def _predict_time_series_top_down(self, dictionary, X=None):
for key in self.level_nodes[self.level]:
submodel = HierarchicalTopDown(
model=self.model,
hierarchy_tree=self.sub_trees_dictionary[key],
root=key,
method=self.method)
submodel.models_ = self._extract_fitted_models(key)
submodel.proportions = self._extract_proportions(key)
dictionary.update(submodel.predict(X))
return
def hierarchical_top_down_model_sgf(draw,
time_series_forecasting_model1_no_cache):
dataframes = draw(n_time_series_with_same_index(min_n=5))
tree = draw(tree_construction(dataframes))
method = "tdsgf"
return HierarchicalTopDown(
model=time_series_forecasting_model1_no_cache,
hierarchy_tree=tree,
method=method,
)
def _predict_new_time_series(self,
X: pd.DataFrame) -> Dict[str, pd.DataFrame]:
new_time_series_dictionary = {}
if self.level == 0:
new_time_series_dictionary = HierarchicalTopDown._predict_new_time_series(
self, X)
elif self.level == max(list(self.level_nodes.keys())):
new_time_series_dictionary = HierarchicalBottomUp._predict_new_time_series(
self, X)
else:
self._predict_time_series_top_down(new_time_series_dictionary, X)
self._predict_new_time_series_bottom_up(new_time_series_dictionary,
X)
return new_time_series_dictionary
def hierarchical_top_down_model_sga_tree_by_hand(
draw, time_series_forecasting_model1_no_cache):
tree_adjacency_list = {
"0": {},
"1": {},
"2": {
"1": {},
"4": {}
},
"3": {
"0": {},
"2": {}
},
"4": {},
}
tree = nx.DiGraph(tree_adjacency_list)
root = "3"
return HierarchicalTopDown(model=time_series_forecasting_model1_no_cache,
hierarchy_tree=tree,
root=root)
def hierarchical_basic_top_down_model_fp_method(
time_series_forecasting_model1_no_cache, ):
return HierarchicalTopDown(time_series_forecasting_model1_no_cache,
hierarchy_tree="infer",
method="tdfp")
def hierarchical_basic_top_down_model(time_series_forecasting_model1_no_cache):
return HierarchicalTopDown(time_series_forecasting_model1_no_cache,
"infer")
def test_prediction_values_top_down_tdfp(self):
index = [
"2000-01-01 00:00:00",
"2000-01-01 00:00:01",
"2000-01-01 00:00:02",
"2000-01-01 00:00:03",
"2000-01-01 00:00:04",
"2000-01-01 00:00:05",
"2000-01-01 00:00:06",
"2000-01-01 00:00:07",
"2000-01-01 00:00:08",
"2000-01-01 00:00:09",
"2000-01-01 00:00:10",
"2000-01-01 00:00:11",
"2000-01-01 00:00:12",
"2000-01-01 00:00:13",
"2000-01-01 00:00:14",
"2000-01-01 00:00:15",
"2000-01-01 00:00:16",
"2000-01-01 00:00:17",
"2000-01-01 00:00:18",
"2000-01-01 00:00:19",
]
time_series_model = AR(p=2, horizon=3)
data1_list = [1 for i in range(20)]
data2_list = [1 for i in range(10)] + [0 for i in range(10)]
data3_list = [0 for i in range(10)] + [1 for i in range(10)]
data1 = pd.DataFrame(index=index, data=data1_list)
data2 = pd.DataFrame(index=index, data=data2_list)
data3 = pd.DataFrame(index=index, data=data3_list)
data = {"data1": data1, "data2": data2, "data3": data3}
tree_adj = {"data1": ["data2", "data3"], "data2": [], "data3": []}
values_prediction_data1 = [[1.0, 1.0, 1.0], [1.0, 1.0, 1.0],
[1.0, 1.0, 1.0]]
values_prediction_child = [[0.0, 0.0, 0.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]]
test_prediction_data1 = pd.DataFrame(
data=values_prediction_data1,
columns=["y_1", "y_2", "y_3"],
index=[
"2000-01-01 00:00:17", "2000-01-01 00:00:18",
"2000-01-01 00:00:19"
],
)
test_prediction_data2 = pd.DataFrame(
data=values_prediction_child,
columns=["y_1", "y_2", "y_3"],
index=[
"2000-01-01 00:00:17", "2000-01-01 00:00:18",
"2000-01-01 00:00:19"
],
)
test_prediction_data3 = pd.DataFrame(
data=values_prediction_data1,
columns=["y_1", "y_2", "y_3"],
index=[
"2000-01-01 00:00:17", "2000-01-01 00:00:18",
"2000-01-01 00:00:19"
],
)
prediction = HierarchicalTopDown(
model=time_series_model,
hierarchy_tree=tree_adj,
root="data1",
method="tdfp",
)
test_prediction_data = {
"data1": test_prediction_data1,
"data2": test_prediction_data2,
"data3": test_prediction_data3,
}
prediction = prediction.fit(data).predict()
for key in test_prediction_data.keys():
assert_frame_equal(test_prediction_data[key], prediction[key])
def test_constructor(self, time_series_forecasting_model1_no_cache,
dataframes):
tree = tree_construction(dataframes)
HierarchicalTopDown(time_series_forecasting_model1_no_cache, tree)
def test_basic_constructor(self, time_series_forecasting_model1_no_cache):
HierarchicalTopDown(model=time_series_forecasting_model1_no_cache,
hierarchy_tree="infer")