-
Notifications
You must be signed in to change notification settings - Fork 0
/
preprocess.py
190 lines (155 loc) · 5.83 KB
/
preprocess.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
import io
import logging
import os
import subprocess
import sys
import time
import boto3
from botocore.exceptions import ClientError
FORMAT = f"%(filename)-13s|%(funcName)-10s|%(lineno)-3d|%(message)s"
logging.basicConfig(format=FORMAT)
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
"""
This script is used to preprocess raw data set, modify it if it can not parse raw data set correctly.
It is requested to run on a machine pre-installed metis(gpmetis), and python3.6+
"""
start_time = time.time()
def shell_command(command: str = 'ls'):
process = subprocess.run([command], capture_output=True, shell=True)
return process.stdout, process.stderr
def upload_file(file_name, bucket, object_name=None):
"""Upload a file to an S3 bucket
:param file_name: File to upload
:param bucket: Bucket to upload to
:param object_name: S3 object name. If not specified then file_name is used
:return: True if file was uploaded, else False
"""
# If S3 object_name was not specified, use file_name
if object_name is None:
object_name = file_name
# Upload the file
s3_client = boto3.client('s3')
try:
response = s3_client.upload_file(file_name, bucket, object_name)
except ClientError as e:
logger.error(e)
return False
return True
PEERS_NUM = 20 if len(sys.argv) == 1 else int(sys.argv[1])
print(f'PEERS_NUM = {PEERS_NUM}')
"""input"""
# delimiter = ' '
delimiter = '\t'
# input_files_fold = './facebook/'
input_files_fold = './'
# all_input_files_name = [path for path in os.listdir(input_files_fold) if path.endswith('.edges')]
all_input_files_name = ('soc-pokec-relationships.txt',)
print(f'input files: {all_input_files_name}')
"""output"""
tmp_undirected_graph = './tmp_metis'
tmp_metis_result = f'{tmp_undirected_graph}.part.{PEERS_NUM}'
result_dataset = 'soc-pokec_dataset'
result_r_dataset = f'r_{result_dataset}'
result_partition = f'{result_dataset}_partition_{PEERS_NUM}'
"""upload"""
bucket = 'pagerankdataset'
s3_key_dataset = result_dataset
s3_key_r_dataset = result_r_dataset
s3_key_partition = result_partition
"""start pre-process"""
all_raw_files_path = [os.path.join(input_files_fold, path) for path in all_input_files_name]
if os.path.isfile(tmp_metis_result):
os.remove(tmp_metis_result)
print(f'remove old "{tmp_metis_result}"')
if os.path.isfile(tmp_undirected_graph):
os.remove(tmp_undirected_graph)
print(f'remove old "{tmp_undirected_graph}"')
# aggregate all the contents from all raw files
contents = []
for file_path in all_raw_files_path:
with open(file_path, 'r') as raw_file:
contents += raw_file.readlines()
# find all the nodes
exist_nodes = set()
try:
for line in contents:
a, b = line.split(delimiter)
exist_nodes.add(int(a))
exist_nodes.add(int(b))
except Exception as e:
print(line)
raise e
# reassign continuous index for each node
node_mapping = dict(zip(exist_nodes, range(1, len(exist_nodes) + 1)))
matrix = [[] for _ in range(len(exist_nodes))]
r_matrix = [[] for _ in range(len(exist_nodes))]
tmp_matrix = [set() for _ in range(len(exist_nodes))]
# reformat raw data set to the format fit for metis
undirected_edges = set()
for line in contents:
a, b = line.split(delimiter)
a = node_mapping[int(a)]
b = node_mapping[int(b)]
matrix[a - 1].append(b)
r_matrix[b - 1].append(a)
# make relation undirected
tmp_matrix[a - 1].add(b)
undirected_edges.add((a, b))
tmp_matrix[b - 1].add(a)
undirected_edges.add((b, a))
# to fit input of distributed page rank
buffer = io.StringIO()
for edges in matrix:
buffer.write(' '.join(map(str, edges)))
buffer.write('\n')
with open(result_dataset, 'w') as data_set:
data_set.write(buffer.getvalue())
buffer.close()
buffer = io.StringIO()
for edges in r_matrix:
buffer.write(' '.join(map(str, edges)))
buffer.write('\n')
with open(result_r_dataset, 'w') as data_set:
data_set.write(buffer.getvalue())
buffer.close()
# to fit input format of metis
buffer = io.StringIO()
buffer.write(f'{len(exist_nodes)} {len(undirected_edges) // 2}\n')
for edges in tmp_matrix:
buffer.write(' '.join(map(str, edges)))
buffer.write('\n')
with open(tmp_undirected_graph, 'w') as data_set_for_metis:
data_set_for_metis.write(buffer.getvalue())
buffer.close()
# call metis to partition graph
stdout, stderr = shell_command('sudo find / -iname gpmetis')
metis_path = stdout.decode().strip()
print(f'{metis_path} {tmp_undirected_graph} {PEERS_NUM}')
stdout, stderr = shell_command(f'{metis_path} {tmp_undirected_graph} {PEERS_NUM}')
if stderr:
for line in stderr.decode().split('\n'):
print(line)
exit(-1)
for line in stdout.decode().split('\n'):
print(line)
print(f'wait for {tmp_metis_result}')
while not os.path.isfile(tmp_metis_result):
time.sleep(1)
print(f'{tmp_metis_result} is ready')
# reformat partition file to the format fit for distributed page rank, reassign subgruph index 0 to (largest index + 1)
with open(tmp_metis_result, 'r') as metis_partition_file:
partitions = tuple(int(partition) for partition in metis_partition_file.readlines())
max_subgraph = max(partitions) + 1
partitions = (str(partition if partition != 0 else max_subgraph) for partition in partitions)
buffer = '\n'.join(partitions)
with open(result_partition, 'w') as result_partition_file:
result_partition_file.write(buffer)
print(f'time consumption {time.time() - start_time}s')
# upload files to S3
upload_file(result_dataset, bucket, s3_key_dataset)
print(f'upload {result_dataset} to S3 bucket {bucket}, key = {s3_key_dataset}')
upload_file(result_r_dataset, bucket, s3_key_r_dataset)
print(f'upload {result_r_dataset} to S3 bucket {bucket}, key = {s3_key_r_dataset}')
upload_file(result_partition, bucket, s3_key_partition)
print(f'upload {result_partition} to S3 bucket {bucket}, key = {s3_key_partition}')