while dataset_test: x, y = util.line_toseq(dataset_test.pop(0), charstop) if dense: dataset.append(util.seq_to_densevec(x, y, vdict)) else: dataset.append(util.seq_to_sparsevec(x,y,charset)) if not len(dataset_test)%1000: print "len(dataset_test)", len(dataset_test) dataset_test = dataset #This is the final test. print "This is the test for the PEAK value." mylstm.load(modelname + "/saving-" + str(peak)) testcost, act, aco, atp, p, r, f = mylstm.test(dataset_test) print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", act, "Total in Output:", aco, "True Positive:", atp, "Loss:", testcost print "\t@@VALIDATE ON TEST@@\tP, R, F:", p, r, f print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", datetime.datetime.now(), datetime.datetime.now()-starttime print "\tpeak =", peak generate = [line for line in util.file_to_lines(glob.glob('qualitative/allover-sjw-gold*'))] dataset_generate = [] for line in generate: x, y = util.line_toseq(line, charstop) if dense: dataset_generate.append(util.seq_to_densevec(x, y, vdict)) else: dataset_generate.append(util.seq_to_sparsevec(x,y,charset)) if not len(dataset_generate)%1000: print "len(dataset_generate)", len(dataset_generate) generate = [util.line_toraw(line) for line in generate] result = util.decode_totext(generate, mylstm.generate(dataset_generate), charstop) for line in result: print line.encode('utf8')
# -*- coding: utf8 -*- import util import sys import glob material = '../data/24s/*' #material = 'data/24s/*' i = 0 for line in util.file_to_lines(glob.glob(material)): #i = i+1 #print (" ".join(util.line_toraw(line)).encode('utf8')) print(" ".join(util.line_toraw(line))) #print i
else: dataset.append(util.seq_to_sparsevec(x,y,charset)) print "len(dataset_test)", len(dataset_test) dataset_test = dataset #This is the final test. print "This is the test for the PEAK value." mylstm.load(modelname + "saving" + str(peak)) testcost, act, aco, atp, p, r, f = mylstm.test(dataset_test) print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", act, "Total in Output:", aco, "True Positive:", atp, "Loss:", testcost print "\t@@VALIDATE ON TEST@@\tP, R, F:", p, r, f print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", datetime.datetime.now(), datetime.datetime.now()-starttime print "\tpeak =", peak generate = [line for line in util.file_to_lines(glob.glob('qualitative/allover-sjw-gold.*'))] dataset_generate = [] while li: x, y = util.line_toseq(generate.pop(0), charstop) if dense: dataset_generate.append(util.seq_to_densevec(x, y, vdict)) else: dataset_generate.append(util.seq_to_sparsevec(x,y,charset)) print "len(dataset_generate)", len(dataset_generate) li_generate = [util.line_toraw(line) for line in dataset_generate] result = util.decode_totext(li_generate, mylstm.generate(dataset_generate), charstop) for line in result: print line.encode('utf8')
dataset_test = dataset #This is the final test. print "This is the test for the PEAK value." mylstm.load(modelname + "/saving-" + str(peak)) testcost, act, aco, atp, p, r, f = mylstm.test(dataset_test) print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", act, "Total in Output:", aco, "True Positive:", atp, "Loss:", testcost print "\t@@VALIDATE ON TEST@@\tP, R, F:", p, r, f print "\t@@VALIDATE ON TEST@@\tTotal in Gold:", datetime.datetime.now( ), datetime.datetime.now() - starttime print "\tpeak =", peak generate = [ line for line in util.file_to_lines(glob.glob('qualitative/allover-sjw-gold*')) ] dataset_generate = [] for line in generate: x, y = util.line_toseq(line, charstop) if dense: dataset_generate.append(util.seq_to_densevec(x, y, vdict)) else: dataset_generate.append(util.seq_to_sparsevec(x, y, charset)) if not len(dataset_generate) % 1000: print "len(dataset_generate)", len(dataset_generate) generate = [util.line_toraw(line) for line in generate] result = util.decode_totext(generate, mylstm.generate(dataset_generate), charstop) for line in result: print line.encode('utf8')
# -*- coding: utf8 -*- import util import sys import glob material = '../data/24s/*' #material = 'data/24s/*' i=0 for line in util.file_to_lines(glob.glob(material)): #i = i+1 print " ".join(util.line_toraw(line)).encode('utf8') #print i