HowBadIsMyBatch/src/HowBadIsMyBatch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9de5907f-18f5-4cb1-903e-26028ff1fa03",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "pd.set_option('display.max_rows', 100)\n",
    "pd.set_option('display.max_columns', None)\n",
    "\n",
    "import os\n",
    "from VAERSFileDownloader import updateVAERSFiles\n",
    "from datetime import datetime\n",
    "from DateProvider import DateProvider\n",
    "from InternationalVaersCovid19Provider import getInternationalVaersCovid19, get_international_VAERSVAX_VAERSSYMPTOMS_Covid19\n",
    "from BatchCodeTableHtmlUpdater import updateBatchCodeTableHtmlFile\n",
    "from BatchCodeTablePersister import createAndSaveBatchCodeTables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d1e4fa9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(datetime.now().strftime(\"%d.%m.%Y, %H:%M:%S Uhr\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffad1c04",
   "metadata": {},
   "outputs": [],
   "source": [
    "dateProvider = DateProvider()\n",
    "print('          lastUpdated:', dateProvider.getLastUpdated())\n",
    "print('lastUpdatedDataSource:', dateProvider.getLastUpdatedDataSource())\n",
    "needsUpdate = dateProvider.needsUpdate()\n",
    "print('needsUpdate:', needsUpdate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "194b7357",
   "metadata": {},
   "outputs": [],
   "source": [
    "years_from_2020_to_present = list(range(2020, datetime.now().year + 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a793dff0",
   "metadata": {},
   "outputs": [],
   "source": [
    "updateVAERSFiles(\n",
    "        years = years_from_2020_to_present,\n",
    "        workingDirectory = os.getcwd())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "58333a19",
   "metadata": {},
   "outputs": [],
   "source": [
    "international_VAERSVAX_Covid19, international_VAERSSYMPTOMS = get_international_VAERSVAX_VAERSSYMPTOMS_Covid19(years = years_from_2020_to_present)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f10b558f",
   "metadata": {},
   "outputs": [],
   "source": [
    "international_VAERSVAX_Covid19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3e0908fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "international_VAERSSYMPTOMS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "204af94d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from SymptomByBatchcodeTableFactory import SymptomByBatchcodeTableFactory\n",
    "\n",
    "symptomByBatchcodeTable = SymptomByBatchcodeTableFactory.createSymptomByBatchcodeTable(international_VAERSVAX_Covid19, international_VAERSSYMPTOMS)\n",
    "symptomByBatchcodeTable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5997ee59",
   "metadata": {},
   "outputs": [],
   "source": [
    "isGlobal = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9268d60d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from SymptomHistogramByBatchcodeTableFactory import SymptomHistogramByBatchcodeTableFactory\n",
    "\n",
    "symptomHistogramByBatchcodeTable = SymptomHistogramByBatchcodeTableFactory.createGlobalSymptomHistogramByBatchcodeTable(symptomByBatchcodeTable) if isGlobal else SymptomHistogramByBatchcodeTableFactory.createSymptomHistogramByBatchcodeTable(symptomByBatchcodeTable)\n",
    "symptomHistogramByBatchcodeTable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5170efad",
   "metadata": {},
   "outputs": [],
   "source": [
    "from HistogramTable2DictTableConverter import HistogramTable2DictTableConverter\n",
    "\n",
    "dictByBatchcodeTable = HistogramTable2DictTableConverter.convertGlobalHistogramTable2DictTable(symptomHistogramByBatchcodeTable) if isGlobal else HistogramTable2DictTableConverter.convertHistogramTable2DictTable(symptomHistogramByBatchcodeTable)\n",
    "dictByBatchcodeTable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a57e0fd5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from HtmlUtils import getCountries\n",
    "from MultiIndexValuesProvider import MultiIndexValuesProvider\n",
    "from HistogramDescriptionPersister import HistogramDescriptionPersister\n",
    "import shutil\n",
    "\n",
    "if isGlobal:\n",
    "    batchcodes = MultiIndexValuesProvider.getValues(dictByBatchcodeTable.index)\n",
    "    batchcodes = {batchcode for batchcode in batchcodes if batchcode != 'nan'}\n",
    "    # directory = f'../docs/data/histograms/global'\n",
    "    directory = f'../docs/data/histograms'\n",
    "    shutil.rmtree(directory, ignore_errors = True)\n",
    "    histogramDescriptionPersister = HistogramDescriptionPersister(directory)\n",
    "    histogramDescriptionPersister.saveHistogramDescriptionsForBatchcodes(\n",
    "        batchcodes,\n",
    "        dictByBatchcodeTable,\n",
    "        progress = lambda count, size, batchcode: print(f'{count}/{size}: {batchcode}'))\n",
    "else:\n",
    "    countries = getCountries(international_VAERSVAX_Covid19)\n",
    "    for count, country in enumerate(countries, start = 1):\n",
    "        # FK-TODO: use https://github.com/tqdm/tqdm\n",
    "        print(f'saving histograms for country {count}/{len(countries)}: {country}')\n",
    "        dictByBatchcodeTable4Country = dictByBatchcodeTable[dictByBatchcodeTable['COUNTRY'] == country]\n",
    "        batchcodes = MultiIndexValuesProvider.getValues(dictByBatchcodeTable4Country.index)\n",
    "        batchcodes = {batchcode for batchcode in batchcodes if batchcode != 'nan'}\n",
    "        directory = f'../docs/data/histograms/{country}'\n",
    "        shutil.rmtree(directory, ignore_errors = True)\n",
    "        histogramDescriptionPersister = HistogramDescriptionPersister(directory)\n",
    "        histogramDescriptionPersister.saveHistogramDescriptionsForBatchcodes(\n",
    "            batchcodes,\n",
    "            dictByBatchcodeTable4Country,\n",
    "            progress = lambda count, size, batchcode: print(f'{count}/{size}: {batchcode}'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86918e80",
   "metadata": {},
   "outputs": [],
   "source": [
    "# batchcodes = international_VAERSVAX_Covid19['VAX_LOT'].dropna().drop_duplicates().to_list()\n",
    "# batchcodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "781ac80e",
   "metadata": {},
   "outputs": [],
   "source": [
    "internationalVaersCovid19 = getInternationalVaersCovid19(years = years_from_2020_to_present)\n",
    "internationalVaersCovid19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8d6507ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "updateBatchCodeTableHtmlFile(internationalVaersCovid19, batchCodeTableHtmlFile=\"../docs/batchCodeTable.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0915aa5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "createAndSaveBatchCodeTables(\n",
    "    internationalVaersCovid19,\n",
    "    minADRsForLethality = 100,\n",
    "    onCountryProcessed = display)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "howbadismybatch-venv-kernel",
   "language": "python",
   "name": "howbadismybatch-venv-kernel"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.8"
  },
  "vscode": {
   "interpreter": {
    "hash": "1bce2b9b19ce5f16d695ff75ac05095b3e564c169ff454b58b87cb796c0695b8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}