#bsub < job.bsub import numpy as np from sklearn.cluster import AffinityPropagation from sklearn import cluster import os import threading from job_thread_executioner import ThreadExecutioner from job_basic import createFolders from job_basic import getParameters #this is for windows: #os.chdir("C:/MiCluster.Test/") methodName = "affinity_propagation" dataset, thread_limit, rounds = getParameters() executioner = ThreadExecutioner(thread_limit) createFolders(methodName, dataset) #for multi-threading def worker(X, damping): method = AffinityPropagation(damping=damping) method.fit(X) key = methodName + "/length_" + length + "/" + deg + "/individuals/affinity_propagation_damping_" + str( damping) np.savetxt(key + "_labels.csv", method.labels_, fmt="%d") X = np.loadtxt("cs_datasets/" + dataset + ".csv", delimiter=",") for damping in [0.5, 0.6, 0.7, 0.8, 0.9]:
#thread_limit = 10 #length_reactivity = "_21" #deg = "wt" #k = 3 #init = 'random' thread_limit = 10 rounds = 100 if (len(sys.argv) > 1): thread_limit = int(sys.argv[1].strip()) if (len(sys.argv) > 2): rounds = int(sys.argv[2].strip()) executioner = ThreadExecutioner(thread_limit) methodName = "kmeans" #this is for multithreading def worker(length_reactivity, deg, X, k, init, round): """multithreading worker""" method = KMeans(k, init=init) method.fit(X) key = methodName + "/length" + length_reactivity + "/" + deg + "/individuals/kmeans_k_" + str( k) + "_init_" + init + "_round_" + str(round) np.savetxt(key + "_labels.csv", method.labels_, fmt="%d") #create folders
import os import numpy as np from scipy.cluster import hierarchy from job_thread_executioner import ThreadExecutioner from job_basic import createFolders from job_basic import getParameters os.chdir("C:\\Icas.Test\\") methodName = "hierarchical" dataset, thread_limit, rounds = getParameters() executioner = ThreadExecutioner(5) createFolders(methodName, dataset) #thread_limit = 10 #rounds = 100 #dataset = "cs_rna_distance_triangle_wt_71" upper_triangle = np.loadtxt("cs_datasets/" + dataset + ".csv", delimiter=",") hierarchy_result = hierarchy.linkage(upper_triangle) for k in range(3, 15): print "k=" + str(k) cutree = hierarchy.cut_tree(hierarchy_result, n_clusters=[k]) alist = [] for i in range(0, len(cutree)):