def onlyForFirstIncrement(self, local_min_sup):
previous_threshold = UserDefined.min_sup
previous_data_set = Variable.size_of_dataset
# set the min_sup by given local min sup
UserDefined.min_sup = local_min_sup
Variable.size_of_dataset = len(ProgramVariable.uSDB)
tmp_root_node, tot_candidates = FUWSequence().douWSequence()
self.cur_ls_trie = Trie(tmp_root_node)
# load and update upto threshold and the size of the dataset
for i in range(0, len(ProgramVariable.uSDB)):
self.fssfs_trie.update_support(self.fssfs_trie.root_node, None, 0.0, 0, i)
previous_data_set += len(ProgramVariable.uSDB)
ProgramVariable.uSDB = ProgramVariable.pre_uSDB
for i in range(0, len(ProgramVariable.uSDB)):
self.cur_ls_trie.update_support(self.cur_ls_trie.root_node, None, 0.0, 0, i)
# Merging tries
self.fssfs_trie.merge_ls_with_fssfs_trie(self.cur_ls_trie.root_node, self.fssfs_trie.root_node)
self.cur_ls_trie = None
# update tries
self.fssfs_trie.traverse_trie(self.fssfs_trie.root_node)
self.fssfs_trie.update_trie(self.fssfs_trie.root_node)
UserDefined.min_sup = previous_threshold
Variable.size_of_dataset = previous_data_set
Variable.WAM = ProgramVariable.pre_upto_wSum/ProgramVariable.pre_upto_sum
# writing tries to file
self.fssfs_trie.trie_into_file(self.fssfs_trie.root_node, '')
return
def uWSIncPlusMethod(self, local_min_sup):
# store upto threshold and the size of the dataset
previous_threshold = UserDefined.min_sup
previous_data_set = Variable.size_of_dataset
# set the min_sup by given local min sup
UserDefined.min_sup = local_min_sup
Variable.size_of_dataset = len(ProgramVariable.uSDB)
tmp_root_node, tot_candidates = FUWSequence().douWSequence()
self.cur_ls_trie = Trie(tmp_root_node)
self.cur_ls_trie.printFSSFS(self.cur_ls_trie.root_node, '')
for i in range(0, len(ProgramVariable.uSDB)):
self.fssfs_trie.update_support(self.fssfs_trie.root_node, None, 0.0, 0, i)
# Merging tries
self.fssfs_trie.merge_ls_with_fssfs_trie(self.cur_ls_trie.root_node, self.fssfs_trie.root_node)
self.cur_ls_trie = None
# update tries
self.fssfs_trie.traverse_trie(self.fssfs_trie.root_node)
self.fssfs_trie.update_trie(self.fssfs_trie.root_node)
current_threshold = ThresholdCalculation.get_semi_wgt_exp_sup()
UserDefined.min_sup = previous_threshold
Variable.size_of_dataset = previous_data_set
Variable.size_of_dataset += len(ProgramVariable.uSDB)
Variable.WAM = ProgramVariable.pre_upto_wSum / ProgramVariable.pre_upto_sum
# writing tries to file
self.fssfs_trie.printPFS(self.fssfs_trie.root_node, '', ThresholdCalculation.get_semi_wgt_exp_sup(), current_threshold)
self.fssfs_trie.trie_into_file(self.fssfs_trie.root_node, '')
return
#WAM calculation && DataBase size update
WAMCalculation.update_WAM()
Variable.size_of_dataset = len(ProgramVariable.uSDB)
print(Variable.size_of_dataset, ' size of dataset')
print('WAM Done')
for mn_sup in min_sup_lst:
UserDefined.min_sup = mn_sup
for fct in wgt_fct_lst:
UserDefined.wgt_factor = fct
for mu in mu_lst:
Variable.mu = mu
start_time = time.time()
root_node, total_candidates = FUWSequence().douWSequence()
fssfs_trie = Trie(root_node)
fssfs_trie.update_trie(fssfs_trie.root_node)
total_patterns = fssfs_trie.trie_into_file(fssfs_trie.root_node, '')
FileInfo.fs.close()
FileInfo.sfs.close()
end_time = time.time()
print('FUWS: Total Candidates - ', total_candidates)
print('FUWS: Total Patterns - ', total_patterns)
print(start_time, end_time, end_time-start_time)
FileInfo.fs = open('initialFS_UWS.txt', 'w')
FileInfo.sfs = open('initialSFS_UWS.txt', 'w')
#WAM calculation && DataBase size update
WAMCalculation.update_WAM()
Variable.size_of_dataset += len(ProgramVariable.uSDB)
print(Variable.size_of_dataset, ' : Size of dataset')
ProgramVariable.uSDB += previous_usdb
ProgramVariable.pSDB += previous_psdb
print(len(ProgramVariable.pSDB),
' : check the value with previous one')
# print()
print('WAM: ', Variable.WAM)
print('Threshold: ', ThresholdCalculation.get_wgt_exp_sup(),
ThresholdCalculation.get_semi_wgt_exp_sup(), ' at uWSeq')
root_node = FUWSequence().douWSequence()
fssfs_trie = Trie(root_node)
fssfs_trie.update_trie(fssfs_trie.root_node)
fssfs_trie.trie_into_file(fssfs_trie.root_node, '')
FileInfo.fs.close()
FileInfo.sfs.close()
end_time = time.time()
FileInfo.time_info.write(str(end_time - previous_time))
FileInfo.time_info.write('\n')
previous_psdb = ProgramVariable.pSDB
previous_usdb = ProgramVariable.uSDB
print('Increment No', i)
FileInfo.fs = open(prefix_all + '/fs_naive' + str(i) + '.txt', 'r')
count_fs = 0
for seq in FileInfo.fs:
# we provided a file named as 'manual_weights.txt' in 'Files' folder
# In that file you can assign weight to each item manually. This is to inform you that
# file contains an item and its weight separated by an space in each line
# Don't forget to update accordingly if you wish
# WAM will be calculated && DataBase size will be been updated
WAMCalculation.update_WAM()
Variable.size_of_dataset = len(ProgramVariable.uSDB)
start_time = time.time()
root_node, total_candidates = FUWSequence().douWSequence() # To find the potential candidate patterns
fssfs_trie = Trie(root_node) # Patterns are stored into USeq-Trie
fssfs_trie.update_trie(fssfs_trie.root_node) # calculate actual WES and remove false patterns
fssfs_trie.trie_into_file(fssfs_trie.root_node, '') # write the desired patterns (FS & SFS) into files
FileInfo.fs.close()
FileInfo.sfs.close()
end_time = time.time()
print('Total required time(in Seconds): ', end_time-start_time)
PreProcess().doProcess()
# for seq in ProgramVariable.pSDB:
# print(seq, ' Print at Nothing')
print('Preprocess Done!')
# Weight Assigning
WeightAssign.assign(ProgramVariable.itemList)
# WeightAssign.manual_assign()
print('Weight Assign Done')
#WAM calculation && DataBase size update
WAMCalculation.update_WAM()
Variable.size_of_dataset = len(ProgramVariable.uSDB)
print(Variable.size_of_dataset, ' size of dataset')
print('WAM Done')
start_time = time.time()
root_node, total_candidates = UWSequence().douWSequence()
fssfs_trie = Trie(root_node)
# fssfs_trie.update_trie(fssfs_trie.root_node)
total_num_patterns = fssfs_trie.trie_into_file(fssfs_trie.root_node, '')
FileInfo.fs.close()
FileInfo.sfs.close()
end_time = time.time()
print('Total Candidates: ', total_candidates)
print('Total Patterns: ', total_num_patterns)
print('Total Times: ', start_time, end_time, end_time - start_time)
# You can change the file but make sure that you name the file exactly same as the given name
# wgt_assign_obj.manual_assign() # this is for manual way to assign weight
# we provided a file named as 'manual_weights.txt' in 'Files' folder
# In that file you can assign weight to each item manually. This is to inform you that
# file contains an item and its weight separated by an space in each line
# Don't forget to update accordingly if you wish
# WAM will be calculated && DataBase size will be initialized
WAMCalculation.update_WAM()
Variable.size_of_dataset = len(ProgramVariable.uSDB)
# Run FUWS algorithm to get FS and SFS from initial datasets and store store them into USeq-Trie
fsfss_trie_root_node, total_candidates = FUWSequence().douWSequence()
fsfss_trie = Trie(fsfss_trie_root_node)
fsfss_trie.update_trie(fsfss_trie.root_node)
fsfss_trie.trie_into_file(fsfss_trie.root_node, '')
cur_time = time.time()
FileInfo.time_info.write(str(cur_time - previous_time))
FileInfo.time_info.write('\n')
ls_trie = Trie(TrieNode(False, None, None, 0.0,
False)) # Initialize the trie to store LFS
FileInfo.fs.write('\n ----------- \n')
FileInfo.sfs.write('\n ----------- \n')
FileInfo.pfs.write('\n ----------- \n')
FileInfo.ls.write('\n ----------- \n')
# Dataset Preprocessing
PreProcess().doProcess()
WeightAssign.assign(ProgramVariable.itemList)
# WAM calculation && DataBase size update
WAMCalculation.upto_sum = 0
WAMCalculation.upto_wSum = 0
WAMCalculation.update_WAM()
Variable.size_of_dataset += len(ProgramVariable.uSDB)
ProgramVariable.uSDB += previous_usdb
ProgramVariable.pSDB += previous_psdb
root_node = UWSequence().douWSequence()
fssfs_trie = Trie(root_node)
fssfs_trie.update_trie(fssfs_trie.root_node)
fssfs_trie.trie_into_file(fssfs_trie.root_node, '')
FileInfo.fs.close()
FileInfo.sfs.close()
cnt = 0
f = open(
input_folder + '/result/fs_naive' + str(i) + '.txt',
'r')
for line in f:
if line == '\n' or line.strip() == '':
break
cnt += 1
patterns0.append(cnt)