] # TSSPG] + chipPG # +cagePG+metPG+chipPG # Generate train train_chrs = [] [train_chrs.append("chr" + chr) for chr in chr_nums] if write_all_chrms_in_file: train_file_name = "training.RandOn" + str(params) params.out_file = output_folder + "_".join( train_chrs) + train_file_name for trainChrName in train_chrs: training_file_name = "training.RandOn" + trainChrName + str( params) + ".txt" # set it if you want to use all contacts of chromosome for training: # params.sample_size = len(params.contacts_reader.data[trainChrName]) # if you want to use only an interval of chromosome, set its coordinates: params.interval = Interval( trainChrName, params.contacts_reader.get_min_contact_position(trainChrName), params.contacts_reader.get_max_contact_position(trainChrName)) if not write_all_chrms_in_file: train_file_name = "training.RandOn" + str(params) + ".txt" params.out_file = output_folder + params.interval.toFileName( ) + train_file_name generate_data(params, saveFileDescription=True) if not write_all_chrms_in_file: del (params.out_file) del (params.sample_size)
# # Interval("chr10",36000000,41000000), # # Interval("chr1", 100000000, 110000000)]: # params.interval = interval validate_chrs = ["chr19", "chrX"] for validateChrName in validate_chrs: params.sample_size = len( params.contacts_reader.data[validateChrName]) #print(params.sample_size) validation_file_name = "validatingOrient." + str(params) + ".txt" params.interval = Interval( validateChrName, params.contacts_reader.get_min_contact_position( validateChrName), params.contacts_reader.get_max_contact_position( validateChrName)) logging.getLogger( __name__).info("Generating validation dataset for interval " + str(params.interval)) params.out_file = output_folder + params.interval.toFileName( ) + validation_file_name generate_data(params) del (params.out_file) del (params.sample_size) # for object in [params.contacts_reader]+params.pgs: # lostInterval = Interval("chr1",103842568,104979840) # object.delete_region(lostInterval) # params.interval = Interval("chr1",100000000,109000000) # logging.getLogger(__name__).info("Saving data to file "+params.interval.toFileName() + "DEL." + lostInterval.toFileName()+validation_file_name) # params.out_file = params.interval.toFileName() + "DEL." + lostInterval.toFileName()+validation_file_name # generate_data(params)
# Training parapmeters data_batch_size = 32 mask_batch_size = 32 # final batch_size is data_batch_size x mask_batch_size s = 5 # size of optimal subset that we are looking for s_p = 2 # number of flipped bits in a mask when looking around m_opt phase_2_start = 6000 # after how many batches phase 2 will begin max_batches = 15000 # how many batches if the early stopping condition not satisfied early_stopping_patience = 600 # how many patience batches (after phase 2 starts) # before the training stops # Generate data for XOR dataset: # First three features are used to create the target (y) # All the following features are gaussian noise # In total 10 features X_tr, y_tr = generate_data(n=N_TRAIN_SAMPLES, seed=0) X_val, y_val = generate_data(n=N_VAL_SAMPLES, seed=0) X_te, y_te = generate_data(n=N_TEST_SAMPLES, seed=0) # Get one hot encoding of the labels y_tr = get_one_hot(y_tr.astype(np.int8), 4) y_te = get_one_hot(y_te.astype(np.int8), 4) y_val = get_one_hot(y_val.astype(np.int8), 4) # Create the framework, needs number of features and batch_sizes, str_id for tensorboard fs = FeatureSelector(FEATURE_SHAPE, s, data_batch_size, mask_batch_size, str_id=dataset_label)
import pickle import matplotlib.pyplot as plt import seaborn as sns from DataGenerator import generate_data from ShapeGenerator import generate_images for i in range(1, 5): file_name = 'shapes_' + str(i) generate_images(file_name) generate_data(file_name) for i in range(1, 501): sns.boxplot(x='Standard Deviation', y='Sørensen–Dice Coefficient', hue='Method', data=pickle.load( open('Data/shapes_' + str(i) + '_data.pkl', 'rb'))) plt.show()
else: #print("Route flow " + str(i) + " through " + str(r_star)) pie[i] = fl[i] - (fl[i] * k_r) throughput = throughput + fl_e[i] for edge in ce: if edge in gre[ind]: lamb[edge] = lamb[edge] + lamb[edge]*float(fl[i]*gre[ind][edge]/ce[edge]) + float(fl[i]*gre[ind][edge]/(chi*ce[edge])) for mbox in pm: if mbox in qrm[ind]: theta[mbox] = theta[mbox] + theta[mbox] * float(fl[i] * qrm[ind][mbox] / pm[mbox]) + float(fl[i] * qrm[ind][ mbox] / (chi * pm[mbox])) print("Throughput: " + str(throughput)) if __name__ == '__main__': rls, gre, ce, fl, pm, qrm, fl_e, fl_pm = generate_data() # print(rls) # print(gre) # print(ce) # print(fl) # print(pm) # print(qrm) pda(rls, gre, ce, fl, pm, qrm, fl_e) sum = 0 for i in range(0, len(fl_e)): sum = sum + fl_e[i] print("Maximum Throughtput: " + str(sum))