From 3cef3637ca64bd2cf4ef777b1297f929a0baf9ea Mon Sep 17 00:00:00 2001
From: HarrisonWilde <harrisondwilde@outlook.com>
Date: Tue, 17 Oct 2023 14:32:12 +0100
Subject: [PATCH] Apply black formatting to notebooks

---
 auxiliary/MIMIC_preproc.ipynb    | 223 +++++++++++++++++++------------
 auxiliary/generate_SUPPORT.ipynb |   9 +-
 2 files changed, 140 insertions(+), 92 deletions(-)

diff --git a/auxiliary/MIMIC_preproc.ipynb b/auxiliary/MIMIC_preproc.ipynb
index 3d8db638..9125f1d5 100644
--- a/auxiliary/MIMIC_preproc.ipynb
+++ b/auxiliary/MIMIC_preproc.ipynb
@@ -60,8 +60,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "mimic_path = \"../data/\" # Path containing MIMIC data\n",
-    "output_path = \"../data/\" # Path to save the created datasets\n",
+    "mimic_path = \"../data/\"  # Path containing MIMIC data\n",
+    "output_path = \"../data/\"  # Path to save the created datasets\n",
     "\n",
     "admissions = pd.read_csv(mimic_path + \"ADMISSIONS.csv\")\n",
     "chartevents = pd.read_csv(mimic_path + \"CHARTEVENTS.csv\", nrows=10000000)\n",
@@ -113,27 +113,20 @@
     "        < pd.to_datetime(\"1930-01-01 00:00:00\", format=\"%Y-%m-%d %H:%M:%S\")\n",
     "    ):\n",
     "        num_years = randrange(years_diff_behind - 80, years_diff_behind - 40)\n",
-    "        new_dobs.append(\n",
-    "            pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
-    "            + pd.DateOffset(years=num_years)\n",
-    "        )\n",
+    "        new_dobs.append(pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\") + pd.DateOffset(years=num_years))\n",
     "        dob_offset.append(num_years)\n",
     "    elif (years_diff_ahead != 0) and (\n",
     "        pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
     "        > pd.to_datetime(\"2021-12-01 00:00:00\", format=\"%Y-%m-%d %H:%M:%S\")\n",
     "    ):\n",
     "        num_years = randrange(years_diff_ahead + 30, years_diff_ahead + 50)\n",
-    "        new_dobs.append(\n",
-    "            pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
-    "            - pd.DateOffset(years=num_years)\n",
-    "        )\n",
+    "        new_dobs.append(pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\") - pd.DateOffset(years=num_years))\n",
     "        dob_offset.append(-num_years)\n",
     "    else:\n",
     "        new_dobs.append(pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\"))\n",
     "        dob_offset.append(0)\n",
     "patients[\"DOB\"] = new_dobs\n",
-    "patients[\"DOB_offset\"] = dob_offset\n",
-    "\n"
+    "patients[\"DOB_offset\"] = dob_offset"
    ]
   },
   {
@@ -163,39 +156,57 @@
    "source": [
     "seed(2021)\n",
     "# Generate small input data file\n",
-    "mimic_table = admissions[['SUBJECT_ID','ETHNICITY','ADMITTIME','DISCHTIME','DISCHARGE_LOCATION']]\n",
-    "mimic_table = mimic_table.merge(patients[['SUBJECT_ID','GENDER','DOB']])\n",
-    "mimic_table = mimic_table.merge(icustays[['SUBJECT_ID','ICUSTAY_ID','FIRST_CAREUNIT']])\n",
-    "one_per_pat = chartevents.drop_duplicates(subset=['SUBJECT_ID','ICUSTAY_ID'])\n",
-    "mimic_table = mimic_table.merge(one_per_pat[['SUBJECT_ID','ICUSTAY_ID','CHARTTIME','ITEMID','VALUE','VALUEUOM']],on=['SUBJECT_ID','ICUSTAY_ID'])\n",
+    "mimic_table = admissions[[\"SUBJECT_ID\", \"ETHNICITY\", \"ADMITTIME\", \"DISCHTIME\", \"DISCHARGE_LOCATION\"]]\n",
+    "mimic_table = mimic_table.merge(patients[[\"SUBJECT_ID\", \"GENDER\", \"DOB\"]])\n",
+    "mimic_table = mimic_table.merge(icustays[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"FIRST_CAREUNIT\"]])\n",
+    "one_per_pat = chartevents.drop_duplicates(subset=[\"SUBJECT_ID\", \"ICUSTAY_ID\"])\n",
+    "mimic_table = mimic_table.merge(\n",
+    "    one_per_pat[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"CHARTTIME\", \"ITEMID\", \"VALUE\", \"VALUEUOM\"]],\n",
+    "    on=[\"SUBJECT_ID\", \"ICUSTAY_ID\"],\n",
+    ")\n",
     "\n",
     "new_admits = []\n",
     "new_dischs = []\n",
     "new_chart = []\n",
     "\n",
     "for index, row in tqdm(mimic_table.iterrows(), total=mimic_table.shape[0]):\n",
-    "    admit_min = len(pd.date_range(start=pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime('2021-12-01 00:00:00',format='%Y-%m-%d %H:%M:%S'),freq='D'))\n",
-    "    stay_len = len(pd.date_range(start=pd.to_datetime(row['ADMITTIME'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime(row['DISCHTIME'],format='%Y-%m-%d %H:%M:%S'),freq='S'))\n",
+    "    admit_min = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(\"2021-12-01 00:00:00\", format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"D\",\n",
+    "        )\n",
+    "    )\n",
+    "    stay_len = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"ADMITTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(row[\"DISCHTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"S\",\n",
+    "        )\n",
+    "    )\n",
     "\n",
-    "    num_days_admit = randrange(np.round(admit_min*0.25).astype(int),np.round(admit_min*0.9).astype(int)+5)\n",
-    "    num_days_disch = randrange(0,50)\n",
-    "    num_secs_chart = randrange(np.round(stay_len*0.01).astype(int)+1,np.round(stay_len*0.99).astype(int)+10)\n",
-    "    new_admit_date = pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S') + pd.DateOffset(days=num_days_admit)\n",
+    "    num_days_admit = randrange(np.round(admit_min * 0.25).astype(int), np.round(admit_min * 0.9).astype(int) + 5)\n",
+    "    num_days_disch = randrange(0, 50)\n",
+    "    num_secs_chart = randrange(np.round(stay_len * 0.01).astype(int) + 1, np.round(stay_len * 0.99).astype(int) + 10)\n",
+    "    new_admit_date = pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\") + pd.DateOffset(days=num_days_admit)\n",
     "    new_admits.append(new_admit_date)\n",
     "    new_dischs.append(new_admit_date + pd.DateOffset(days=num_days_disch))\n",
     "    new_chart.append(new_admit_date + pd.DateOffset(seconds=num_secs_chart))\n",
     "\n",
-    "mimic_table['ADMITTIME'] = new_admits\n",
-    "mimic_table['DISCHTIME'] = new_dischs\n",
-    "mimic_table['CHARTTIME'] = new_chart\n",
-    "\n",
-    "mimic_table = mimic_table[(pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME)) & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))]\n",
-    "mimic_table = mimic_table.merge(items[['ITEMID','LABEL']],on=['ITEMID'])\n",
-    "mimic_table.drop(['ICUSTAY_ID','ITEMID'],axis=1,inplace=True)\n",
-    "age_calc = pd.Timestamp('2021-12-01 00:00:00')\n",
-    "mimic_table['DOB'] = pd.to_datetime(mimic_table['DOB'], format='%Y-%m-%d %H:%M:%S')\n",
-    "mimic_table['age'] = (age_calc - mimic_table['DOB']).astype('<m8[Y]')\n",
-    "mimic_table = mimic_table.groupby('SUBJECT_ID').head(4)\n",
+    "mimic_table[\"ADMITTIME\"] = new_admits\n",
+    "mimic_table[\"DISCHTIME\"] = new_dischs\n",
+    "mimic_table[\"CHARTTIME\"] = new_chart\n",
+    "\n",
+    "mimic_table = mimic_table[\n",
+    "    (pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "    & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "]\n",
+    "mimic_table = mimic_table.merge(items[[\"ITEMID\", \"LABEL\"]], on=[\"ITEMID\"])\n",
+    "mimic_table.drop([\"ICUSTAY_ID\", \"ITEMID\"], axis=1, inplace=True)\n",
+    "age_calc = pd.Timestamp(\"2021-12-01 00:00:00\")\n",
+    "mimic_table[\"DOB\"] = pd.to_datetime(mimic_table[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
+    "mimic_table[\"age\"] = (age_calc - mimic_table[\"DOB\"]).astype(\"<m8[Y]\")\n",
+    "mimic_table = mimic_table.groupby(\"SUBJECT_ID\").head(4)\n",
     "mimic_table.to_csv(output_path + \"MIMIC_{}_{}.csv\".format(mimic_table.shape[0], time.strftime(\"%Y%m%d\")), index=False)\n",
     "print(f\"Small input table saved, number of columns:  {mimic_table.shape[1]}, number of rows: {mimic_table.shape[0]}\")"
    ]
@@ -239,65 +250,85 @@
    "source": [
     "seed(2021)\n",
     "# Generate mid-sized input file\n",
-    "mimic_table = admissions[['SUBJECT_ID','ETHNICITY','ADMITTIME','DISCHTIME','DISCHARGE_LOCATION']]\n",
-    "mimic_table = mimic_table.merge(patients[['SUBJECT_ID','GENDER','DOB']])\n",
-    "mimic_table = mimic_table.merge(icustays[['SUBJECT_ID','ICUSTAY_ID','FIRST_CAREUNIT']])\n",
+    "mimic_table = admissions[[\"SUBJECT_ID\", \"ETHNICITY\", \"ADMITTIME\", \"DISCHTIME\", \"DISCHARGE_LOCATION\"]]\n",
+    "mimic_table = mimic_table.merge(patients[[\"SUBJECT_ID\", \"GENDER\", \"DOB\"]])\n",
+    "mimic_table = mimic_table.merge(icustays[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"FIRST_CAREUNIT\"]])\n",
     "\n",
     "total_subjects = list(chartevents.SUBJECT_ID.unique())\n",
     "\n",
     "df_list = []\n",
     "\n",
-    "split_one = int(np.round(len(total_subjects)*0.3).astype(int))\n",
-    "split_two = int(np.round(len(total_subjects)*0.9).astype(int))\n",
+    "split_one = int(np.round(len(total_subjects) * 0.3).astype(int))\n",
+    "split_two = int(np.round(len(total_subjects) * 0.9).astype(int))\n",
     "\n",
-    "for sub_index in range(0,split_one):\n",
+    "for sub_index in range(0, split_one):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(1))\n",
     "\n",
-    "for sub_index in range(split_one,split_two):\n",
+    "for sub_index in range(split_one, split_two):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(2))\n",
     "\n",
-    "for sub_index in range(split_two,len(total_subjects)):\n",
+    "for sub_index in range(split_two, len(total_subjects)):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(100))\n",
     "\n",
     "one_per_pat = pd.concat(df_list)\n",
     "one_per_pat.shape\n",
     "\n",
-    "mimic_table = mimic_table.merge(one_per_pat[['SUBJECT_ID','ICUSTAY_ID','CHARTTIME','ITEMID','VALUE','VALUEUOM']],on=['SUBJECT_ID','ICUSTAY_ID'])\n",
+    "mimic_table = mimic_table.merge(\n",
+    "    one_per_pat[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"CHARTTIME\", \"ITEMID\", \"VALUE\", \"VALUEUOM\"]],\n",
+    "    on=[\"SUBJECT_ID\", \"ICUSTAY_ID\"],\n",
+    ")\n",
     "\n",
     "new_admits = []\n",
     "new_dischs = []\n",
     "new_chart = []\n",
     "\n",
-    "patient_stays = mimic_table[['SUBJECT_ID','ICUSTAY_ID','DOB','ADMITTIME']]\n",
+    "patient_stays = mimic_table[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"DOB\", \"ADMITTIME\"]]\n",
     "patient_stays.drop_duplicates(inplace=True)\n",
     "\n",
     "for index, row in tqdm(mimic_table.iterrows(), total=mimic_table.shape[0]):\n",
-    "    admit_min = len(pd.date_range(start=pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime('2021-12-01 00:00:00',format='%Y-%m-%d %H:%M:%S'),freq='D'))\n",
-    "    stay_len = len(pd.date_range(start=pd.to_datetime(row['ADMITTIME'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime(row['DISCHTIME'],format='%Y-%m-%d %H:%M:%S'),freq='S'))\n",
+    "    admit_min = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(\"2021-12-01 00:00:00\", format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"D\",\n",
+    "        )\n",
+    "    )\n",
+    "    stay_len = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"ADMITTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(row[\"DISCHTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"S\",\n",
+    "        )\n",
+    "    )\n",
     "\n",
-    "    num_days_admit = randrange(np.round(admit_min*0.25).astype(int),np.round(admit_min*0.9).astype(int)+5)\n",
-    "    num_days_disch = randrange(0,50)\n",
-    "    num_secs_chart = randrange(np.round(stay_len*0.01).astype(int)+1,np.round(stay_len*0.99).astype(int)+10)\n",
-    "    new_admit_date = pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S') + pd.DateOffset(days=num_days_admit)\n",
+    "    num_days_admit = randrange(np.round(admit_min * 0.25).astype(int), np.round(admit_min * 0.9).astype(int) + 5)\n",
+    "    num_days_disch = randrange(0, 50)\n",
+    "    num_secs_chart = randrange(np.round(stay_len * 0.01).astype(int) + 1, np.round(stay_len * 0.99).astype(int) + 10)\n",
+    "    new_admit_date = pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\") + pd.DateOffset(days=num_days_admit)\n",
     "    new_admits.append(new_admit_date)\n",
     "    new_dischs.append(new_admit_date + pd.DateOffset(days=num_days_disch))\n",
     "    new_chart.append(new_admit_date + pd.DateOffset(seconds=num_secs_chart))\n",
     "\n",
-    "mimic_table['ADMITTIME'] = new_admits\n",
-    "mimic_table['DISCHTIME'] = new_dischs\n",
-    "mimic_table['CHARTTIME'] = new_chart\n",
-    "\n",
-    "mimic_table = mimic_table[(pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME)) & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))]\n",
-    "mimic_table = mimic_table.merge(items[['ITEMID','LABEL']],on=['ITEMID'])\n",
-    "mimic_table.drop(['ICUSTAY_ID','ITEMID'],axis=1,inplace=True)\n",
-    "age_calc = pd.Timestamp('2021-12-01 00:00:00')\n",
-    "mimic_table['DOB'] = pd.to_datetime(mimic_table['DOB'], format='%Y-%m-%d %H:%M:%S')\n",
-    "mimic_table['age'] = (age_calc - mimic_table['DOB']).astype('<m8[Y]')\n",
+    "mimic_table[\"ADMITTIME\"] = new_admits\n",
+    "mimic_table[\"DISCHTIME\"] = new_dischs\n",
+    "mimic_table[\"CHARTTIME\"] = new_chart\n",
+    "\n",
+    "mimic_table = mimic_table[\n",
+    "    (pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "    & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "]\n",
+    "mimic_table = mimic_table.merge(items[[\"ITEMID\", \"LABEL\"]], on=[\"ITEMID\"])\n",
+    "mimic_table.drop([\"ICUSTAY_ID\", \"ITEMID\"], axis=1, inplace=True)\n",
+    "age_calc = pd.Timestamp(\"2021-12-01 00:00:00\")\n",
+    "mimic_table[\"DOB\"] = pd.to_datetime(mimic_table[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
+    "mimic_table[\"age\"] = (age_calc - mimic_table[\"DOB\"]).astype(\"<m8[Y]\")\n",
     "mimic_table.to_csv(output_path + \"MIMIC_{}_{}.csv\".format(mimic_table.shape[0], time.strftime(\"%Y%m%d\")), index=False)\n",
-    "print(f\"Mid sized input table saved, number of columns:  {mimic_table.shape[1]}, number of rows: {mimic_table.shape[0]}\")"
+    "print(\n",
+    "    f\"Mid sized input table saved, number of columns:  {mimic_table.shape[1]}, number of rows: {mimic_table.shape[0]}\"\n",
+    ")"
    ]
   },
   {
@@ -319,63 +350,81 @@
    "source": [
     "seed(2021)\n",
     "# Generate large-sized input file\n",
-    "mimic_table = admissions[['SUBJECT_ID','ETHNICITY','ADMITTIME','DISCHTIME','DISCHARGE_LOCATION']]\n",
-    "mimic_table = mimic_table.merge(patients[['SUBJECT_ID','GENDER','DOB']])\n",
-    "mimic_table = mimic_table.merge(icustays[['SUBJECT_ID','ICUSTAY_ID','FIRST_CAREUNIT']])\n",
+    "mimic_table = admissions[[\"SUBJECT_ID\", \"ETHNICITY\", \"ADMITTIME\", \"DISCHTIME\", \"DISCHARGE_LOCATION\"]]\n",
+    "mimic_table = mimic_table.merge(patients[[\"SUBJECT_ID\", \"GENDER\", \"DOB\"]])\n",
+    "mimic_table = mimic_table.merge(icustays[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"FIRST_CAREUNIT\"]])\n",
     "\n",
     "total_subjects = list(chartevents.SUBJECT_ID.unique())\n",
     "\n",
     "df_list = []\n",
     "\n",
-    "split_one = int(np.round(len(total_subjects)*0.3).astype(int))\n",
-    "split_two = int(np.round(len(total_subjects)*0.7).astype(int))\n",
+    "split_one = int(np.round(len(total_subjects) * 0.3).astype(int))\n",
+    "split_two = int(np.round(len(total_subjects) * 0.7).astype(int))\n",
     "\n",
-    "for sub_index in range(0,split_one):\n",
+    "for sub_index in range(0, split_one):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(1))\n",
     "\n",
-    "for sub_index in range(split_one,split_two):\n",
+    "for sub_index in range(split_one, split_two):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(2))\n",
     "\n",
-    "for sub_index in range(split_two,len(total_subjects)):\n",
+    "for sub_index in range(split_two, len(total_subjects)):\n",
     "    search_id = total_subjects[sub_index]\n",
     "    df_list.append(chartevents[chartevents.SUBJECT_ID == int(search_id)].head(100))\n",
     "\n",
     "one_per_pat = pd.concat(df_list)\n",
     "one_per_pat.shape\n",
     "\n",
-    "mimic_table = mimic_table.merge(one_per_pat[['SUBJECT_ID','ICUSTAY_ID','CHARTTIME','ITEMID','VALUE','VALUEUOM']],on=['SUBJECT_ID','ICUSTAY_ID'])\n",
+    "mimic_table = mimic_table.merge(\n",
+    "    one_per_pat[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"CHARTTIME\", \"ITEMID\", \"VALUE\", \"VALUEUOM\"]],\n",
+    "    on=[\"SUBJECT_ID\", \"ICUSTAY_ID\"],\n",
+    ")\n",
     "\n",
     "new_admits = []\n",
     "new_dischs = []\n",
     "new_chart = []\n",
     "\n",
-    "patient_stays = mimic_table[['SUBJECT_ID','ICUSTAY_ID','DOB','ADMITTIME']]\n",
+    "patient_stays = mimic_table[[\"SUBJECT_ID\", \"ICUSTAY_ID\", \"DOB\", \"ADMITTIME\"]]\n",
     "patient_stays.drop_duplicates(inplace=True)\n",
     "\n",
     "for index, row in tqdm(mimic_table.iterrows(), total=mimic_table.shape[0]):\n",
-    "    admit_min = len(pd.date_range(start=pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime('2021-12-01 00:00:00',format='%Y-%m-%d %H:%M:%S'),freq='D'))\n",
-    "    stay_len = len(pd.date_range(start=pd.to_datetime(row['ADMITTIME'],format='%Y-%m-%d %H:%M:%S'),end=pd.to_datetime(row['DISCHTIME'],format='%Y-%m-%d %H:%M:%S'),freq='S'))\n",
+    "    admit_min = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(\"2021-12-01 00:00:00\", format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"D\",\n",
+    "        )\n",
+    "    )\n",
+    "    stay_len = len(\n",
+    "        pd.date_range(\n",
+    "            start=pd.to_datetime(row[\"ADMITTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            end=pd.to_datetime(row[\"DISCHTIME\"], format=\"%Y-%m-%d %H:%M:%S\"),\n",
+    "            freq=\"S\",\n",
+    "        )\n",
+    "    )\n",
     "\n",
-    "    num_days_admit = randrange(np.round(admit_min*0.25).astype(int),np.round(admit_min*0.9).astype(int)+5)\n",
-    "    num_days_disch = randrange(0,50)\n",
-    "    num_secs_chart = randrange(np.round(stay_len*0.01).astype(int)+1,np.round(stay_len*0.99).astype(int)+10)\n",
-    "    new_admit_date = pd.to_datetime(row['DOB'],format='%Y-%m-%d %H:%M:%S') + pd.DateOffset(days=num_days_admit)\n",
+    "    num_days_admit = randrange(np.round(admit_min * 0.25).astype(int), np.round(admit_min * 0.9).astype(int) + 5)\n",
+    "    num_days_disch = randrange(0, 50)\n",
+    "    num_secs_chart = randrange(np.round(stay_len * 0.01).astype(int) + 1, np.round(stay_len * 0.99).astype(int) + 10)\n",
+    "    new_admit_date = pd.to_datetime(row[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\") + pd.DateOffset(days=num_days_admit)\n",
     "    new_admits.append(new_admit_date)\n",
     "    new_dischs.append(new_admit_date + pd.DateOffset(days=num_days_disch))\n",
     "    new_chart.append(new_admit_date + pd.DateOffset(seconds=num_secs_chart))\n",
     "\n",
-    "mimic_table['ADMITTIME'] = new_admits\n",
-    "mimic_table['DISCHTIME'] = new_dischs\n",
-    "mimic_table['CHARTTIME'] = new_chart\n",
-    "\n",
-    "mimic_table = mimic_table[(pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME)) & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))]\n",
-    "mimic_table = mimic_table.merge(items[['ITEMID','LABEL']],on=['ITEMID'])\n",
-    "mimic_table.drop(['ICUSTAY_ID','ITEMID'],axis=1,inplace=True)\n",
-    "age_calc = pd.Timestamp('2021-12-01 00:00:00')\n",
-    "mimic_table['DOB'] = pd.to_datetime(mimic_table['DOB'], format='%Y-%m-%d %H:%M:%S')\n",
-    "mimic_table['age'] = (age_calc - mimic_table['DOB']).astype('<m8[Y]')\n",
+    "mimic_table[\"ADMITTIME\"] = new_admits\n",
+    "mimic_table[\"DISCHTIME\"] = new_dischs\n",
+    "mimic_table[\"CHARTTIME\"] = new_chart\n",
+    "\n",
+    "mimic_table = mimic_table[\n",
+    "    (pd.to_datetime(mimic_table.ADMITTIME) < pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "    & (pd.to_datetime(mimic_table.DISCHTIME) > pd.to_datetime(mimic_table.CHARTTIME))\n",
+    "]\n",
+    "mimic_table = mimic_table.merge(items[[\"ITEMID\", \"LABEL\"]], on=[\"ITEMID\"])\n",
+    "mimic_table.drop([\"ICUSTAY_ID\", \"ITEMID\"], axis=1, inplace=True)\n",
+    "age_calc = pd.Timestamp(\"2021-12-01 00:00:00\")\n",
+    "mimic_table[\"DOB\"] = pd.to_datetime(mimic_table[\"DOB\"], format=\"%Y-%m-%d %H:%M:%S\")\n",
+    "mimic_table[\"age\"] = (age_calc - mimic_table[\"DOB\"]).astype(\"<m8[Y]\")\n",
     "mimic_table.to_csv(output_path + \"MIMIC_{}_{}.csv\".format(mimic_table.shape[0], time.strftime(\"%Y%m%d\")), index=False)\n",
     "print(f\"Large input table saved, number of columns:  {mimic_table.shape[1]}, number of rows: {mimic_table.shape[0]}\")"
    ]
diff --git a/auxiliary/generate_SUPPORT.ipynb b/auxiliary/generate_SUPPORT.ipynb
index e2349d8e..94da4b92 100644
--- a/auxiliary/generate_SUPPORT.ipynb
+++ b/auxiliary/generate_SUPPORT.ipynb
@@ -44,8 +44,8 @@
    "outputs": [],
    "source": [
     "ref_date = datetime(2023, 1, 1)\n",
-    "df['x0'] = ref_date - pd.to_timedelta(df['x0'] * 365.2425, unit='D')\n",
-    "df = df.rename(columns={'x0': 'dob'})"
+    "df[\"x0\"] = ref_date - pd.to_timedelta(df[\"x0\"] * 365.2425, unit=\"D\")\n",
+    "df = df.rename(columns={\"x0\": \"dob\"})"
    ]
   },
   {
@@ -62,8 +62,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "_map = {x: chr(65 + int(x)) for x in df['x3'].unique().tolist()}\n",
-    "df['x3'] = df['x3'].map(_map)"
+    "_map = {x: chr(65 + int(x)) for x in df[\"x3\"].unique().tolist()}\n",
+    "df[\"x3\"] = df[\"x3\"].map(_map)"
    ]
   },
   {
@@ -88,7 +88,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "\n",
     "# Create a column sampled from a multi-modal gaussian distribution with means 1.0, 50.0 and 100.0 and stds 1.0, 2.0 and 3.0\n",
     "means = [1.0, 50.0, 100.0]\n",
     "stds = [1.0, 2.0, 3.0]\n",