def test_emdat_df_2020_load(self):
"""load selected sub sample from CSV, return DataFrame
here: from 2020 EM-DAT version to 2020 target_version"""
df = im_d.clean_emdat_df(EMDAT_2020_CSV_DEMO,
countries=['THA', 'Viet Nam'],
hazard='TC',
year_range=[2005, 2008],
target_version=2020)
self.assertIn('ISO', df.columns)
self.assertIn('Year', df.columns)
iso3 = list(df.ISO.unique())
years = np.arange(df.Year.min(), df.Year.max() + 1)
self.assertIn('THA', iso3)
self.assertIn('VNM', iso3)
self.assertNotIn('USA', iso3)
self.assertNotIn('TWN', iso3)
self.assertEqual(4, len(years))
self.assertEqual(2008, years[-1])
self.assertEqual(2006, years[1])
self.assertEqual(43, df.columns.size)
self.assertEqual(688, df.size)
self.assertEqual(624000, df["Total Damages ('000 US$)"].max())
self.assertIn('Tropical cyclone', list(df['Disaster Subtype']))
self.assertFalse(False in list(
df['Disaster Subtype'] == 'Tropical cyclone'))
self.assertFalse('Flood' in list(df['Disaster Subtype']))
def test_emdat_to_impact_2020format(self):
"""test import TC EM-DAT to Impact() from new 2020 EMDAT format CSV"""
df1 = im_d.clean_emdat_df(EMDAT_2020_CSV_DEMO,
hazard='TC',
countries='PHL',
year_range=(2013, 2013))
df2 = im_d.emdat_impact_event(EMDAT_2020_CSV_DEMO,
countries='PHL',
hazard='TC',
year_range=(2013, 2013),
reference_year=None,
imp_str='Total Affected')
impact_emdat, _countries = im_d.emdat_to_impact(
EMDAT_2020_CSV_DEMO,
'TC',
countries='PHL',
year_range=(2013, 2013),
imp_str="Total Affected")
# compare number of entries for all steps:
self.assertEqual(len(df1.index), len(df2.index))
self.assertEqual(impact_emdat.event_id.size, len(df1.index))
# TC events in EM-DAT in the Philipppines, 2013:
self.assertEqual(8, impact_emdat.event_id.size)
# People affected by TC events in the Philippines in 2013 (AAI):
self.assertAlmostEqual(17944571., impact_emdat.aai_agg, places=0)
# People affected by Typhoon Hayian in the Philippines:
self.assertAlmostEqual(1.610687e+07,
impact_emdat.at_event[4],
places=0)
def test_emdat_df_2018_to_2020_load(self):
"""load selected sub sample from CSV, return DataFrame
here: from 2018 EM-DAT version to 2020 target_version"""
df = im_d.clean_emdat_df(EMDAT_TEST_CSV, countries=['USA'], hazard='TC',
year_range=[2000, 2017], target_version=2020)
self.assertIn('ISO', df.columns)
self.assertIn('Year', df.columns)
iso3 = list(df.ISO.unique())
years = np.arange(df.Year.min(), df.Year.max() + 1)
self.assertListEqual(['USA'], iso3)
self.assertEqual(18, len(years))
self.assertEqual(2017, years[-1])
self.assertEqual(2010, years[10])
self.assertEqual(1634, df.size)
self.assertEqual(60000000, df["Insured Damages ('000 US$)"].max())
self.assertIn('Tropical cyclone', list(df['Disaster Subtype']))
self.assertFalse(False in list(df['Disaster Subtype'] == 'Tropical cyclone'))
self.assertFalse('Flood' in list(df['Disaster Subtype']))
def test_clean_emdat_df_2018_load(self):
"""load selected sub sample from CSV, return DataFrame.
here: from 2018 EM-DAT version to 2018 target_version"""
df = im_d.clean_emdat_df(EMDAT_TEST_CSV, countries=['Bangladesh'], hazard='TC',
year_range=[2000, 2017], target_version=2018)
self.assertIn('ISO', df.columns)
self.assertIn('Year', df.columns)
iso3 = list(df.ISO.unique())
years = np.arange(df.Year.min(), df.Year.max() + 1)
self.assertListEqual(['BGD'], iso3)
self.assertEqual(18, len(years))
self.assertEqual(2017, years[-1])
self.assertEqual(2010, years[10])
self.assertEqual(450, df.size)
self.assertEqual(8978541, df['Total affected'].max())
self.assertIn('Tropical cyclone', list(df['Disaster subtype']))
self.assertFalse(False in list(df['Disaster subtype'] == 'Tropical cyclone'))
self.assertFalse('Flood' in list(df['Disaster subtype']))