def full_test(): # CREATE CHAR_NUM CHARACTERISTICS console.write_header("Creating Characteristics") characteristics = [] data.generate_characteristic(characteristics) # CREATE CLASS_NUM SYMBOL CLASSES console.write_header(" Creating Symbol Classes") symbolClasses = [] data.generate_symbol_classes(symbolClasses, characteristics) console.write_header("Computing Homogeneous Distortion") Distorter().create_homogeneus_cloud(symbolClasses) for c in range(0, util.global_variables.CLASS_NUM): console.write_header("Computing Cluster Evaluation") best_k = ps.cluster_evaluation(util.global_variables.MAX_K_CLUS_EVALUATION, symbolClasses[c:c+1]) util.global_variables.K = best_k[0] console.write_header("Computing Clusters with K:", str(util.global_variables.K)) Clusterer().computeClusters(symbolClasses[c:c+1]) console.write_header("Creating Non Homogeneous Foreign") foreignClassesNonHomo = f_creator.create_non_homogeneous_foreign(symbolClasses) console.write_header("Creating Homogeneous Foreign") foreignClassesHomo = f_creator.create_homogeneous_foreign(symbolClasses, characteristics) console.write_header(" Synthetic Data Calculations") synth_calc.ambiguity_for_different_radiuses(symbolClasses[:], foreignClassesHomo, foreignClassesNonHomo)
def grouping_assessment(): # CREATE CHAR_NUM CHARACTERISTICS console.write_header("Creating Characteristics") characteristics = [] data.generate_characteristic(characteristics) # CREATE CLASS_NUM SYMBOL CLASSES console.write_header(" Creating Symbol Classes") symbolClasses = [] data.generate_symbol_classes(symbolClasses, characteristics) console.write_header("Computing K cloud Distortion") #Distorter().create_k_clouds(util.global_variables.K_CLOUD_DISTORTION,symbolClasses) Distorter().create_non_homogeneus_cloud(symbolClasses) console.write_header("Computing Cluster Evaluation") ps.cluster_evaluation(util.global_variables.MAX_K_CLUS_EVALUATION,symbolClasses)
def synthetic_test_paper_2_old(): # CREATE CHAR_NUM CHARACTERISTICS console.write_header("Creating Characteristics") characteristics = [] data.generate_characteristic(characteristics) # CREATE CLASS_NUM SYMBOL CLASSES console.write_header(" Creating Symbol Classes") symbolClasses = [] data.generate_symbol_classes(symbolClasses, characteristics) # CREATE CLOUD DISTORTION IN NATIVE SET console.write_header("Computing K cloud Distortion") Distorter().create_cluster_assessment_cloud(util.global_variables.K_CLOUD_DISTORTION,symbolClasses) #Distorter().create_k_clouds(util.global_variables.K_CLOUD_DISTORTION,symbolClasses) Clusterer().computeClusters(symbolClasses[:]) Plot3D().renderPlot(symbolClasses) # COMPUTE CLUSTER EVALUATION console.write_header("Computing Cluster Evaluation") ps.cluster_evaluation(util.global_variables.MAX_K_CLUS_EVALUATION,symbolClasses)
def real_data(): console.write_header("Loading Native symbols") symbolClasses = loader.load_native_xls() console.write_header("Loading Foreign symbols") foreignClasses = loader.load_foreign_xls() util.global_variables.CLASS_NUM = len(symbolClasses) util.global_variables.CHAR_NUM = len(symbolClasses[0].learning_set[0].characteristicsValues) for c in range(0, util.global_variables.CLASS_NUM): console.write_header("Computing Cluster Evaluation") best_k = ps.cluster_evaluation(util.global_variables.MAX_K_CLUS_EVALUATION, symbolClasses[c:c+1]) util.global_variables.K = best_k[0] console.write_header("Computing Clusters with K:", str(util.global_variables.K)) Clusterer().computeClusters(symbolClasses[c:c+1]) console.write_header(" Synthetic Data Calculations") synth_calc.ambiguity_for_different_radiuses_real_data(symbolClasses[:], foreignClasses)
def semisynthetic_test_paper_2(): console.write_header("Loading Native symbols") symbolClasses = loader.load_native_xls() console.write_header("Loading Foreign symbols") foreignClasses = loader.load_foreign_xls() util.global_variables.CLASS_NUM = len(symbolClasses) util.global_variables.CHAR_NUM = len(symbolClasses[0].learning_set[0].characteristicsValues) # COMPUTE CLUSTER EVALUATION for c in range(0, util.global_variables.CLASS_NUM): console.write_header("Computing Cluster Evaluation") best_k = ps.cluster_evaluation(util.global_variables.MAX_K_CLUS_EVALUATION, symbolClasses[c:c+1]) util.global_variables.K = best_k[0] console.write_header("Computing Clusters with K:", str(util.global_variables.K)) Clusterer().computeClusters(symbolClasses[c:c+1]) paper2.compute(symbolClasses, foreignClasses)