"number of feature levels of years and nid should match number"
assert len(df_agg['age_start'].unique()) == len(df_agg['age_end'].unique()),\
"number of feature levels age start should match number of feature " +\
r"levels age end"
assert len(df_agg['diagnosis_id'].unique()) <= 2,\
"diagnosis_id should have 2 or less feature levels"
assert len(df_agg['code_system_id'].unique()) <= 2,\
"code_system_id should have 2 or less feature levels"
assert len(df_agg['source'].unique()) == 1,\
"source should only have one feature level"
assert round(val_sum, 3) == round(df_agg.val.sum(), 3),\
"some cases were lost"
fpath = r"FILEPATH"
compare_df = pd.read_hdf(fpath)
test_results = stage_hosp_prep.test_case_counts(df_agg, compare_df)
if test_results == "test_case_counts has passed!!":
pass
else:
msg = " --- ".join(test_results)
assert False, msg
write_path = root + r"FILENAME" +\
r"FILEPATH"
write_hosp_file(df_agg, write_path, backup=True)
assert len(df_agg['sex_id'].unique()) == 2,\
"There should only be two feature levels to sex_id"
assert len(df_agg['code_system_id'].unique()) <= 2,\
"code_system_id should have 2 or less feature levels"
assert len(df_agg['source'].unique()) == 1,\
"source should only have one feature level"
assert (df.val >= 0).all(), ("for some reason there are negative case counts")
updated_locs = [4761, 4750, 4760]
test_df = df_agg[~df_agg.location_id.isin(updated_locs)]
compare_df = pd.read_hdf("FILEPATH")
compare_df = compare_df[~compare_df.location_id.isin(updated_locs)]
test_results = stage_hosp_prep.test_case_counts(test_df, compare_df)
if test_results == "test_case_counts has passed!!":
pass
else:
msg = " --- ".join(test_results)
assert False, msg
write_path = root + r"FILENAME"\
"FILEPATH"
write_hosp_file(df_agg, write_path, backup=True)