Add topofeatures (#1241)

TopologicalFeaturesImplementation was added

cases/dataset_preparation.py

@@ -8,24 +8,23 @@ def unpack_archived_data(archive_name: str):
             os.path.basename(archive_name) not in os.listdir(os.path.dirname(archive_path))):
         with tarfile.open(archive_path) as file:
             def is_within_directory(directory, target):
-                
+
                 abs_directory = os.path.abspath(directory)
                 abs_target = os.path.abspath(target)
-            
+
                 prefix = os.path.commonprefix([abs_directory, abs_target])
-                
+
                 return prefix == abs_directory
-            
+
             def safe_extract(tar, path=".", members=None, *, numeric_owner=False):
-            
+
                 for member in tar.getmembers():
                     member_path = os.path.join(path, member.name)
                     if not is_within_directory(path, member_path):
                         raise Exception("Attempted Path Traversal in Tar File")
-
-                tar.extractall(path, members, numeric_owner=numeric_owner) 
-
-
+
+                tar.extractall(path, members, numeric_owner=numeric_owner)
+
             safe_extract(file, path=os.path.dirname(archive_path))
         print('Unpacking finished')
     else:

cases/time_series_gapfilling_case.py

@@ -80,7 +80,7 @@ def get_composite_pipeline():
     node_linear_2 = PipelineNode('linear', nodes_from=[node_2])
 
     node_final = PipelineNode('ridge', nodes_from=[node_linear_1,
-                                                    node_linear_2])
+                                                   node_linear_2])
     pipeline = Pipeline(node_final)
     return pipeline
 

docs/source/introduction/fedot_features/automation_features.rst

@@ -69,6 +69,7 @@ FEDOT supports bunch of dimensionality preprocessing operations that can be be a
    `one_hot_encoding`,One-Hot Encoder, Feature encoding
    `label_encoding`,Label Encoder, Feature encoding
    `resample`,Imbalanced binary class transformation in classification, Data transformation
+   `topological_features`,Calculation of topological features, only for time series,Data transformation
 
 
 .. csv-table:: Feature transformation operations implementations
@@ -104,6 +105,7 @@ FEDOT supports bunch of dimensionality preprocessing operations that can be be a
    `one_hot_encoding`,`sklearn.preprocessing.OneHotEncoder`,
    `label_encoding`,`sklearn.preprocessing.LabelEncoder`,`fast_train` `*tree`
    `resample`,`FEDOT model using sklearn.utils.resample`,
+   `topological_features`,FEDOT model,`ts`
 
 
 Models used

examples/advanced/automl/tpot_vs_fedot.py

@@ -1,18 +1,16 @@
+from fedot.core.data.data import InputData
+from fedot.core.pipelines.node import PipelineNode
+from fedot.core.pipelines.pipeline import Pipeline
+from tpot.export_utils import set_param_recursive
+from tpot.builtins import StackingEstimator
+from sklearn.pipeline import make_pipeline
+from sklearn.naive_bayes import BernoulliNB
+from sklearn.metrics import roc_auc_score as roc_auc
+from sklearn.ensemble import RandomForestClassifier
 import numpy
 
 numpy.float = numpy.float64  # tmp patch before TPOT could fix this: https://github.com/EpistasisLab/tpot/issues/1281
 
-from sklearn.ensemble import RandomForestClassifier
-from sklearn.metrics import roc_auc_score as roc_auc
-from sklearn.naive_bayes import BernoulliNB
-from sklearn.pipeline import make_pipeline
-from tpot.builtins import StackingEstimator
-from tpot.export_utils import set_param_recursive
-
-from fedot.core.pipelines.pipeline import Pipeline
-from fedot.core.pipelines.node import PipelineNode
-from fedot.core.data.data import InputData
-
 
 def run_tpot_vs_fedot_example(train_file_path: str, test_file_path: str):
     train_data = InputData.from_csv(train_file_path)

examples/advanced/time_series_forecasting/multistep.py

@@ -4,6 +4,7 @@
 import pandas as pd
 
 from examples.advanced.time_series_forecasting.composing_pipelines import get_border_line_info
+from examples.simple.time_series_forecasting.api_forecasting import TS_DATASETS
 from examples.simple.time_series_forecasting.ts_pipelines import ts_ar_pipeline
 from examples.simple.time_series_forecasting.tuning_pipelines import visualise
 from fedot.core.data.data import InputData
@@ -19,13 +20,6 @@
 
 _TS_EXAMPLES_DATA_PATH = fedot_project_root().joinpath('examples/data/ts')
 
-TS_DATASETS = {
-    'australia': _TS_EXAMPLES_DATA_PATH.joinpath('australia.csv'),
-    'beer': _TS_EXAMPLES_DATA_PATH.joinpath('beer.csv'),
-    'salaries': _TS_EXAMPLES_DATA_PATH.joinpath('salaries.csv'),
-    'stackoverflow': _TS_EXAMPLES_DATA_PATH.joinpath('stackoverflow.csv'),
-    'test_sea': fedot_project_root().joinpath('test', 'data', 'simple_sea_level.csv')}
-
 
 def run_multistep(dataset: str, pipeline: Pipeline, step_forecast: int = 10, future_steps: int = 5,
                   visualisation=False):

examples/advanced/time_series_forecasting/nemo.py

@@ -117,11 +117,11 @@ def return_working_pipeline():
 
 pipeline = get_arima_pipeline()
 train_dataset = MultiModalData({
-        'arima': deepcopy(train_input),
-    })
+    'arima': deepcopy(train_input),
+})
 predict_dataset = MultiModalData({
-        'arima': deepcopy(predict_input),
-    })
+    'arima': deepcopy(predict_input),
+})
 pipeline.fit_from_scratch(train_dataset)
 predicted_values = pipeline.predict(predict_dataset)
 predicted_values = predicted_values.predict
@@ -140,13 +140,13 @@ def return_working_pipeline():
 # arima with nemo ensemble
 pipeline = return_working_pipeline()
 train_dataset = MultiModalData({
-        'lagged/1': deepcopy(train_input),
-        'exog_ts': deepcopy(train_input_exog)
-    })
+    'lagged/1': deepcopy(train_input),
+    'exog_ts': deepcopy(train_input_exog)
+})
 predict_dataset = MultiModalData({
-        'lagged/1': deepcopy(predict_input),
-        'exog_ts': deepcopy(predict_input_exog)
-    })
+    'lagged/1': deepcopy(predict_input),
+    'exog_ts': deepcopy(predict_input_exog)
+})
 pipeline.fit_from_scratch(train_dataset)
 predicted_values = pipeline.predict(predict_dataset).predict
 
@@ -165,13 +165,13 @@ def return_working_pipeline():
 # arima with nemo ensemble
 pipeline = get_arima_nemo_pipeline()
 train_dataset = MultiModalData({
-        'arima': deepcopy(train_input),
-        'exog_ts': deepcopy(train_input_exog)
-    })
+    'arima': deepcopy(train_input),
+    'exog_ts': deepcopy(train_input_exog)
+})
 predict_dataset = MultiModalData({
-        'arima': deepcopy(predict_input),
-        'exog_ts': deepcopy(predict_input_exog)
-    })
+    'arima': deepcopy(predict_input),
+    'exog_ts': deepcopy(predict_input_exog)
+})
 pipeline.fit_from_scratch(train_dataset)
 predicted_values = pipeline.predict(predict_dataset).predict
 

examples/advanced/time_series_forecasting/prediction_intervals.py

@@ -35,10 +35,10 @@ def build_pred_ints(start=5000, end=7000, horizon=200):
                             task=task,
                             data_type=DataTypesEnum.ts)
     model = Fedot(problem='ts_forecasting',
-              task_params=task.task_params,
-              timeout=3,
-              preset='ts',
-              show_progress=False)
+                  task_params=task.task_params,
+                  timeout=3,
+                  preset='ts',
+                  show_progress=False)
 
     model.fit(train_input)
     model.forecast()