francois/HowBadIsMyBatch
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactoring

Browse Source
This commit is contained in:
frankknoll
2022-02-07 12:52:14 +01:00
parent 26f050b0d9
commit a2db3a869f
1 changed files with 45 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

100
HowBadIsMyBatch.ipynb
View File

@@ -370,6 +370,20 @@
"import unittest"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "fcc855dd",
"metadata": {},
"outputs": [],
"source": [
"class TestHelper:\n",
"\n",
" @staticmethod\n",
" def createDataFrame(index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
@@ -383,7 +397,7 @@
"\n",
" def test_convertVAX_LOTColumnToUpperCase(self):\n",
" # Given\n",
" dataFrame = self.createDataFrame(\n",
" dataFrame = TestHelper.createDataFrame(\n",
" columns = ['VAX_LOT'],\n",
" data = [ ['037K20A'],\n",
" ['025l20A'],\n",
@@ -397,7 +411,7 @@
" DataFrameNormalizer.convertVAX_LOTColumnToUpperCase(dataFrame)\n",
" \n",
" # Then\n",
" dataFrameExpected = self.createDataFrame(\n",
" dataFrameExpected = TestHelper.createDataFrame(\n",
" columns = ['VAX_LOT'],\n",
" data = [ ['037K20A'],\n",
" ['025L20A'],\n",
@@ -406,10 +420,7 @@
" \"0916600\",\n",
" \"0916601\",\n",
" \"1996874\"])\n",
" assert_frame_equal(dataFrame, dataFrameExpected, check_dtype = False)\n",
"\n",
" def createDataFrame(self, index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
" assert_frame_equal(dataFrame, dataFrameExpected, check_dtype = False)\n"
]
},
{
@@ -428,14 +439,14 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE'],\n",
" data = [ [1, 0, 0],\n",
" [0, 0, 1]],\n",
" index = [\n",
" \"0916600\",\n",
" \"0916601\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '037K20A', '1'],\n",
" ['COVID19', 'MODERNA', '025L20A', '1']],\n",
@@ -445,14 +456,14 @@
" dtypes = {'VAX_DOSE_SERIES': \"string\"})\n",
" },\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE'],\n",
" data = [ [0, 0, 0],\n",
" [0, 0, 1]],\n",
" index = [\n",
" \"1996873\",\n",
" \"1996874\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['HPV9', 'MERCK & CO. INC.', 'R017624', 'UNK'],\n",
" ['COVID19', 'MODERNA', '025L20A', '1']],\n",
@@ -469,7 +480,7 @@
" dataFrame = dataFrameFilter.filterBy(dataFrame, manufacturer = \"MODERNA\", dose = '1')\n",
" \n",
" # Then\n",
" dataFrameExpected = self.createDataFrame(\n",
" dataFrameExpected = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [1, 0, 0, 'COVID19', 'MODERNA', '037K20A', '1'],\n",
" [0, 0, 1, 'COVID19', 'MODERNA', '025L20A', '1'],\n",
@@ -486,14 +497,14 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'HOSPITAL', 'ER_VISIT'],\n",
" data = [ [1, 1, 0, 1, 1],\n",
" [0, 0, 1, 0, 1]],\n",
" index = [\n",
" \"0916600\",\n",
" \"0916601\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '037K20A', '1'],\n",
" ['COVID19', 'PFIZER\\BIONTECH', '025L20A', '1']],\n",
@@ -510,7 +521,7 @@
" dataFrame = dataFrameFilter.filterBy(dataFrame, dose = '1')\n",
" \n",
" # Then\n",
" dataFrameExpected = self.createDataFrame(\n",
" dataFrameExpected = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'HOSPITAL', 'ER_VISIT', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [1, 1, 0, 1, 1, 'COVID19', 'MODERNA', '037K20A', '1'],\n",
" [0, 0, 1, 0, 1, 'COVID19', 'PFIZER\\BIONTECH', '025L20A', '1']],\n",
@@ -525,12 +536,12 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE'],\n",
" data = [ [1, 0, 0]],\n",
" index = [\n",
" \"1048786\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '016M20A', '2'],\n",
" ['COVID19', 'MODERNA', '030L20A', '1']],\n",
@@ -547,7 +558,7 @@
" dataFrame = dataFrameFilter.filterBy(dataFrame, manufacturer = \"MODERNA\", dose = '1')\n",
" \n",
" # Then\n",
" dataFrameExpected = self.createDataFrame(\n",
" dataFrameExpected = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [1, 0, 0, 'COVID19', 'MODERNA', '030L20A', '1']],\n",
" index = [\n",
@@ -560,12 +571,12 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE'],\n",
" data = [ [1, 0, 0]],\n",
" index = [\n",
" \"1048786\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '016M20A', '2'],\n",
" ['COVID19', 'MODERNA', '030L20A', '1']],\n",
@@ -582,16 +593,13 @@
" dataFrame = dataFrameFilter.filterBy(dataFrame, manufacturer = \"MODERNA\", dose = '2')\n",
" \n",
" # Then\n",
" dataFrameExpected = self.createDataFrame(\n",
" dataFrameExpected = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [1, 0, 0, 'COVID19', 'MODERNA', '016M20A', '2']],\n",
" index = [\n",
" \"1048786\"],\n",
" dtypes = {'VAX_DOSE_SERIES': \"string\"})\n",
" assert_frame_equal(dataFrame, dataFrameExpected, check_dtype = False)\n",
"\n",
" def createDataFrame(self, index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
" assert_frame_equal(dataFrame, dataFrameExpected, check_dtype = False)\n"
]
},
{
@@ -610,14 +618,14 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'HOSPITAL', 'ER_VISIT'],\n",
" data = [ [1, 1, 0, 1, 1],\n",
" [0, 0, 1, 0, 1]],\n",
" index = [\n",
" \"0916600\",\n",
" \"0916601\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '037K20A', '1'],\n",
" ['COVID19', 'PFIZER\\BIONTECH', '025L20A', '1']],\n",
@@ -649,14 +657,14 @@
" dataFrame = VaersDescr2DataFrameConverter.createDataFrameFromDescrs(\n",
" [\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'HOSPITAL', 'ER_VISIT'],\n",
" data = [ [1, 0, 0, 0, 0],\n",
" [0, 0, 1, 0, 0]],\n",
" index = [\n",
" \"0916600\",\n",
" \"0916601\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['COVID19', 'MODERNA', '037K20A', '1'],\n",
" ['COVID19', 'MODERNA', '025L20A', '1']],\n",
@@ -666,14 +674,14 @@
" dtypes = {'VAX_DOSE_SERIES': \"string\"})\n",
" },\n",
" {\n",
" 'VAERSDATA': self.createDataFrame(\n",
" 'VAERSDATA': TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'HOSPITAL', 'ER_VISIT'],\n",
" data = [ [0, 0, 0, 0, 0],\n",
" [0, 0, 1, 0, 0]],\n",
" index = [\n",
" \"1996873\",\n",
" \"1996874\"]),\n",
" 'VAERSVAX': self.createDataFrame(\n",
" 'VAERSVAX': TestHelper.createDataFrame(\n",
" columns = ['VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ ['HPV9', 'MERCK & CO. INC.', 'R017624', 'UNK'],\n",
" ['COVID19', 'MODERNA', '025L20A', '1']],\n",
@@ -704,10 +712,7 @@
" 'LIFE THREATENING ILLNESSES': [0, 0]\n",
" },\n",
" index = pd.Index(['025L20A', '037K20A'], name = 'VAX_LOT'))\n",
" assert_frame_equal(batchCodeTable, batchCodeTableExpected, check_dtype = False)\n",
"\n",
" def createDataFrame(self, index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
" assert_frame_equal(batchCodeTable, batchCodeTableExpected, check_dtype = False)\n"
]
},
{
@@ -723,7 +728,7 @@
"\n",
" def test_getDoseTable(self):\n",
" # Given\n",
" dataFrame = self.createDataFrame(\n",
" dataFrame = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [1, 0, 0, 'COVID19', 'MODERNA', '016M20A', '2'],\n",
" [1, 0, 0, 'COVID19', 'MODERNA', '030L20A', '1'],\n",
@@ -753,7 +758,7 @@
" def test_getDoseByMonthTable(self):\n",
" # Given\n",
" parseDate = lambda dateStr: pd.to_datetime(dateStr, format = \"%m/%d/%Y\")\n",
" dataFrame = self.createDataFrame(\n",
" dataFrame = TestHelper.createDataFrame(\n",
" columns = ['RECVDATE', 'DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES'],\n",
" data = [ [parseDate('01/01/2021'), 1, 0, 0, 'COVID19', 'MODERNA', '016M20A', '2'],\n",
" [parseDate('01/01/2021'), 1, 0, 0, 'COVID19', 'MODERNA', '030L20A', '1'],\n",
@@ -784,10 +789,7 @@
" (2021, 1, '2'),\n",
" ],\n",
" names = ('Year', 'Month', 'Dose'))),\n",
" check_index_type = False)\n",
"\n",
" def createDataFrame(self, index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
" check_index_type = False)\n"
]
},
{
@@ -803,7 +805,7 @@
"\n",
" def test_getInternationalLotTable(self):\n",
" # Given\n",
" dataFrame = self.createDataFrame(\n",
" dataFrame = TestHelper.createDataFrame(\n",
" columns = ['DIED', 'L_THREAT', 'DISABLE', 'VAX_TYPE', 'VAX_MANU', 'VAX_LOT', 'VAX_DOSE_SERIES', 'SPLTTYPE'],\n",
" data = [ [1, 0, 0, 'COVID19', 'MODERNA', '016M20A', '2', 'GBPFIZER INC2020486806'],\n",
" [1, 0, 0, 'COVID19', 'MODERNA', '030L20A', '1', 'FRMODERNATX, INC.MOD20224'],\n",
@@ -823,7 +825,7 @@
" # Then\n",
" assert_frame_equal(\n",
" internationalLotTable,\n",
" self.createDataFrame(\n",
" TestHelper.createDataFrame(\n",
" columns = ['Total reports', 'Deaths', 'Disabilities', 'Life Threatening Illnesses', 'Severe reports (%)'],\n",
" data = [ [2, 2, 1, 1, (2 + 1 + 1) / 2 * 100],\n",
" [1, 1, 0, 0, (1 + 0 + 0) / 1 * 100],\n",
@@ -834,11 +836,7 @@
" 'United Kingdom',\n",
" 'Unknown Country'\n",
" ],\n",
" name = 'Country')))\n",
"\n",
" # FK-TODO: createDataFrame() is defined in almost every test class: DRY \n",
" def createDataFrame(self, index, columns, data, dtypes = {}):\n",
" return pd.DataFrame(index = index, columns = columns, data = data).astype(dtypes)\n"
" name = 'Country')))\n"
]
},
{
@@ -1056,14 +1054,6 @@
"internationalLotTable.to_excel('results/International_Deadly_Lots.xlsx')\n",
"internationalLotTable"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "664ebc44",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {