13. adding of eeg dataset with bayesian tests

imputegap/imputation/imputation.py

@@ -12,7 +12,6 @@
 
 class Imputation:
 
-    @classmethod
     def load_parameters(query: str = "default", algorithm: str = "cdrec"):
         """
         Load default values of algorithms
@@ -34,8 +33,6 @@ 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)
 
@@ -111,7 +108,7 @@ def cdrec(ground_truth, contamination, params=None):
             if params is not None:
                 truncation_rank, epsilon, iterations = params
             else:
-                truncation_rank, epsilon, iterations = Imputation.load_parameters(algorithm="cdrec")
+                truncation_rank, epsilon, iterations = Imputation.load_parameters(query="default", algorithm="cdrec")
 
             imputed_matrix = cdrec(contamination=contamination, truncation_rank=truncation_rank, iterations=iterations, epsilon=epsilon)
 
@@ -168,7 +165,7 @@ def iim_imputation(ground_truth, contamination, params=None):
             if params is not None:
                 neighbors, algo_code = params
             else:
-                neighbors, algo_code = Imputation.load_parameters(algorithm="iim")
+                neighbors, algo_code = Imputation.load_parameters(query="default", algorithm="iim")
 
             imputed_matrix = iim(contamination=contamination, number_neighbor=neighbors, algo_code=algo_code)
 
@@ -191,7 +188,7 @@ def mrnn_imputation(ground_truth, contamination, params=None):
             if params is not None:
                 hidden_dim, learning_rate, iterations, sequence_length = params
             else:
-                hidden_dim, learning_rate, iterations, sequence_length = Imputation.load_parameters(algorithm="mrnn")
+                hidden_dim, learning_rate, iterations, sequence_length = Imputation.load_parameters(query="default", algorithm="mrnn")
 
             imputed_matrix = mrnn(contamination=contamination, hidden_dim=hidden_dim, learning_rate=learning_rate, iterations=iterations, sequence_length=sequence_length)
 
@@ -216,7 +213,7 @@ def stmvl_imputation(ground_truth, contamination, params=None):
             if params is not None:
                 window_size, gamma, alpha = params
             else:
-                window_size, gamma, alpha = Imputation.load_parameters(algorithm="stmvl")
+                window_size, gamma, alpha = Imputation.load_parameters(query="default", algorithm="stmvl")
 
             imputed_matrix = stmvl(contamination=contamination, window_size=window_size, gamma=gamma, alpha=alpha)
 

imputegap/manager/utils.py

@@ -0,0 +1,28 @@
+import os
+
+
+def get_file_path_dataset(set_name="test"):
+    """
+    Find the accurate path for loading files of tests
+    :return: correct file paths
+    """
+
+    filepath = "../imputegap/dataset/" + set_name + ".txt"
+
+    if not os.path.exists(filepath):
+        filepath = filepath[1:]
+
+    return filepath
+
+
+def get_save_path_asset():
+    """
+    Find the accurate path for saving files of tests
+    :return: correct file paths
+    """
+    filepath = "../tests/assets"
+
+    if not os.path.exists(filepath):
+        filepath = filepath[1:]
+
+    return filepath

imputegap/runner_optimization.py

@@ -37,11 +37,11 @@ def check_block_size(filename):
     gap.print()
     gap.plot(ts_type="contamination", title="test", save_path="assets", limitation=3, display=False)
 
-    optimal_params, yi = Optimization.Bayesian.bayesian_optimization(ground_truth=gap.ts, contamination=gap.ts_contaminate, algorithm="iim")
-
-    print("\nOptical Params : ", optimal_params)
-    print("\nyi : ", yi, "\n")
-
-    Optimization.save_optimization(optimal_params=optimal_params, algorithm="cdrec")
-
-    print("\n", "_"*95, "end")
+    for algo in ["cdrec", "stmvl", "iim", "mrnn"]:
+        print("RUN OPTIMIZATION FOR : ", algo, "...")
+        optimal_params, yi = Optimization.Bayesian.bayesian_optimization(ground_truth=gap.ts, contamination=gap.ts_contaminate, algorithm=algo)
+        print("\nOptical Params : ", optimal_params)
+        print("\nyi : ", yi, "\n")
+        Optimization.save_optimization(optimal_params=optimal_params, algorithm=algo)
+        print("\n", "_"*95, "end")
+    print("\n", "_" * 95, "end")