12. adding of eeg dataset with bayesian tests

imputegap/contamination/contamination.py

@@ -145,7 +145,7 @@ def scenario_missing_percentage(ts, series_impacted=0.2, missing_rate=0.2, prote
             W = int((N - P) * missing_rate)  # number of data to remove
 
             for to_remove in range(0, W):
-                I = P + to_remove
-                ts_contaminated[S, I] = np.nan
+                index = P + to_remove
+                ts_contaminated[S, index] = np.nan
 
         return ts_contaminated

imputegap/imputation/imputation.py

@@ -12,6 +12,7 @@
 
 class Imputation:
 
+    @classmethod
     def load_parameters(query: str = "default", algorithm: str = "cdrec"):
         """
         Load default values of algorithms
@@ -22,6 +23,7 @@ def load_parameters(query: str = "default", algorithm: str = "cdrec"):
         :return: tuples of optimal parameters and the config of default values
         """
 
+        filepath = ""
         if query == "default":
             filepath = "../env/default_values.toml"
         elif query == "optimal":
@@ -32,34 +34,34 @@ def load_parameters(query: str = "default", algorithm: str = "cdrec"):
         if not os.path.exists(filepath):
             filepath = filepath[1:]
 
+        print(filepath)
+
         with open(filepath, "r") as _:
             config = toml.load(filepath)
 
-        params = None
         if algorithm == "cdrec":
             truncation_rank = int(config['cdrec']['rank'])
             epsilon = config['cdrec']['epsilon']
             iterations = int(config['cdrec']['iteration'])
-            params = (truncation_rank, epsilon, iterations)
+            return (truncation_rank, epsilon, iterations)
         elif algorithm == "stmvl":
             window_size = int(config['stmvl']['window_size'])
             gamma = float(config['stmvl']['gamma'])
             alpha = int(config['stmvl']['alpha'])
-            params = (window_size, gamma, alpha)
+            return (window_size, gamma, alpha)
         elif algorithm == "iim":
             learning_neighbors = int(config['iim']['learning_neighbors'])
             algo_code = config['iim']['algorithm_code']
-            params = (learning_neighbors, algo_code)
+            return (learning_neighbors, algo_code)
         elif algorithm == "mrnn":
             hidden_dim = int(config['mrnn']['hidden_dim'])
             learning_rate = float(config['mrnn']['learning_rate'])
             iterations = int(config['mrnn']['iterations'])
             sequence_length = int(config['mrnn']['sequence_length'])
-            params = (hidden_dim, learning_rate, iterations, sequence_length)
+            return (hidden_dim, learning_rate, iterations, sequence_length)
         else :
             print("Default/Optimal config not found for this algorithm")
-
-        return params
+            return None
 
     def evaluate_params(ground_truth, contamination, configuration, algorithm="cdrec"):
         """

tests/test_contamination_mcar.py

@@ -142,8 +142,7 @@ def test_mcar_selection_datasets(self):
 
                         # 2) Check if the correct percentage of series are contaminated
                         contaminated_series = np.isnan(ts_contaminate).any(axis=1).sum()
-                        self.assertEqual(contaminated_series, expected_contaminated_series,
-                                         f"Expected {expected_contaminated_series} contaminated series but found {contaminated_series}")
+                        self.assertEqual(contaminated_series, expected_contaminated_series, f"Expected {expected_contaminated_series} contaminated series but found {contaminated_series}")
 
     def test_mcar_position_datasets(self):
         """

tests/test_contamination_mp.py

@@ -1,6 +1,7 @@
 import os
 import unittest
 import numpy as np
+import math
 
 from imputegap.contamination.contamination import Contamination
 from imputegap.manager.manager import TimeSeries
@@ -32,7 +33,7 @@ def get_file_path(set_name="test"):
 
 class TestContamination(unittest.TestCase):
 
-    def test_mcar_selection(self):
+    def test_mp_selection(self):
         """
         the goal is to test if only the selected values are contaminated
         """
@@ -41,13 +42,17 @@ def test_mcar_selection(self):
         series_impacted = [0.4]
         missing_rates = [0.4]
         seeds_start, seeds_end = 42, 43
-        series_check = ["1", "2", "3", "4"]
         protection = 0.1
 
+        length_of_gap_ts = len(impute_gap.ts[0])
+        len_expected = math.ceil(missing_rates[0] * length_of_gap_ts)
+        series_check = [str(i) for i in range(len_expected)]
+
         for seed_value in range(seeds_start, seeds_end):
             for series_sel in series_impacted:
                 for missing_rate in missing_rates:
 
+
                     ts_contaminate = Contamination.scenario_missing_percentage(ts=impute_gap.ts,
                                                                  series_impacted=series_sel,
                                                                  missing_rate=missing_rate,
@@ -69,7 +74,7 @@ def test_mcar_selection(self):
 
                     self.assertTrue(check_nan_series, True)
 
-    def test_mcar_position(self):
+    def test_mp_position(self):
         """
         the goal is to test if the starting position is always guaranteed
         """

tests/test_opti_bayesian_cdrec.py

@@ -57,5 +57,5 @@ def test_optimization_bayesian_cdrec(self):
 
         Optimization.save_optimization(optimal_params=optimal_params, algorithm=algorithm+"_test")
 
-        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], True, f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
+        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
         self.assertTrue(yi > 0, True)

tests/test_opti_bayesian_iim.py

@@ -60,5 +60,5 @@ def test_optimization_bayesian_iim(self):
 
         Optimization.save_optimization(optimal_params=optimal_params, algorithm=algorithm+"_test")
 
-        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], True, f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
+        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
         self.assertTrue(yi > 0, True)

tests/test_opti_bayesian_mrnn.py

@@ -61,5 +61,5 @@ def test_optimization_bayesian_mrnn(self):
 
         Optimization.save_optimization(optimal_params=optimal_params, algorithm=algorithm+"_test")
 
-        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], True, f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
+        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
         self.assertTrue(yi > 0, True)

tests/test_opti_bayesian_stmvl.py

@@ -61,5 +61,5 @@ def test_optimization_bayesian_stmvl(self):
 
         Optimization.save_optimization(optimal_params=optimal_params, algorithm=algorithm+"_test")
 
-        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], True, f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
+        self.assertTrue(metrics_optimal["RMSE"] < metrics_default["RMSE"], f"Expected {metrics_optimal['RMSE']} > {metrics_default['RMSE']}")
         self.assertTrue(yi > 0, True)