td>\n",
+       "      <td>0.832685</td>\n",
+       "      <td>0.102464</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.447059</td>\n",
+       "      <td>0.499443</td>\n",
+       "      <td>0.482759</td>\n",
+       "      <td>0.304348</td>\n",
+       "      <td>0.373333</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>0</td>\n",
+       "      <td>772</td>\n",
+       "      <td>0.506477</td>\n",
+       "      <td>0.493523</td>\n",
+       "      <td>0.375305</td>\n",
+       "      <td>85</td>\n",
+       "      <td>0.541176</td>\n",
+       "      <td>0.458824</td>\n",
+       "      <td>0.394830</td>\n",
+       "      <td>0</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.011658</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.517647</td>\n",
+       "      <td>0.486065</td>\n",
+       "      <td>0.567568</td>\n",
+       "      <td>0.456522</td>\n",
+       "      <td>0.506024</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>1</td>\n",
+       "      <td>772</td>\n",
+       "      <td>0.506477</td>\n",
+       "      <td>0.493523</td>\n",
+       "      <td>0.375305</td>\n",
+       "      <td>85</td>\n",
+       "      <td>0.541176</td>\n",
+       "      <td>0.458824</td>\n",
+       "      <td>0.394830</td>\n",
+       "      <td>0</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.011658</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   test_acc  test_roc_auc  test_precision  test_recall  test_f1_score  \\\n",
+       "0  0.767442      0.879348        0.700000     0.875000       0.777778   \n",
+       "1  0.790698      0.884239        0.711538     0.925000       0.804348   \n",
+       "2  0.767442      0.882065        0.692308     0.900000       0.782609   \n",
+       "0  0.447059      0.499443        0.482759     0.304348       0.373333   \n",
+       "1  0.517647      0.486065        0.567568     0.456522       0.506024   \n",
+       "\n",
+       "  model_type  test_model_id  train_len  train_active_perc  \\\n",
+       "0    XGBoost              0        771           0.514916   \n",
+       "1    XGBoost              1        771           0.514916   \n",
+       "2    XGBoost              2        771           0.514916   \n",
+       "0    XGBoost              0        772           0.506477   \n",
+       "1    XGBoost              1        772           0.506477   \n",
+       "\n",
+       "   train_inactive_perc  train_avg_tanimoto_dist  test_len  test_active_perc  \\\n",
+       "0             0.485084                 0.376806        86          0.465116   \n",
+       "1             0.485084                 0.376806        86          0.465116   \n",
+       "2             0.485084                 0.376806        86          0.465116   \n",
+       "0             0.493523                 0.375305        85          0.541176   \n",
+       "1             0.493523                 0.375305        85          0.541176   \n",
+       "\n",
+       "   test_inactive_perc  test_avg_tanimoto_dist  num_leaking_uniprot_train_test  \\\n",
+       "0            0.534884                0.381147                              34   \n",
+       "1            0.534884                0.381147                              34   \n",
+       "2            0.534884                0.381147                              34   \n",
+       "0            0.458824                0.394830                               0   \n",
+       "1            0.458824                0.394830                               0   \n",
+       "\n",
+       "   num_leaking_smiles_train_test  perc_leaking_uniprot_train_test  \\\n",
+       "0                             44                         0.832685   \n",
+       "1                             44                         0.832685   \n",
+       "2                             44                         0.832685   \n",
+       "0                              6                         0.000000   \n",
+       "1                              6                         0.000000   \n",
+       "\n",
+       "   perc_leaking_smiles_train_test split_type  \n",
+       "0                        0.102464     random  \n",
+       "1                        0.102464     random  \n",
+       "2                        0.102464     random  \n",
+       "0                        0.011658    uniprot  \n",
+       "1                        0.011658    uniprot  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "xgboost_hparam: (3, 7)\n"
+     ]
+    },
+    {
+     "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>eta</th>\n",
+       "      <th>max_depth</th>\n",
+       "      <th>min_child_weight</th>\n",
+       "      <th>gamma</th>\n",
+       "      <th>subsample</th>\n",
+       "      <th>colsample_bytree</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.098975</td>\n",
+       "      <td>9</td>\n",
+       "      <td>0.013528</td>\n",
+       "      <td>0.000218</td>\n",
+       "      <td>0.851506</td>\n",
+       "      <td>0.630858</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.025540</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0.051810</td>\n",
+       "      <td>0.007490</td>\n",
+       "      <td>0.514577</td>\n",
+       "      <td>0.883731</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.000391</td>\n",
+       "      <td>10</td>\n",
+       "      <td>0.027727</td>\n",
+       "      <td>0.013472</td>\n",
+       "      <td>0.649744</td>\n",
+       "      <td>0.511964</td>\n",
+       "      <td>tanimoto</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        eta  max_depth  min_child_weight     gamma  subsample  \\\n",
+       "0  0.098975          9          0.013528  0.000218   0.851506   \n",
+       "0  0.025540          3          0.051810  0.007490   0.514577   \n",
+       "0  0.000391         10          0.027727  0.013472   0.649744   \n",
+       "\n",
+       "   colsample_bytree split_type  \n",
+       "0          0.630858     random  \n",
+       "0          0.883731    uniprot  \n",
+       "0          0.511964   tanimoto  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "xgboost_majority_vote: (3, 7)\n"
+     ]
+    },
+    {
+     "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>test_acc</th>\n",
+       "      <th>test_roc_auc</th>\n",
+       "      <th>test_precision</th>\n",
+       "      <th>test_recall</th>\n",
+       "      <th>test_f1</th>\n",
+       "      <th>model_type</th>\n",
+       "      <th>split_type</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.779070</td>\n",
+       "      <td>0.880978</td>\n",
+       "      <td>0.723404</td>\n",
+       "      <td>0.850000</td>\n",
+       "      <td>0.781609</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>random</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.447059</td>\n",
+       "      <td>0.487179</td>\n",
+       "      <td>0.481481</td>\n",
+       "      <td>0.282609</td>\n",
+       "      <td>0.356164</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>uniprot</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.717647</td>\n",
+       "      <td>0.831081</td>\n",
+       "      <td>0.842105</td>\n",
+       "      <td>0.432432</td>\n",
+       "      <td>0.571429</td>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>tanimoto</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   test_acc  test_roc_auc  test_precision  test_recall   test_f1 model_type  \\\n",
+       "0  0.779070      0.880978        0.723404     0.850000  0.781609    XGBoost   \n",
+       "0  0.447059      0.487179        0.481481     0.282609  0.356164    XGBoost   \n",
+       "0  0.717647      0.831081        0.842105     0.432432  0.571429    XGBoost   \n",
+       "\n",
+       "  split_type  \n",
+       "0     random  \n",
+       "0    uniprot  \n",
+       "0   tanimoto  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "active_col = 'Active (Dmax 0.6, pDC50 6.0)'\n",
+    "test_split = 0.1\n",
+    "n_models_for_test = 3\n",
+    "cv_n_folds = 5\n",
+    "\n",
+    "active_name = active_col.replace(' ', '_').replace('(', '').replace(')', '').replace(',', '')\n",
+    "report_base_name = f'{active_name}_test_split_{test_split}'\n",
+    "\n",
+    "# Load the data\n",
+    "reports = {\n",
+    "    'cv_train': pd.concat([\n",
+    "        pd.read_csv(f'reports/cv_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/cv_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/cv_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'test': pd.concat([\n",
+    "        pd.read_csv(f'reports/test_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/test_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/test_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    # 'ablation': pd.concat([\n",
+    "    #     pd.read_csv(f'reports/ablation_report_{report_base_name}_random.csv'),\n",
+    "    #     pd.read_csv(f'reports/ablation_report_{report_base_name}_uniprot.csv'),\n",
+    "    #     pd.read_csv(f'reports/ablation_report_{report_base_name}_tanimoto.csv'),\n",
+    "    # ]),\n",
+    "    'hparam': pd.concat([\n",
+    "        pd.read_csv(f'reports/hparam_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/hparam_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/hparam_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'majority_vote': pd.concat([\n",
+    "        pd.read_csv(f'reports/majority_vote_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/majority_vote_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/majority_vote_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'xgboost_cv_train': pd.concat([\n",
+    "        pd.read_csv(f'reports/xgboost_cv_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_cv_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_cv_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'xgboost_test': pd.concat([\n",
+    "        pd.read_csv(f'reports/xgboost_test_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_test_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_test_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'xgboost_hparam': pd.concat([\n",
+    "        pd.read_csv(f'reports/xgboost_hparam_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_hparam_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_hparam_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "    'xgboost_majority_vote': pd.concat([\n",
+    "        pd.read_csv(f'reports/xgboost_majority_vote_report_{report_base_name}_random.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_majority_vote_report_{report_base_name}_uniprot.csv'),\n",
+    "        pd.read_csv(f'reports/xgboost_majority_vote_report_{report_base_name}_tanimoto.csv'),\n",
+    "    ]),\n",
+    "}\n",
+    "for k, report in reports.items():\n",
+    "    print(f'{k}: {report.shape}')\n",
+    "    display(report.head())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "|   fold | split_type   |   train_len |   val_len |   test_len |   train_active_perc |   val_active_perc |   test_active_perc | perc_leaking_uniprot_train_test   | perc_leaking_smiles_train_test   |   test_avg_tanimoto_dist |\n",
+      "|-------:|:-------------|------------:|----------:|-----------:|--------------------:|------------------:|-------------------:|:----------------------------------|:---------------------------------|-------------------------:|\n",
+      "|      0 | random       |         616 |       155 |         86 |            0.51461  |          0.516129 |           0.465116 | 82.5%                             | 11.2%                            |                    0.381 |\n",
+      "|      1 | random       |         617 |       154 |         86 |            0.513776 |          0.519481 |           0.465116 | 84.0%                             | 10.2%                            |                    0.381 |\n",
+      "|      2 | random       |         617 |       154 |         86 |            0.515397 |          0.512987 |           0.465116 | 83.8%                             | 9.4%                             |                    0.381 |\n",
+      "|      3 | random       |         617 |       154 |         86 |            0.515397 |          0.512987 |           0.465116 | 82.3%                             | 10.4%                            |                    0.381 |\n",
+      "|      4 | random       |         617 |       154 |         86 |            0.515397 |          0.512987 |           0.465116 | 83.8%                             | 10.0%                            |                    0.381 |\n",
+      "|      0 | uniprot      |         560 |       212 |         85 |            0.544643 |          0.40566  |           0.541176 | 0.0%                              | 1.1%                             |                    0.395 |\n",
+      "|      1 | uniprot      |         627 |       145 |         85 |            0.516746 |          0.462069 |           0.541176 | 0.0%                              | 0.8%                             |                    0.395 |\n",
+      "|      2 | uniprot      |         662 |       110 |         85 |            0.506042 |          0.509091 |           0.541176 | 0.0%                              | 1.2%                             |                    0.395 |\n",
+      "|      3 | uniprot      |         546 |       226 |         85 |            0.483516 |          0.561947 |           0.541176 | 0.0%                              | 1.5%                             |                    0.395 |\n",
+      "|      4 | uniprot      |         693 |        79 |         85 |            0.484848 |          0.696203 |           0.541176 | 0.0%                              | 1.3%                             |                    0.395 |\n",
+      "|      0 | tanimoto     |         660 |       112 |         85 |            0.515152 |          0.535714 |           0.435294 | 57.7%                             | 0.0%                             |                    0.42  |\n",
+      "|      1 | tanimoto     |         589 |       183 |         85 |            0.497453 |          0.584699 |           0.435294 | 56.4%                             | 0.0%                             |                    0.42  |\n",
+      "|      2 | tanimoto     |         616 |       156 |         85 |            0.542208 |          0.423077 |           0.435294 | 57.3%                             | 0.0%                             |                    0.42  |\n",
+      "|      3 | tanimoto     |         598 |       174 |         85 |            0.528428 |          0.482759 |           0.435294 | 56.5%                             | 0.0%                             |                    0.42  |\n",
+      "|      4 | tanimoto     |         625 |       147 |         85 |            0.5072   |          0.564626 |           0.435294 | 57.0%                             | 0.0%                             |                    0.42  |\n"
+     ]
+    }
+   ],
+   "source": [
+    "cols_to_show = [\n",
+    "    'fold',\n",
+    "    'split_type',\n",
+    "    'train_len',\n",
+    "    'val_len',\n",
+    "    'test_len',\n",
+    "    'train_active_perc',\n",
+    "    'val_active_perc',\n",
+    "    'test_active_perc',\n",
+    "    # 'train_unique_groups',\n",
+    "    # 'val_unique_groups',\n",
+    "    'perc_leaking_uniprot_train_test',\n",
+    "    'perc_leaking_smiles_train_test',\n",
+    "    'test_avg_tanimoto_dist',\n",
+    "]\n",
+    "# print(reports['cv_train'][cols_to_show].to_markdown(index=False))\n",
+    "# Print a subset of columns (that contain the string \"perc_\") as percentages in format: .1%\n",
+    "tmp = reports['cv_train'][cols_to_show].copy()\n",
+    "for col in tmp.columns:\n",
+    "    if 'perc_' in col:\n",
+    "        tmp[col] = tmp[col].apply(lambda x: f'{x:.1%}')\n",
+    "    if 'dist' in col:\n",
+    "        tmp[col] = tmp[col].apply(lambda x: f'{x:.3f}')\n",
+    "print(tmp[cols_to_show].to_markdown(index=False))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1400x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "def plot_performance_metrics(df_cv, df_test, df_test_majority=None, title=None, show_plot=False):\n",
+    "    # Extract and prepare CV data\n",
+    "    cv_data = df_cv[['model_type', 'fold', 'split_type', 'val_acc', 'val_roc_auc']] #, 'test_acc', 'test_roc_auc']]\n",
+    "    cv_data = cv_data.melt(id_vars=['model_type', 'fold', 'split_type'], var_name='Metric', value_name='Score')\n",
+    "    cv_data['Metric'] = cv_data['Metric'].replace({\n",
+    "        'val_acc': 'Validation Accuracy',\n",
+    "        'val_roc_auc': 'Validation ROC AUC',\n",
+    "        'test_acc': 'Test Accuracy',\n",
+    "        'test_roc_auc': 'Test ROC AUC'\n",
+    "    })\n",
+    "    cv_data['Stage'] = cv_data['Metric'].apply(lambda x: 'Validation' if 'Val' in x else 'Test')\n",
+    "    # Remove test data from CV data\n",
+    "    cv_data = cv_data[cv_data['Stage'] == 'Validation']\n",
+    "\n",
+    "    # Extract and prepare test data\n",
+    "    test_data = df_test[['model_type', 'test_acc', 'test_roc_auc', 'split_type']]\n",
+    "    test_data = test_data.melt(id_vars=['model_type', 'split_type'], var_name='Metric', value_name='Score')\n",
+    "    test_data['Metric'] = test_data['Metric'].replace({\n",
+    "        'test_acc': 'Test Accuracy\\n(Average Score)',\n",
+    "        'test_roc_auc': 'Test ROC AUC\\n(Average Score)'\n",
+    "    })\n",
+    "    test_data['Stage'] = 'Test'\n",
+    "\n",
+    "    # Combine CV and test data\n",
+    "    combined_data = pd.concat([cv_data, test_data], ignore_index=True)\n",
+    "\n",
+    "    if df_test_majority is not None:\n",
+    "        # Extract and prepare test data\n",
+    "        test_data_majority = df_test_majority[['model_type', 'test_acc', 'test_roc_auc', 'split_type']]\n",
+    "        test_data_majority = test_data_majority.melt(id_vars=['model_type', 'split_type'], var_name='Metric', value_name='Score')\n",
+    "        test_data_majority['Metric'] = test_data_majority['Metric'].replace({\n",
+    "            'test_acc': 'Test Accuracy\\n(Majority Vote)',\n",
+    "            'test_roc_auc': 'Test ROC AUC\\n(Majority Vote)'\n",
+    "        })\n",
+    "        test_data_majority['Stage'] = 'Test (Majority Vote)'\n",
+    "        combined_data = pd.concat([combined_data, test_data_majority], ignore_index=True)\n",
+    "\n",
+    "    # Rename 'split_type' values according to a predefined map for clarity\n",
+    "    group2name = {\n",
+    "        'random': 'Standard Split',\n",
+    "        'uniprot': 'Target Split',\n",
+    "        'tanimoto': 'Similarity Split',\n",
+    "    }\n",
+    "    combined_data['Split Type'] = combined_data['split_type'].map(group2name)\n",
+    "\n",
+    "    # Add dummy model data\n",
+    "    dummy_val_acc = []\n",
+    "    dummy_test_acc = []\n",
+    "    for i, group in enumerate(group2name.keys()):\n",
+    "        # Get the majority class in group_df\n",
+    "        group_df = df_cv[df_cv['split_type'] == group]\n",
+    "        major_col = 'inactive' if group_df['val_inactive_perc'].mean() > 0.5 else 'active'\n",
+    "        dummy_val_acc.append(group_df[f'val_{major_col}_perc'].mean())\n",
+    "\n",
+    "        group_df = df_test[df_test['split_type'] == group]\n",
+    "        major_col = 'inactive' if group_df['test_inactive_perc'].mean() > 0.5 else 'active'\n",
+    "        dummy_test_acc.append(group_df[f'test_{major_col}_perc'].mean())\n",
+    "\n",
+    "    dummy_scores = []\n",
+    "    for i in range(len(dummy_val_acc)):\n",
+    "        metrics = {\n",
+    "            'Validation Accuracy': dummy_val_acc[i],\n",
+    "            'Validation ROC AUC': 0.5,\n",
+    "            'Test Accuracy\\n(Average Score)': dummy_test_acc[i],\n",
+    "            'Test ROC AUC\\n(Average Score)': 0.5,\n",
+    "        }\n",
+    "        if df_test_majority is not None:\n",
+    "            metrics['Test Accuracy\\n(Majority Vote)'] = dummy_test_acc[i]\n",
+    "            metrics['Test ROC AUC\\n(Majority Vote)'] = 0.5\n",
+    "        for metric, score in metrics.items():\n",
+    "            dummy_scores.append({\n",
+    "                'Experiment': i,\n",
+    "                'Metric': metric,\n",
+    "                'Score': score,\n",
+    "                'Split Type': 'Dummy model',\n",
+    "            })\n",
+    "    dummy_model = pd.DataFrame(dummy_scores)\n",
+    "    combined_data = pd.concat([combined_data, dummy_model], ignore_index=True)\n",
+    "\n",
+    "    # Plotting\n",
+    "    plt.figure(figsize=(12, 6))\n",
+    "    sns.barplot(\n",
+    "        data=combined_data,\n",
+    "        x='Metric',\n",
+    "        y='Score',\n",
+    "        hue='Split Type',\n",
+    "        errorbar=('sd', 1),\n",
+    "        palette=palette)\n",
+    "    plt.title('')\n",
+    "    plt.ylabel('')\n",
+    "    plt.xlabel('')\n",
+    "    plt.ylim(0, 1.0)  # Assuming scores are normalized between 0 and 1\n",
+    "    plt.grid(axis='y', alpha=0.5, linewidth=0.5)\n",
+    "\n",
+    "    # Make the y-axis as percentage\n",
+    "    plt.gca().yaxis.set_major_formatter(plt.matplotlib.ticker.PercentFormatter(1, decimals=0))\n",
+    "    # Plot the legend below the x-axis, outside the plot, and divided in two columns\n",
+    "    plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.08), ncol=4)\n",
+    "\n",
+    "    # For each bar, add the rotated value (as percentage), inside the bar\n",
+    "    for i, p in enumerate(plt.gca().patches):\n",
+    "        # TODO: For some reasons, there are 4 additional rectangles being\n",
+    "        # plotted... I suspect it's because the dummy_df doesn't have the same\n",
+    "        # shape as the df containing all the evaluation data...\n",
+    "        if p.get_height() < 0.01:\n",
+    "            continue\n",
+    "        if i % 2 == 0:\n",
+    "            value = f'{p.get_height():.1%}'\n",
+    "        else:\n",
+    "            value = f'{p.get_height():.3f}'\n",
+    "        \n",
+    "        # print(f'Plotting value: {p.get_height():.5f} -> {value}')\n",
+    "        x = p.get_x() + p.get_width() / 2\n",
+    "        y = 0.4 # p.get_height() - p.get_height() / 2\n",
+    "        plt.annotate(value, (x, y), ha='center', va='center', color='black', fontsize=10, rotation=90, alpha=0.8)\n",
+    "\n",
+    "    # plt.savefig(f'plots/{title}.pdf', bbox_inches='tight')\n",
+    "    if show_plot:\n",
+    "        plt.show()\n",
+    "\n",
+    "    # Plot in the same above the accuracy and the ROC AUC in two different subplots\n",
+    "    fig, axes = plt.subplots(1, 2, figsize=(14, 4))\n",
+    "    sns.barplot(\n",
+    "        data=combined_data[combined_data['Metric'].str.contains('Accuracy')],\n",
+    "        x='Metric',\n",
+    "        y='Score',\n",
+    "        hue='Split Type',\n",
+    "        errorbar=('sd', 1),\n",
+    "        palette=palette,\n",
+    "        ax=axes[0])\n",
+    "    sns.barplot(\n",
+    "        data=combined_data[combined_data['Metric'].str.contains('ROC AUC')],\n",
+    "        x='Metric',\n",
+    "        y='Score',\n",
+    "        hue='Split Type',\n",
+    "        errorbar=('sd', 1),\n",
+    "        palette=palette,\n",
+    "        ax=axes[1])\n",
+    "    # axes[0].set_title('Accuracy')\n",
+    "    axes[0].set_ylabel('')\n",
+    "    axes[0].set_xlabel('')\n",
+    "    axes[0].set_ylim(0, 1.0)\n",
+    "    axes[0].grid(axis='y', alpha=0.5, linewidth=0.5)\n",
+    "    axes[0].yaxis.set_major_formatter(plt.matplotlib.ticker.PercentFormatter(1, decimals=0))\n",
+    "    axes[0].legend().remove()\n",
+    "\n",
+    "    # axes[1].set_title('ROC AUC')\n",
+    "    axes[1].set_ylabel('')\n",
+    "    axes[1].set_xlabel('')\n",
+    "    axes[1].set_ylim(0, 1.0)\n",
+    "    axes[1].grid(axis='y', alpha=0.5, linewidth=0.5)\n",
+    "    # axes[1].yaxis.set_major_formatter(plt.matplotlib.ticker.PercentFormatter(1, decimals=0))\n",
+    "    axes[1].legend().remove()\n",
+    "\n",
+    "    # For each bar in both subplots, add the rotated value (as percentage), inside the bar\n",
+    "    for i, ax in enumerate(axes):\n",
+    "        for p in ax.patches:\n",
+    "            if p.get_height() < 0.01:\n",
+    "                continue\n",
+    "            if i % 2 == 0:\n",
+    "                value = f'{p.get_height():.1%}'\n",
+    "            else:\n",
+    "                value = f'{p.get_height():.3f}'\n",
+    "            \n",
+    "            x = p.get_x() + p.get_width() / 2\n",
+    "            y = 0.3 # p.get_height() - p.get_height() / 2\n",
+    "            ax.annotate(value, (x, y), ha='center', va='center', color='black', fontsize=10, rotation=90, alpha=0.8)\n",
+    "\n",
+    "    # # For each bar, add the rotated value (as percentage), inside the bar\n",
+    "    # for i, p in enumerate(plt.gca().patches):\n",
+    "    #     # TODO: For some reasons, there are 4 additional rectangles being\n",
+    "    #     # plotted... I suspect it's because the dummy_df doesn't have the same\n",
+    "    #     # shape as the df containing all the evaluation data...\n",
+    "    #     if p.get_height() < 0.01:\n",
+    "    #         continue\n",
+    "    #     if i % 2 == 0:\n",
+    "    #         value = f'{p.get_height():.1%}'\n",
+    "    #     else:\n",
+    "    #         value = f'{p.get_height():.3f}'\n",
+    "        \n",
+    "    #     # print(f'Plotting value: {p.get_height():.5f} -> {value}')\n",
+    "    #     x = p.get_x() + p.get_width() / 2\n",
+    "    #     y = 0.4 # p.get_height() - p.get_height() / 2\n",
+    "    #     plt.annotate(value, (x, y), ha='center', va='center', color='black', fontsize=10, rotation=90, alpha=0.8)\n",
+    "\n",
+    "    plt.legend(loc='upper center', bbox_to_anchor=(-0.1, -0.15), ncol=4)\n",
+    "    plt.savefig(f'plots/{title}.pdf', bbox_inches='tight')\n",
+    "    if show_plot:\n",
+    "        plt.show()\n",
+    "\n",
+    "plot_performance_metrics(\n",
+    "    df_cv=reports['cv_train'],\n",
+    "    df_test=reports['test'],\n",
+    "    df_test_majority=reports['majority_vote'][reports['majority_vote']['cv_models'].isna()],\n",
+    "    title=f'summary_performance-best_models_as_test',\n",
+    "    show_plot=True,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1400x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plot_performance_metrics(\n",
+    "    df_cv=reports['xgboost_cv_train'],\n",
+    "    df_test=reports['xgboost_test'],\n",
+    "    df_test_majority=reports['xgboost_majority_vote'],\n",
+    "    title=f'xgboost_summary_performance-best_models_as_test',\n",
+    "    show_plot=True,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Plotting value: 0.85734 -> 85.7%\n",
+      "Plotting value: 0.92236 -> 0.922\n",
+      "Plotting value: 0.82558 -> 82.6%\n",
+      "Plotting value: 0.84783 -> 0.848\n",
+      "Plotting value: 0.70521 -> 70.5%\n",
+      "Plotting value: 0.74610 -> 0.746\n",
+      "Plotting value: 0.61176 -> 61.2%\n",
+      "Plotting value: 0.61483 -> 0.615\n",
+      "Plotting value: 0.79090 -> 79.1%\n",
+      "Plotting value: 0.86691 -> 0.867\n",
+      "Plotting value: 0.70588 -> 70.6%\n",
+      "Plotting value: 0.82376 -> 0.824\n",
+      "Plotting value: 0.52003 -> 52.0%\n",
+      "Plotting value: 0.50000 -> 0.500\n",
+      "Plotting value: 0.54692 -> 54.7%\n",
+      "Plotting value: 0.50000 -> 0.500\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plot_performance_metrics(\n",
+    "    df_cv=reports['cv_train'],\n",
+    "    df_test=reports['majority_vote'][reports['majority_vote']['cv_models'].isna()],\n",
+    "    title=f'majority_vote_performance-best_models_as_test',\n",
+    "    show_plot=True,\n",
+    ")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}