def generate_insights(filename,config_file=None,threshold=0.3):
print('Generating insights for the file ' + str(filename))
data_frame = load_data(filename)
unique_data,categorical_attributes,measure_attributes,categorical_index,timeseries_attributes,discarded_attributes = data_preprocessing(data_frame,config_file,threshold)
# Multi processing for correlation
# TODO: Change logic based on the number of cores for correlation
out_queue = Queue()
correlation_p = Process(target=generate_correlation_insights, args=(data_frame,measure_attributes,out_queue))
correlation_p.start()
tau = 2
top_k = 40
maxOneInsPerSub = True
result = ins.Insights(data_frame, tau, top_k, categorical_attributes, measure_attributes, timeseries_attributes, maxOneInsPerSub)
result = sorted(result, key=lambda elem: elem[0], reverse=True)
# print("\n")
# for r in result:
# print(r, "\n")
corr_result = out_queue.get()
if correlation_p.is_alive():
correlation_p.join()
print('Generating insight graph')
print("--- %s Minutes ---" % str(round(float((int(time.time()) - int(start_time)) /60), 2)))
generate_graphs(result,categorical_index,top_k,corr_result,filename)
print("--- %s Minutes ---" % str(round(float((int(time.time()) - int(start_time)) /60), 2)))
def main_10_cities():
# Create the TSP
tsp = create_tsp_problem(length=200, width=200, cities=10)
plot_tsp_grid(tsp, "Genetic_Algorithm")
# Random Hillclimbing
rhc_run_stats, rhc_run_curves = hillclimbing.solve_with_hillclimbing(
tsp, "RHC_TSP")
rhc_data_strings = {
'title': 'RHC - 10 Cities',
'Parameters': ['Restarts'],
'limit_time': 0,
'limit_iterations': 0
}
graphs.generate_graphs(rhc_run_stats, rhc_run_curves, rhc_data_strings)
# Mimic
mimic_run_stats, mimic_run_curves = mimic.solve_with_mimic(
tsp, "MIMIC_TSP")
mimic_data_strings = {
'title': 'MIMIC - 10 Cities',
'Parameters': ['Population Size', 'Keep Percent'],
'limit_time': 5,
'limit_iterations': 100
}
graphs.generate_graphs(mimic_run_stats, mimic_run_curves,
mimic_data_strings)
# Solve with Genetic Algorithm
ga_run_stats, ga_run_curves = genetic_algorithm.solve_with_ga(
tsp, "GA_TSP")
ga_data_strings = {
'title': 'Genetic Algorithms - 10 Cities',
'Parameters': ['Mutation Rate', 'Population Size'],
'limit_time': 1,
'limit_iterations': 800
}
graphs.generate_graphs(ga_run_stats, ga_run_curves, ga_data_strings)
# Simulated Annealing
sa_run_stats, sa_run_curves = simulated_annealing.solve_with_sim_annealing(
tsp, "SA_TSP")
sa_data_strings = {
'title': 'Simulated Annealing - 10 Cities',
'Parameters': ['Temperature'],
'limit_time': 0.4,
'limit_iterations': 1000
}
graphs.generate_graphs(sa_run_stats, sa_run_curves, sa_data_strings)
def main_20_items():
t_pct = 0.1
knapsack = create_knapsack_problem(20)
# Random Hillclimbing
rhc_run_stats, rhc_run_curves = hillclimbing.solve_with_hillclimbing(
knapsack, "RHC_Knapsack")
rhc_data_strings = {
'title': 'RHC - Knapsack',
'Parameters': ['Restarts'],
'limit_time': 0,
'limit_iterations': 0
}
graphs.generate_graphs(rhc_run_stats, rhc_run_curves, rhc_data_strings)
# Mimic
mimic_run_stats, mimic_run_curves = mimic.solve_with_mimic(
knapsack, "MIMIC_Knapsack")
mimic_data_strings = {
'title': 'MIMIC - Knapsack',
'Parameters': ['Population Size', 'Keep Percent'],
'limit_time': 10,
'limit_iterations': 50
}
graphs.generate_graphs(mimic_run_stats, mimic_run_curves,
mimic_data_strings)
# Solve with Genetic Algorithm
ga_run_stats, ga_run_curves = genetic_algorithm.solve_with_ga(
knapsack, "GA_Knapsack")
ga_data_strings = {
'title': 'Genetic Algorithms - Knapsack',
'Parameters': ['Mutation Rate', 'Population Size'],
'limit_time': 11,
'limit_iterations': 800
}
graphs.generate_graphs(ga_run_stats, ga_run_curves, ga_data_strings)
# Simulated Annealing
sa_run_stats, sa_run_curves = simulated_annealing.solve_with_sim_annealing(
knapsack, "SA_Knapsack")
sa_data_strings = {
'title': 'Simulated Annealing - Knapsack',
'Parameters': ['Temperature'],
'limit_time': 0.5,
'limit_iterations': 1500
}
graphs.generate_graphs(sa_run_stats, sa_run_curves, sa_data_strings)
def main_20_items():
t_pct = 0.1
six_peaks = create_six_peeks_problem(20, 0.1)
# Random Hillclimbing
rhc_run_stats, rhc_run_curves = hillclimbing.solve_with_hillclimbing(six_peaks, "RHC_6P")
rhc_data_strings = {
'title': 'RHC - 6 Peaks',
'Parameters': ['Restarts'],
'limit_time': 0,
'limit_iterations': 0
}
graphs.generate_graphs(rhc_run_stats, rhc_run_curves, rhc_data_strings)
# Mimic
mimic_run_stats, mimic_run_curves = mimic.solve_with_mimic(six_peaks, "MIMIC_6P")
mimic_data_strings = {
'title': 'MIMIC - 6 Peaks',
'Parameters': ['Population Size', 'Keep Percent'],
'limit_time': 10,
'limit_iterations': 100
}
graphs.generate_graphs(mimic_run_stats, mimic_run_curves, mimic_data_strings)
# Solve with Genetic Algorithm
ga_run_stats, ga_run_curves = genetic_algorithm.solve_with_ga(six_peaks, "GA_6P")
ga_data_strings = {
'title': 'Genetic Algorithms - 6 Peaks',
'Parameters': ['Mutation Rate', 'Population Size'],
'limit_time': 0.35,
'limit_iterations': 100
}
graphs.generate_graphs(ga_run_stats, ga_run_curves, ga_data_strings)
# Simulated Annealing
sa_run_stats, sa_run_curves = simulated_annealing.solve_with_sim_annealing(six_peaks, "SA_6P")
sa_data_strings = {
'title': 'Simulated Annealing - 6 Peaks',
'Parameters': ['Temperature'],
'limit_time': 0.3,
'limit_iterations': 1500
}
graphs.generate_graphs(sa_run_stats, sa_run_curves, sa_data_strings)