-
Notifications
You must be signed in to change notification settings - Fork 2
/
utils.py
376 lines (329 loc) · 11.5 KB
/
utils.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
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
# Load required modules
import csv
# import random
# import matplotlib.pyplot as plt
# from matplotlib import cm
# import matplotlib
import networkx as nx
# import metis
# from collections import Counter
import numpy as np
import time
import argparse
import copy
# import numpy.linalg as la
# import scipy.cluster.vq as vq
# import itertools
# import operator
# import math
# import collections
# from mpmath import *
# from itertools import chain
# from itertools import product
# from itertools import starmap
# from functools import partial
# import os
# import seaborn as sns
# import shutil
# from networkx.drawing.nx_agraph import graphviz_layout
# import ujson
# from pycallgraph import PyCallGraph
# from pycallgraph.output import GraphvizOutput
# import numpy.linalg as la
# import scipy.cluster.vq as vq
# import scipy
import re
# from pycallgraph2.output import GraphvizOutput
##########################################
### create file names ###
##########################################
def edgelist_filename (settings, sample):
return settings[sample]['graph_path']+'/DAG.edgelist'
##########################################
### load files ###
##########################################
def read_json(inputfile):
lines = [open(inputfile, 'r').read().strip("\n")][0].split('\n')
ports, gates = {}, {}
for idx, line in enumerate(lines):
line = line.strip()
if line.startswith('"ports"'):
p_s = idx
searchlines = lines[idx+1:]
for i, sl in enumerate(searchlines, idx):
if sl.strip().startswith('"cells"'):
p_e = i+1
if line.startswith('"cells"'):
g_s = idx
searchlines = lines[idx+1:]
for i, sl in enumerate(searchlines, idx):
if sl.strip().startswith('"netnames"'):
g_e = i
# get information of inputs and outputs
spacer = [idx + p_s + 1 for idx, line in enumerate(lines[p_s + 1:p_e]) if ': {' in line.strip()]
for i, v in enumerate(spacer):
# get names
s = lines[v].strip()
s = re.search('"(.*)"', s)
el = s.group(1)
ports[el] = {}
# get directions
s = lines[v + 1].strip()
s = re.search('"direction": "(.*)"', s)
direction = s.group(1)
ports[el]['direction'] = direction
# get offset if it exists
s = lines[v + 2].strip()
if s[-1] == ',':
offset = s.split(':')[1].split(',')[0].strip()
ports[el]['offset'] = int(offset)
s = lines[v + 3].strip()
# get bits
bits = s.split('[')[1].split(']')[0].strip()
if ',' in bits:
bit_list = bits.replace(' ', '').split(',')
for bit_element in bit_list:
SubPortInfo = {'direction': ports[el]['direction'], 'bits': int(bit_element)}
SubPortID = f"{el}_{bit_element}"
ports[SubPortID] = SubPortInfo
del ports[el]
continue
else:
if bits == '"0"':
ports[el]['bits'] = 0
else:
ports[el]['bits'] = int(bits)
# get information of gates
spacer = [idx+g_s+1 for idx, line in enumerate(lines[g_s+1:g_e]) if '$abc$' in line.strip()]
for i, v in enumerate(spacer):
# get names
s = int(lines[v].strip().split('"')[1].split('$')[-1])
gates[s] = {}
gates[s]['input'] = {}
gates[s]['output'] = {}
# search for attributes of this gate
if i != len(spacer)-1:
searchlines = lines[v:spacer[i+1]]
else:
searchlines = lines[v:]
for sl in searchlines:
# get gate type
if sl.strip().startswith('"type"'):
gatetype = re.search('_(.*)_', sl.strip())
if not gatetype:
continue
gates[s]['type'] = gatetype.group(1)
# get input(s)
if sl.strip().startswith('"A": [') or sl.strip().startswith('"B": [') or sl.strip().startswith('"C": [') or sl.strip().startswith('"D": [') \
or sl.strip().startswith('"S": ['):
port = re.search('"(.*)"', sl).group(1)
bits = sl.split('[')[1].split(']')[0].strip()
gates[s]['input'][port] = int(bits)
# get output
# stop loop after getting output edge ID
if sl.strip().startswith('"Y": ['):
port = re.search('"(.*)"', sl).group(1)
bits = sl.split('[')[1].split(']')[0].strip()
gates[s]['output'][port] = int(bits)
if sl.strip().startswith('"$auto$'):
break
return ports, gates
def synthesize_graph(ports, gates, outdir, t):
G = nx.DiGraph()
# start from the output, add edges
edges = []
for p in ports:
if ports[p]['direction'] == 'output':
b = ports[p]['bits']
for g in gates:
if b == gates[g]['output']['Y']:
edges.append((g, p))
for p in ports:
if ports[p]['direction'] == 'input':
b = ports[p]['bits']
for g in gates:
if b == gates[g]['input']['A']:
edges.append((p, g))
if gates[g]['type'] != 'NOT':
if b == gates[g]['input']['B']:
edges.append((p, g))
if gates[g]['type'] in ['MUX', 'NMUX']:
if b == gates[g]['input']['S']:
edges.append((p, g))
if gates[g]['type'] in ['AOI3', 'OAI3', 'AOI4', 'OAI4']:
if b == gates[g]['input']['C']:
edges.append((p, g))
if gates[g]['type'] in ['AOI4', 'OAI4']:
if b == gates[g]['input']['D']:
edges.append((p, g))
for g in gates:
op = gates[g]['output']['Y']
for sg in gates:
# Ron update on 09/05/2023, add more gate types inside
if gates[sg]['type'] == 'NOT':
gin = [gates[sg]['input']['A']]
elif gates[sg]['type'] in ['MUX', 'NMUX']:
gin = [gates[sg]['input']['A'], gates[sg]['input']['B'], gates[sg]['input']['S']]
elif gates[sg]['type'] in ['AOI3', 'OAI3']:
gin = [gates[sg]['input']['A'], gates[sg]['input']['B'], gates[sg]['input']['C']]
elif gates[sg]['type'] in ['AOI4', 'OAI4']:
gin = [gates[sg]['input']['A'], gates[sg]['input']['B'], gates[sg]['input']['C'], gates[sg]['input']['D']]
else:
gin = [gates[sg]['input']['A'], gates[sg]['input']['B']]
if op in gin:
edges.append((g, sg))
for e in edges:
G.add_edge(*e)
nx.write_edgelist(G, outdir+'/DAG.edgelist')
# generate port-gate dictionary.
def generateGateDict(data, outpath):
csv_file = f'{outpath}/port_gate.csv'
with open(csv_file, 'w', newline='') as file:
writer = csv.writer(file)
writer.writerow(['key', 'input', 'output', 'type'])
for key, value in data.items():
input_data = value['input']
output_data = value['output']
row = [key, value['type'], input_data, output_data]
writer.writerow(row)
def load_settings (filename):
"""Load the settings file"""
settings = {}
data_reader = csv.reader(open(filename, 'rU'), delimiter='\t')
# Ignore header
header = next(data_reader)
# Process each line
for row in data_reader:
if len(row) == len(header):
sample = row[0]
sample_data = {}
for el_idx, el in enumerate(header[1:]):
sample_data[el] = row[el_idx+1]
settings[sample] = sample_data
return settings
def load_graph (settings, sample):
"""
read DAG edgelist, return DIRECTED graph, and input/output nodes
"""
G = nx.read_edgelist (edgelist_filename (settings, sample), nodetype = str, create_using=nx.DiGraph())
return G
def load_graph_undirected (settings, sample):
"""
read DAG edgelist, return UNDIRECTED graph, and input/output nodes
"""
G = nx.Graph()
G = nx.read_edgelist (edgelist_filename (settings, sample), nodetype=str)
return G
def load_metis_part_sol (inputfile):
"""
read metis partition result
"""
lines = [open(inputfile, 'r').read().strip("\n")][0].split('\n')
cut = int( lines[0].split('\t')[1] )
partDict = {}
for line in lines[1:]:
tokens = line.split('\t')
part = int( tokens[0].split(' ')[-1] )
nodes = tokens[1].split(',')
partDict[part] = nodes
# print(partDict)
return cut, partDict
def get_nonprimitive_nodes (G):
"""
Obtain nonprimitive nodes of a DAG
input nodes (in_nodes) - in_degree is 0
output nodes (out_nodes) - out_degree is 0
"""
in_nodes, out_nodes = [], []
for node in G.nodes():
indegree = G.in_degree(node)
outdegree = G.out_degree(node)
if outdegree == 0:
out_nodes.append(node)
if indegree == 0:
in_nodes.append(node)
nonprimitives = in_nodes + out_nodes
return in_nodes, out_nodes, nonprimitives
def get_G_primitive (G, nonprimitives):
"""
if primitive only is True, remove input and output nodes
"""
G_primitive = nx.DiGraph()
for edge in G.edges():
if edge[0] not in nonprimitives and edge[1] not in nonprimitives:
G_primitive.add_edge(*edge)
return G_primitive
def loadSettings():
# Parse the command line inputs
parser = argparse.ArgumentParser(description="perform graph partition using metis")
parser.add_argument("-settings", dest="settings", required=True, help="settings.txt", metavar="string")
parser.add_argument("-samples", dest="samples", required=True, help="1,2", metavar="string")
args = parser.parse_args()
# Run the command
samples = args.samples.split(',')
settings = load_settings(args.settings)
return samples, settings
def loadData(s, settings):
print('Processing sample', s)
# print (settings[s])
# obtain user-defined params
tmp = settings[s]['S_bounds'].replace('\"','').split(',')
S_bounds = [eval(i) for i in tmp]
# if target_n = -1, that means we want to find the most optimal partition solution
# target_n can stop running the merging.py when the algorithm found a merge solution that has subgroups less than target_n
# If Color Flag = 1, we suggest users set it as -1.
# Because the edge coloring stage may not find solution with color assignment also has subgroups less than target_n.
target_n = -1
primitive_only = settings[s]['primitive_only']
ConstraintType = settings[s]['high_low_flag'].split(',')[0]
constraint = []
if ConstraintType.lower() == 'high':
tmp = settings[s]['high_constraint'].split(',')
for i in tmp:
constraint.append(eval(i))
else:
constraint.append(int(settings[s]['low_constraint'].split(',')[0]))
loop_free = False
if settings[s]['loop_free'].lower() == 'true':
loop_free = True
color_flag = int(settings[s]['ColorFlag'])
if color_flag == 0:
target_n = int(settings[s]['target_n'].split(',')[0])
bio_flag = int(settings[s]['BioFlag'])
out_path = settings[s]['output_path']
# "attempts" means the number of possible merging paths we will collect, note that each path includes multiple partition results
# time step for verification stage
# time step for merging stage
# depth: searching depth for each community merging propaganda checking
# "depth2": the depth of searching possible merging solution for un-neighbor communities in every propaganda checking
# Upper bound for continuously negative reward path in each propaganda checking
attempt_range = []
tmp = settings[s]['attempt_range'].split(',')
for i in tmp:
attempt_range.append(eval(i))
timestep = int(settings[s]['timestep_v'])
timestep2 = int(settings[s]['timestep_m'])
depth = int(settings[s]['depth'])
depth2 = int(settings[s]['depth2'])
ub = int(settings[s]['ub'])
# parameters for edge coloring.
# Assume we have at most n different cell-cell communication molecular for one benchmark,
# set n as the element in the below color list: "color_upperbounds".
# SingleFlag = True: Only check one solution file
# SingleFlag = False: Check a list of potential solution
color_upperbound = int(settings[s]['color_upperbound'])
trace_back = int(settings[s]['timestep_traceback'])
check_interval = int(settings[s]['check_interval'])
SingleFlag = True
if settings[s]['SingleFlag'].lower() == 'false':
SingleFlag = False
# load graph
# G = load_graph_undirected(settings, s)
DAG = load_graph(settings, s)
in_nodes, out_nodes, nonprimitives = get_nonprimitive_nodes(DAG)
if primitive_only == 'TRUE':
G_primitive = get_G_primitive(DAG, nonprimitives)
else:
G_primitive = copy.deepcopy(DAG)
return G_primitive, S_bounds, target_n, primitive_only, ConstraintType, constraint, loop_free, out_path, timestep, timestep2, \
bio_flag, color_flag, depth, DAG, depth2, attempt_range, ub, color_upperbound, trace_back, check_interval, SingleFlag