def prepare_data(self, data: DataProvider):
df_all = data.get_all_transactions()
df_agg_months = data.aggregate_by_category(df_all, 'MS',
config.CATEGORY_MAIN_COL)
self._prepare_data_monthly_chart(data, df_agg_months)
self._prepare_data_average_table(df_agg_months.reset_index())
def prepare_data(self, data: DataProvider):
df_all = data.get_all_transactions().copy()
df_all.loc[df_all[config.CATEGORY_SUB_COL] == '',
config.CATEGORY_SUB_COL] = config.MISC_CATEGORY
for main_category in config.categories.keys():
df_cat = df_all.loc[
df_all[config.CATEGORY_MAIN_COL] == main_category,
[config.CATEGORY_SUB_COL, config.ABSAMOUNT_COL]]
self._x_axis, sub_category_values = data.aggregate_by_category_as_tuple(
df_cat, 'MS', config.CATEGORY_SUB_COL, main_category)
self._main_category_values[main_category] = sub_category_values
def test_aggregate_transactions_by_year(self):
transactions = [
Transaction('123456', datetime(year=2018, month=3, day=5), 120.0,
'reason', 'recipient', config.INCOME_CATEGORY),
Transaction('123456', datetime(year=2018, month=5, day=18), -50.0,
'reason', 'recipient', self.category1),
Transaction('123456', datetime(year=2018, month=8, day=15), -30.0,
'reason', 'recipient', self.category1),
Transaction('123456', datetime(year=2018, month=9, day=3), -100.0,
'reason', 'recipient', self.category2),
Transaction('123456', datetime(year=2019, month=6, day=21), -150.0,
'reason', 'recipient', self.category2),
]
data_provider = DataProvider.load(transactions)
df_all = data_provider.get_all_transactions()
x_axis, category_values = data_provider.aggregate_by_category_as_tuple(
df_all, 'YS', config.CATEGORY_MAIN_COL)
self.assertEqual(2, len(x_axis))
self.assertEqual(3, len(
category_values.keys())) # includes income category
self.assertListEqual([120.0, 0.0],
list(category_values[config.INCOME_CATEGORY]))
self.assertListEqual([80.0, 0.0],
list(category_values[self.category1]))
self.assertListEqual([100.0, 150.0],
list(category_values[self.category2]))
def test_aggregate_different_expenses_by_month(self):
transactions = [
Transaction('123456', datetime(year=2019, month=3, day=1), -20.0,
'reason', 'recipient', self.category1),
Transaction('123456', datetime(year=2019, month=3, day=18), -50.0,
'reason', 'recipient', self.category2),
Transaction('123456', datetime(year=2019, month=4, day=15), -30.0,
'reason', 'recipient', self.category1),
Transaction('123456', datetime(year=2019, month=6, day=5), -100.0,
'reason', 'recipient', self.category2),
Transaction('123456', datetime(year=2019, month=6, day=21), -150.0,
'reason', 'recipient', self.category2),
]
data_provider = DataProvider.load(transactions)
df_all = data_provider.get_all_transactions()
x_axis, category_values = data_provider.aggregate_by_category_as_tuple(
df_all, 'MS', config.CATEGORY_MAIN_COL)
self.assertEqual(4, len(x_axis))
self.assertEqual(3, len(category_values.keys()))
self.assertListEqual([20.0, 30.0, 0.0, 0.0],
list(category_values[self.category1]))
self.assertListEqual([50.0, 0.0, 0.0, 250.0],
list(category_values[self.category2]))
def assign_category_to_transactions():
global _transactions
transactions = _transactions
category_finder = CategoryFinder()
category_finder.assign_category(transactions)
global _data_provider
_data_provider = DataProvider.load(transactions)
print_transactions_statistics(transactions)
def prepare_data(self, data: DataProvider):
df_out = data.get_out_transactions()
self._df_agg_years = df_out.groupby([
df_out.index.year, config.CATEGORY_MAIN_COL,
config.CATEGORY_SUB_COL
])[config.ABSAMOUNT_COL].sum().reset_index()
self._df_agg_years.loc[
self._df_agg_years.sub_category == '',
config.CATEGORY_SUB_COL] = None # do not plot missing values
def prepare_data(self, data: DataProvider):
column_order = [config.PAYMENT_REASON_COL,
config.RECIPIENT_COL,
config.CATEGORY_MAIN_COL,
config.CATEGORY_SUB_COL,
config.AMOUNT_COL]
self._df_all_ta = data.get_all_transactions(). \
sort_values(by=[config.DATE_COL, config.CATEGORY_MAIN_COL, config.CATEGORY_SUB_COL]). \
loc[:, column_order].reset_index()
def prepare_data(self, data: DataProvider):
"""
Accumulates all transactions by category
"""
df_all = data.get_all_transactions()
self._x_axis = list(
map(lambda date: date,
df_all.resample('D').sum().index))
for category_name in reversed(list(config.categories.keys())):
df_category = df_all[df_all[config.CATEGORY_MAIN_COL] ==
category_name]
df_category = df_category.resample('D').sum().reindex(
df_all.index).resample('D').max().fillna(0)
values = list(df_category[config.ABSAMOUNT_COL].cumsum())
self._cumulative_categories[category_name] = values
def test_aggregate_income_by_month(self):
transactions = [
Transaction('123456', datetime(year=2019, month=3, day=5), 20.0,
'reason', 'recipient', config.INCOME_CATEGORY),
Transaction('123456', datetime(year=2019, month=3, day=18), 50.0,
'reason', 'recipient', config.INCOME_CATEGORY),
Transaction('123456', datetime(year=2019, month=5, day=8), 100.0,
'reason', 'recipient', config.INCOME_CATEGORY),
]
data_provider = DataProvider.load(transactions)
df_all = data_provider.get_all_transactions()
x_axis, category_values = data_provider.aggregate_by_category_as_tuple(
df_all, 'MS', config.CATEGORY_MAIN_COL)
self.assertEqual(3, len(x_axis))
self.assertEqual(3, len(category_values.keys()))
self.assertListEqual([70.0, 0.0, 100.0],
list(category_values[config.INCOME_CATEGORY]))
def prepare_data(self, data: DataProvider):
"""
Preprocesses each transaction and calculates the relative amount within its category
"""
RATIO = 'ratio'
df_all = data.get_all_transactions()
for category_name in config.categories.keys():
df_category = df_all[df_all[config.CATEGORY_MAIN_COL] ==
category_name]
category_total = df_category[config.ABSAMOUNT_COL].sum()
df_category.loc[:, RATIO] = df_category[
config.ABSAMOUNT_COL] / category_total
x_axis = list(
map(lambda datetime: pd.Timestamp(datetime),
pd.DatetimeIndex(df_category.index).values))
if x_axis:
self._category_values[category_name] = (
x_axis, df_category[config.ABSAMOUNT_COL].values,
df_category[RATIO].values,
df_category[config.LABEL].values)
def _prepare_data_monthly_chart(self, data: DataProvider,
df_agg_months: pd.DataFrame):
self._x_axis, self._category_values = data.expand_by_categories(
df_agg_months, config.CATEGORY_MAIN_COL)
def prepare_data(self, data: DataProvider):
df_all = data.get_all_transactions()
self._x_axis, self._category_values = data.aggregate_by_category_as_tuple(
df_all, 'YS', config.CATEGORY_MAIN_COL)