def test_xwalk_state_bgp2000_tr2010_unrounded(self):
knw_str_vals = numpy.array([
["G10000509355299999051304R1", "G1000050051305", "10005051305"],
["G10000509355299999051304R1", "G1000050051306", "10005051306"],
["G10000509355299999051304R1", "G1000050051400", "10005051400"],
["G10000509355299999051304R2", "G1000050051305", "10005051305"],
])
knw_num_vals = numpy.array([
[6.80605382e-01, 6.33909150e-01, 6.57366450e-01, 6.59501671e-01],
[3.19167389e-01, 3.65781711e-01, 3.42281879e-01, 3.40110906e-01],
[2.27229039e-04, 3.09138740e-04, 3.51671251e-04, 3.87423412e-04],
[8.02660754e-01, 8.17567568e-01, 8.20895522e-01, 8.36236934e-01],
])
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=input_var_tags,
stfips=stfips,
vectorized=False,
keep_base=False,
weights_precision=None,
)
ix1, ix2 = 1025, 1029
id_cols = ["bgp2000gj", "tr2010gj", "tr2010ge"]
obs_str_vals = obs_xwalk.xwalk[id_cols][ix1:ix2].values
wgt_cols = ["wt_pop", "wt_fam", "wt_hh", "wt_hu"]
obs_num_vals = obs_xwalk.xwalk[wgt_cols][ix1:ix2].values
numpy.testing.assert_equal(knw_str_vals, obs_str_vals)
numpy.testing.assert_allclose(knw_num_vals, obs_num_vals)
def test_xwalk_extract_state_failure_bgp2000_tr2010(self):
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=input_var_tags,
keep_base=True,
stfips=stfips,
)
with self.assertRaises(RuntimeError):
obs_target_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name, obs_xwalk.target)
with self.assertRaises(RuntimeError):
obs_source_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name, obs_xwalk.source)
with self.assertRaises(RuntimeError):
obs_target_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_target_col)
with self.assertRaises(RuntimeError):
obs_source_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_source_col)
def test_xwalk_source_code_co(self):
# testing for triggered errors
with self.assertRaises(AttributeError):
observed_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=co,
target_geo=tr,
base_source_table=tab_data_path_1990,
input_var=input_vars_1990,
weight_var=input_var_tags,
)
def test_xwalk_uneven_input(self):
# testing for triggered errors
with self.assertRaises(RuntimeError):
observed_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=["one", "two"],
stfips=stfips,
)
def test_xwalk_state_bgp1990_tr2010(self, ):
knw_str_vals = numpy.array([
[
"G100001090444999990421009999999219012",
"G1000010042100",
"10001042100",
],
[
"G100001090444999990421009999999999921",
"G1000010042100",
"10001042100",
],
[
"G100001090444999990421009999999999921",
"G1000010042201",
"10001042201",
],
[
"G100001090444999990421009999999999922",
"G1000010042100",
"10001042100",
],
])
knw_num_vals = numpy.array([
[1.0, 1.0, 1.0, 1.0],
[0.99766436, 0.99716625, 0.99714829, 0.99727768],
[0.00233564, 0.00283375, 0.00285171, 0.00272232],
[1.0, 1.0, 1.0, 1.0],
])
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_1990,
supp_source_table=supplement_data_path_90,
input_var=input_vars_1990,
weight_var=input_var_tags,
stfips=stfips,
vectorized=False,
keep_base=False,
)
ix1, ix2 = 13, 17
id_cols = ["bgp1990gj", "tr2010gj", "tr2010ge"]
obs_str_vals = obs_xwalk.xwalk[id_cols][ix1:ix2].values
wgt_cols = ["wt_pop", "wt_fam", "wt_hh", "wt_hu"]
obs_num_vals = obs_xwalk.xwalk[wgt_cols][ix1:ix2].values
numpy.testing.assert_equal(knw_str_vals, obs_str_vals)
numpy.testing.assert_allclose(knw_num_vals, obs_num_vals, atol=6)
def test_xwalk_bgp1990_tr2010_no_supp_error(self):
with self.assertRaises(RuntimeError):
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_1990,
supp_source_table=None,
input_var=input_vars_1990,
weight_var=input_var_tags,
stfips=stfips,
vectorized=False,
keep_base=False,
)
def test_xwalk_write_read_csv_from_class(self):
write_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=input_var_tags,
keep_base=False,
stfips=stfips,
)
nhgisxwalk.xwalk_df_to_csv(cls=write_xwalk)
read_xwalk = nhgisxwalk.xwalk_df_from_csv(write_xwalk.xwalk_name)
known_values = write_xwalk.xwalk["wt_pop"].values
observed_values = read_xwalk["wt_pop"].values
numpy.testing.assert_allclose(known_values, observed_values)
def test_xwalk_extract_state_bgp1990_tr2010(self):
known_target_nan_xwalk = numpy.empty((0, 7))
known_source_nan_xwalk = numpy.array(
[[numpy.nan, "G1000050990000", "10005990000", 0.0, 0.0, 0.0, 0.0]],
dtype=object,
)
known_target_nan_base = numpy.empty((0, 6))
known_source_nan_base_shape = (149, 6)
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_1990,
supp_source_table=supplement_data_path_90,
input_var=input_vars_1990,
weight_var=input_var_tags,
keep_base=True,
)
obs_target_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name,
obs_xwalk.target).values
numpy.testing.assert_array_equal(known_target_nan_xwalk,
obs_target_nan_xwalk)
obs_source_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name,
obs_xwalk.source).values
numpy.testing.assert_array_equal(known_source_nan_xwalk[0, 0],
obs_source_nan_xwalk[0, 0])
numpy.testing.assert_array_equal(known_source_nan_xwalk[0, 1:3],
obs_source_nan_xwalk[0, 1:3])
obs_target_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_target_col).values
numpy.testing.assert_array_equal(known_target_nan_base,
obs_target_nan_base)
obs_source_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_source_col).values
self.assertEqual(known_source_nan_base_shape,
obs_source_nan_base.shape)
def test_xwalk_extract_unique_stfips_df_bgp2000_tr2010(self):
known_target_fips = set(["10"])
known_source_fips = set(["10", "34"])
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=input_var_tags,
keep_base=False,
stfips=stfips,
)
obs_target_fips = nhgisxwalk.extract_unique_stfips(df=obs_xwalk.xwalk,
endpoint="tr2010gj")
self.assertEqual(known_target_fips, obs_target_fips)
obs_source_fips = nhgisxwalk.extract_unique_stfips(
df=obs_xwalk.xwalk, endpoint="bgp2000gj")
self.assertEqual(known_source_fips, obs_source_fips)
def test_xwalk_extract_unique_stfips_cls_bgp1990_tr2010(self):
known_target_fips = set(["10"])
known_source_fips = set(["10", "34", "nan"])
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_1990,
supp_source_table=supplement_data_path_90,
input_var=input_vars_1990,
weight_var=input_var_tags,
keep_base=False,
stfips=stfips,
)
obs_target_fips = nhgisxwalk.extract_unique_stfips(cls=obs_xwalk,
endpoint="target")
self.assertEqual(known_target_fips, obs_target_fips)
obs_source_fips = nhgisxwalk.extract_unique_stfips(cls=obs_xwalk,
endpoint="source")
self.assertEqual(known_source_fips, obs_source_fips)
def test_xwalk_extract_state_bgp2000_tr2010(self):
known_target_nan_xwalk = numpy.empty((0, 7))
known_source_nan_xwalk = numpy.empty((0, 7))
known_target_nan_base = numpy.empty((0, 6))
known_source_nan_base = numpy.empty((0, 6))
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk2000_blk2010,
source_year=_00,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_2000,
input_var=input_vars_2000_SF1b,
weight_var=input_var_tags,
keep_base=True,
)
obs_target_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name,
obs_xwalk.target).values
numpy.testing.assert_array_equal(known_target_nan_xwalk,
obs_target_nan_xwalk)
obs_source_nan_xwalk = nhgisxwalk.extract_state(
obs_xwalk.xwalk, "nan", obs_xwalk.xwalk_name,
obs_xwalk.source).values
numpy.testing.assert_array_equal(known_source_nan_xwalk,
obs_source_nan_xwalk)
obs_target_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_target_col).values
numpy.testing.assert_array_equal(known_target_nan_base,
obs_target_nan_base)
obs_source_nan_base = nhgisxwalk.extract_state(
obs_xwalk.base, "nan", obs_xwalk.xwalk_name,
obs_xwalk.base_source_col).values
numpy.testing.assert_array_equal(known_source_nan_base,
obs_source_nan_base)
def test_generate_data_product(self):
# records known data values
knw_str_vals = numpy.array([
[
"G100001090444999990421009999999219012",
"G1000010042100",
"10001042100",
],
[
"G100001090444999990421009999999999921",
"G1000010042100",
"10001042100",
],
[
"G100001090444999990421009999999999921",
"G1000010042201",
"10001042201",
],
[
"G100001090444999990421009999999999922",
"G1000010042100",
"10001042100",
],
])
knw_num_vals = numpy.array([
[1.0, 1.0, 1.0, 1.0],
[0.99766436, 0.99716625, 0.99714829, 0.99727768],
[0.00233564, 0.00283375, 0.00285171, 0.00272232],
[1.0, 1.0, 1.0, 1.0],
])
# generate the product
xwalk_args = {
"source_year": _90,
"target_year": _10,
"source_geo": bgp,
"target_geo": tr,
"base_source_table": tab_data_path_1990,
"supp_source_table": supplement_data_path_90,
"input_var": input_vars_1990,
"weight_var": input_var_tags,
"keep_base": False,
"add_geoid": True,
}
nhgisxwalk.generate_data_product(base_xwalk_blk1990_blk2010,
xwalk_args, data_dir)
# record observed
obs_xwalk = nhgisxwalk.GeoCrossWalk(
base_xwalk_blk1990_blk2010,
source_year=_90,
target_year=_10,
source_geo=bgp,
target_geo=tr,
base_source_table=tab_data_path_1990,
supp_source_table=supplement_data_path_90,
input_var=input_vars_1990,
weight_var=input_var_tags,
)
# test
id_cols = ["bgp1990gj", "tr2010gj", "tr2010ge"]
data_types = nhgisxwalk.str_types(id_cols)
from_csv_kws = {
"path": data_dir,
"archived": True,
"remove_unpacked": True,
}
read_csv_kws = {"dtype": data_types}
read_xwalk = nhgisxwalk.xwalk_df_from_csv(obs_xwalk.xwalk_name,
**from_csv_kws,
**read_csv_kws)
ix1, ix2 = 13, 17
obs_str_vals = obs_xwalk.xwalk[id_cols][ix1:ix2].values
wgt_cols = ["wt_pop", "wt_fam", "wt_hh", "wt_hu"]
obs_num_vals = obs_xwalk.xwalk[wgt_cols][ix1:ix2].values
numpy.testing.assert_equal(knw_str_vals, obs_str_vals)
numpy.testing.assert_allclose(knw_num_vals, obs_num_vals, atol=6)
