Merge pull request #45 from luinardi/intermediate_doe_write

Write to csv as soon as the black-box is evaluated.

hypermapper/bo.py

@@ -1,5 +1,4 @@
 import copy
-import csv
 import datetime
 import os
 import random
@@ -23,6 +22,9 @@
         compute_data_array_scalarization,
         get_single_configuration,
         sample_weight_flat,
+        get_output_data_file,
+        create_output_data_file,
+        write_data_array,
     )
 
 except ImportError:
@@ -68,10 +70,13 @@
         compute_data_array_scalarization,
         get_single_configuration,
         sample_weight_flat,
+        get_output_data_file,
+        create_output_data_file,
+        write_data_array,
     )
 
 
-def main(config, black_box_function=None, output_file="", profiling=None):
+def main(config, black_box_function=None, profiling=None):
     """
     Run design-space exploration using bayesian optimization.
     :param config: dictionary containing all the configuration parameters of this optimization.
@@ -98,6 +103,9 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     evaluations_per_optimization_iteration = config[
         "evaluations_per_optimization_iteration"
     ]
+    output_data_file = get_output_data_file(
+        config["output_data_file"], run_directory, application_name
+    )
     batch_mode = evaluations_per_optimization_iteration > 1
     number_of_cpus = config["number_of_cpus"]
     print_importances = config["print_parameter_importance"]
@@ -166,13 +174,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
         for objective in optimization_metrics:
             objective_limits[objective] = [float("inf"), float("-inf")]
 
-    if output_file == "":
-        output_data_file = config["output_data_file"]
-        if output_data_file == "output_samples.csv":
-            output_data_file = application_name + "_" + output_data_file
-    else:
-        output_data_file = output_file
-
     exhaustive_search_data_array = None
     exhaustive_search_fast_addressing_of_data_array = None
     if hypermapper_mode == "exhaustive":
@@ -251,6 +252,11 @@ def main(config, black_box_function=None, output_file="", profiling=None):
             % absolute_configuration_index
         )
 
+    create_output_data_file(
+        output_data_file, param_space.get_input_output_and_timestamp_parameters()
+    )
+    if data_array:  # if it is not empty
+        write_data_array(param_space, data_array, output_data_file)
     ### DoE phase
     if absolute_configuration_index < number_of_doe_samples:
         configurations = []
@@ -278,6 +284,7 @@ def main(config, black_box_function=None, output_file="", profiling=None):
             hypermapper_mode,
             configurations,
             beginning_of_time,
+            output_data_file,
             black_box_function,
             exhaustive_search_data_array,
             exhaustive_search_fast_addressing_of_data_array,
@@ -312,6 +319,7 @@ def main(config, black_box_function=None, output_file="", profiling=None):
                 hypermapper_mode,
                 configurations,
                 beginning_of_time,
+                output_data_file,
                 black_box_function,
                 exhaustive_search_data_array,
                 exhaustive_search_fast_addressing_of_data_array,
@@ -326,20 +334,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
             absolute_configuration_index += 1
             optimization_iterations -= 1
 
-    # Create output file with explored configurations from resumed run and DoE
-    with open(
-        deal_with_relative_and_absolute_path(run_directory, output_data_file), "w"
-    ) as f:
-        w = csv.writer(f)
-        w.writerow(param_space.get_input_output_and_timestamp_parameters())
-        tmp_list = [
-            param_space.convert_types_to_string(j, data_array)
-            for j in param_space.get_input_output_and_timestamp_parameters()
-        ]
-        tmp_list = list(zip(*tmp_list))
-        for i in range(len(data_array[optimization_metrics[0]])):
-            w.writerow(tmp_list[i])
-
     for objective in optimization_metrics:
         lower_bound = min(objective_limits[objective][0], min(data_array[objective]))
         upper_bound = max(objective_limits[objective][1], max(data_array[objective]))
@@ -486,6 +480,7 @@ def main(config, black_box_function=None, output_file="", profiling=None):
                 hypermapper_mode,
                 configurations,
                 beginning_of_time,
+                output_data_file,
                 black_box_function,
                 exhaustive_search_data_array,
                 exhaustive_search_fast_addressing_of_data_array,
@@ -520,21 +515,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
                     for header in configuration:
                         data_array[header].append(configuration[header])
 
-            # and save results
-            with open(
-                deal_with_relative_and_absolute_path(run_directory, output_data_file),
-                "a",
-            ) as f:
-                w = csv.writer(f)
-                tmp_list = [
-                    param_space.convert_types_to_string(j, new_data_array)
-                    for j in list(
-                        param_space.get_input_output_and_timestamp_parameters()
-                    )
-                ]
-                tmp_list = list(zip(*tmp_list))
-                for i in range(len(new_data_array[optimization_metrics[0]])):
-                    w.writerow(tmp_list[i])
             configurations = []
         else:
             # If we have not selected all points in the batch yet, add the model prediction as a 'liar'

hypermapper/evolution.py

@@ -2,7 +2,6 @@
 # This script implements a simplification of the evolutionary process proposed by Real et al.: https://arxiv.org/abs/1802.01548v7#
 ###############################################################################################################################
 import copy
-import csv
 import datetime
 import os
 import sys
@@ -20,6 +19,8 @@
         get_single_configuration,
         concatenate_list_of_dictionaries,
         deal_with_relative_and_absolute_path,
+        get_output_data_file,
+        create_output_data_file,
     )
 except ImportError:
     if os.getenv("HYPERMAPPER_HOME"):  # noqa
@@ -58,6 +59,8 @@
         get_single_configuration,
         concatenate_list_of_dictionaries,
         deal_with_relative_and_absolute_path,
+        get_output_data_file,
+        create_output_data_file,
     )
 
 
@@ -115,6 +118,7 @@ def run_objective_function(
     hypermapper_mode,
     param_space,
     beginning_of_time,
+    output_data_file,
     run_directory,
     evolution_data_array,
     fast_addressing_of_data_array,
@@ -172,6 +176,7 @@ def run_objective_function(
             hypermapper_mode,
             new_configurations,
             beginning_of_time,
+            output_data_file,
             black_box_function,
             run_directory=run_directory,
         )
@@ -385,6 +390,9 @@ def main(config, black_box_function=None, output_file="", profiling=None):
             print("Error: the black box function parameter is not callable")
             raise SystemExit
 
+    output_data_file = get_output_data_file(
+        config["output_data_file"], run_directory, application_name
+    )
     optimization_metrics = config["optimization_objectives"]
     number_of_objectives = len(optimization_metrics)
     if number_of_objectives != 1:
@@ -421,13 +429,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     if hypermapper_mode == "client-server":
         sys.stdout.switch_log_only_on_file(True)
 
-    if output_file == "":
-        output_data_file = config["output_data_file"]
-        if output_data_file == "output_samples.csv":
-            output_data_file = application_name + "_" + output_data_file
-    else:
-        output_data_file = output_file
-
     absolute_configuration_index = 0
     fast_addressing_of_data_array = {}
     evolution_fast_addressing_of_data_array = {}
@@ -440,12 +441,17 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     optimization_function_parameters["param_space"] = param_space
     optimization_function_parameters["beginning_of_time"] = beginning_of_time
     optimization_function_parameters["run_directory"] = run_directory
+    optimization_function_parameters["output_data_file"] = output_data_file
     optimization_function_parameters["black_box_function"] = black_box_function
     optimization_function_parameters["evolution_data_array"] = evolution_data_array
     optimization_function_parameters[
         "fast_addressing_of_data_array"
     ] = evolution_fast_addressing_of_data_array
 
+    create_output_data_file(
+        output_data_file, param_space.get_input_output_and_timestamp_parameters()
+    )
+
     print("Starting evolution...")
     evolution_t0 = datetime.datetime.now()
     all_samples = evolution(
@@ -474,18 +480,5 @@ def main(config, black_box_function=None, output_file="", profiling=None):
         )
     )
 
-    with open(
-        deal_with_relative_and_absolute_path(run_directory, output_data_file), "w"
-    ) as f:
-        w = csv.writer(f)
-        w.writerow(list(evolution_data_array.keys()))
-        tmp_list = [
-            param_space.convert_types_to_string(j, evolution_data_array)
-            for j in list(evolution_data_array.keys())
-        ]
-        tmp_list = list(zip(*tmp_list))
-        for i in range(len(evolution_data_array[optimization_metrics[0]])):
-            w.writerow(tmp_list[i])
-
     print("### End of the evolutionary search")
     return evolution_data_array

hypermapper/local_search.py

@@ -1,5 +1,4 @@
 import copy
-import csv
 import datetime
 import os
 import sys
@@ -28,6 +27,8 @@
         dict_of_lists_to_numpy,
         get_min_configurations,
         get_min_feasible_configurations,
+        get_output_data_file,
+        create_output_data_file,
     )
 except ImportError:
     if os.getenv("HYPERMAPPER_HOME"):  # noqa
@@ -74,6 +75,8 @@
         dict_of_lists_to_numpy,
         get_min_configurations,
         get_min_feasible_configurations,
+        get_output_data_file,
+        create_output_data_file,
     )
 
 
@@ -163,6 +166,7 @@ def run_objective_function(
     hypermapper_mode,
     param_space,
     beginning_of_time,
+    output_data_file,
     run_directory,
     local_search_data_array,
     fast_addressing_of_data_array,
@@ -227,6 +231,7 @@ def run_objective_function(
             hypermapper_mode,
             new_configurations,
             beginning_of_time,
+            output_data_file,
             black_box_function,
             exhaustive_search_data_array,
             exhaustive_search_fast_addressing_of_data_array,
@@ -807,7 +812,7 @@ def parallel_multistart_local_search(
     return data_array, best_configuration
 
 
-def main(config, black_box_function=None, output_file="", profiling=None):
+def main(config, black_box_function=None, profiling=None):
     """
     Run design-space exploration using random scalarizations.
     :param config: dictionary containing all the configuration parameters of this design-space exploration.
@@ -819,8 +824,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     run_directory = config["run_directory"]
     application_name = config["application_name"]
     hypermapper_mode = config["hypermapper_mode"]["mode"]
-    noise = config["noise"]
-
     if hypermapper_mode == "default":
         if black_box_function == None:
             print("Error: the black box function must be provided")
@@ -829,6 +832,10 @@ def main(config, black_box_function=None, output_file="", profiling=None):
             print("Error: the black box function parameter is not callable")
             raise SystemExit
 
+    noise = config["noise"]
+    output_data_file = get_output_data_file(
+        config["output_data_file"], run_directory, application_name
+    )
     optimization_metrics = config["optimization_objectives"]
     number_of_objectives = len(optimization_metrics)
     # local search will not produce reasonable output if run in parallel - it is therefore disabled
@@ -895,13 +902,6 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     if hypermapper_mode == "client-server":
         sys.stdout.switch_log_only_on_file(True)
 
-    if output_file == "":
-        output_data_file = config["output_data_file"]
-        if output_data_file == "output_samples.csv":
-            output_data_file = application_name + "_" + output_data_file
-    else:
-        output_data_file = output_file
-
     absolute_configuration_index = 0
     fast_addressing_of_data_array = {}
     local_search_fast_addressing_of_data_array = {}
@@ -914,6 +914,7 @@ def main(config, black_box_function=None, output_file="", profiling=None):
     optimization_function_parameters["param_space"] = param_space
     optimization_function_parameters["beginning_of_time"] = beginning_of_time
     optimization_function_parameters["run_directory"] = run_directory
+    optimization_function_parameters["output_data_file"] = output_data_file
     optimization_function_parameters[
         "exhaustive_search_data_array"
     ] = exhaustive_search_data_array
@@ -936,6 +937,10 @@ def main(config, black_box_function=None, output_file="", profiling=None):
         "enable_feasible_predictor"
     ] = enable_feasible_predictor
 
+    create_output_data_file(
+        output_data_file, param_space.get_input_output_and_timestamp_parameters()
+    )
+
     print("Starting local search...")
     local_search_t0 = datetime.datetime.now()
     all_samples, best_configuration = local_search(
@@ -957,18 +962,5 @@ def main(config, black_box_function=None, output_file="", profiling=None):
         % (len(local_search_data_array[optimization_metrics[0]]))
     )
 
-    with open(
-        deal_with_relative_and_absolute_path(run_directory, output_data_file), "w"
-    ) as f:
-        w = csv.writer(f)
-        w.writerow(list(local_search_data_array.keys()))
-        tmp_list = [
-            param_space.convert_types_to_string(j, local_search_data_array)
-            for j in list(local_search_data_array.keys())
-        ]
-        tmp_list = list(zip(*tmp_list))
-        for i in range(len(local_search_data_array[optimization_metrics[0]])):
-            w.writerow(tmp_list[i])
-
     print("### End of the local search.")
     return local_search_data_array