-
Notifications
You must be signed in to change notification settings - Fork 0
/
hunger_legacy.py
443 lines (353 loc) · 18.3 KB
/
hunger_legacy.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
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
from hungerDataStructs import *
from datetime import datetime as dt
import os
from statistics import pstdev
from numpy import percentile as pct
class Solver:
def __init__(self, metricID, numDist):
Logger.initialize()
self.reset(metricID, numDist)
def __del__(self):
Logger.cleanup()
# External Getters ----------------------------------------------------------------------------
def reset(self, metricID, numDist):
self.inProgress = False
if isinstance(metricID, str):
self.metricID = metricID
elif isinstance(metricID, int):
self.metricID = allowed[metricID]
self.placements = {region: 0 for region in sorted(regionlist.values(), key=lambda region: region.code)}
self.placedRegions = []
self.failures = set()
self.districts = [District(i+1) for i in range(numDist)]
# Time logging
self.startTime = 0
self.lastTime = 0
self.times = {}
self.occurred = self.times.copy()
# Calculate the minimum acceptable standard deviation
maxRegionMetric = self.__getLargestUnplacedFor().metrics[self.metricID]
allOtherMetrics = sum(region.metrics[self.metricID] for region in self.__getUnplacedRegions()) - maxRegionMetric
if numDist <= 1 or allOtherMetrics/(numDist-1) > maxRegionMetric:
self.minAcceptable = 0.5
else:
hypothetical = [maxRegionMetric] + [allOtherMetrics/(numDist-1)]*(numDist-1)
percentile = 100*pstdev(hypothetical)/sum(hypothetical)
self.minAcceptable = max(0.5, percentile)
def isSolved(self):
return all(placement != 0 for placement in self.placements.values()) and self.getStandardDevAsPercent() < self.minAcceptable
def getStandardDevAsPercent(self):
metrics = [district.metric for district in self.districts]
return 100*pstdev(metrics)/sum(metrics)
def getTimeSinceStarted(self):
if self.startTime == 0:
return -1
else:
return (self.lastTime - self.startTime).total_seconds()
def getEmptyDataFrame():
return {new_list: [] for new_list in ["region","code","district","metric"]}
def getDummyDataFrame():
result = Solver.getEmptyDataFrame()
result["region"].append("none")
result["code"].append("none")
result["district"].append(str(1))
result["metric"].append("none")
return result
def getCurrentDataFrame(self):
result = Solver.getEmptyDataFrame()
for district in self.districts:
for region in district.regions:
result["region"].append(region.name)
result["code"].append(region.code)
result["district"].append(str(district.index))
result["metric"].append(region.metrics[self.metricID])
if len(result["region"]) == 0:
result = Solver.getDummyDataFrame()
return result
def getStarters(self):
self.__seedStarters(True)
return self
# Printers ------------------------------------------------------------------------------------
def printResult(self):
Logger.l("\n\n+---------------------------------------------------------+")
for district in self.districts:
print("District {} ({}):".format(district.index, district.metric))
print("|".join(sorted(region.code for region in district.regions)))
print()
return self
def printConcise(self):
placedRegions = sorted(self.placedRegions, key=lambda region: region.code)
formatstr = "|".join(["{:^" + str(len(placedRegions[0].code)) + "}"]*len(placedRegions))
print(formatstr.format(*(region.code for region in placedRegions)))
print(formatstr.format(*(self.placements[region] for region in placedRegions)))
return self
def printSummary(self):
print("\t{:>10}({}) took {:.3f}s ({:.3f}%, {} failures)".format(self.metricID,
len(self.districts),
self.getTimeSinceStarted(),
self.getStandardDevAsPercent(),
len(self.failures)))
return self
def __doStepLogging(self):
total = sum(self.times.values())
total = 1 if total == 0 else total
#timesToPrint = { key: time for key, time in sorted(self.times.items()) if key in [] }
timesToPrint = { key: time for key, time in sorted(self.times.items()) }
result = []
# Failure count
result.append("failures")
result.append("")
result.append(len(self.failures))
# Time so far
result.append("total")
result.append("")
result.append(self.getTimeSinceStarted())
# Times
for key, time in timesToPrint.items():
result.append(key)
result.append(self.occurred[key])
percent = 100*time/total
if percent > 70:
result.append(color.RED)
elif percent > 20:
result.append(color.YELLOW)
else:
result.append(color.GREEN)
result.append(percent)
# Put it all together
formatStr = " | ".join(["{}: {}{:<6}" + Style.RESET_ALL] + ["{}: {}{:7.3f}" + Style.RESET_ALL] + ["{}({:4}):{}{:6.2f}%" + Style.RESET_ALL]*len(timesToPrint))
resultStr = formatStr.format(*result)
# Subtract out the hidden style characters
numChars = len(resultStr) - 9 * len(timesToPrint)
# Progress bar
# The number of available cells for progress bar
availCells = os.get_terminal_size().columns - numChars - 14
# The number we'll actually use - closest 10, rounding down (no rounding if less than 10)
numCells = availCells if availCells <= 10 else availCells - (availCells%10)
if numCells > 5:
percent = 100*len(self.placedRegions)/len(self.placements)
if percent < 50:
progressColor = color.RED
elif percent < 90:
progressColor = color.YELLOW
else:
progressColor = color.GREEN
progressbar = ("completion: {}+{:" + "{}".format(numCells) + "}+" + Style.RESET_ALL + " | ").format(progressColor, "="*round(percent*(numCells/100)))
suffix = ""
else:
progressbar = ""
suffix = (os.get_terminal_size().columns - numChars)*" "
# Put it all together!
print(progressbar + resultStr + suffix, end="\r")
# Setters -------------------------------------------------------------------------------------
def __addToFailures(self):
self.failures.add(frozenset(self.placements.items()))
def __place(self, region, district):
district.addRegion(region, self.metricID)
self.placedRegions.append(region)
self.placements[region.code] = district.index
def __popLastPlaced(self, district):
for region in reversed(self.placedRegions):
# ignore regions that are already in this district
if region in district.regions:
continue
district = self.districts[self.placements[region]-1]
district.removeRegion(region, self.metricID)
self.placedRegions.remove(region)
self.placements[region.code] = 0
return region
return False
def __updateTime(self, tag = None):
if self.startTime == 0:
self.startTime = dt.now()
if not tag:
self.lastTime = dt.now()
else:
newTime = dt.now()
self.times[tag] = self.times.get(tag, 0) + (newTime - self.lastTime).total_seconds()
self.occurred[tag] = self.occurred.get(tag, 0) + 1
self.lastTime = newTime
def __addUnusedDistricts(self):
unusedDistricts = self.__getUnusedDistricts()
for ud in unusedDistricts:
adjacentSet = set(ud.adj)
for d in self.districts:
# If this unused district's adjacent regions are all in district d, AND there are some adjacent regions (sorry Alaska), add them all!
if adjacentSet and adjacentSet <= d.regions:
if not self.__canAddToDistrict(ud.regions, d):
return False
Logger.s("!", d.index, "enclosed regions:", ud.regions)
for region in ud.regions:
self.__place(region, d)
return True
def __seedStarters(self, doStatus = False):
distances = distanceMatrix.copy()
percentile = pct([region.metrics[self.metricID] for region in regionlist.values()], 50)
for district in self.districts:
minDistances = {}
for code in (region.code for region in self.__getUnplacedRegions() if region.metrics[self.metricID] > percentile):
reachablePlacedRegions = { region.code: distances[code][region.code] for region in self.placedRegions if distances[code][region.code] > 0 }
if len(reachablePlacedRegions) > 0:
minDistances[code] = min(reachablePlacedRegions.items(), key=lambda item: item[1])
if len(minDistances) == 0:
region = self.__getLargestUnusedRegionFor(lambda region: True)
else:
code = max(minDistances, key=lambda code: minDistances[code][1])
region = regionlist[code]
self.__place(region, district)
# Internal getters ----------------------------------------------------------------------------
def __canAddToDistrict(self, regions, district):
if isinstance(regions, Region):
regions = [regions]
currState = self.placements.copy()
for region in regions:
currState[region.code] = district.index
return frozenset(currState.items()) not in self.failures
def __getLargestUnplacedFor(self, district=None):
if district==None:
unplaced = (region for region in self.__getUnplacedRegions())
sorter = lambda region: region.metrics[self.metricID]
else:
unplaced = (region for region in self.__getUnplacedRegions() if (region.code in district.adj or len(region.adj) == 0) and self.__canAddToDistrict(region, district))
sorter = lambda region: (district.adj.get(region.code,0), region.metrics[self.metricID])
return max(unplaced, key=sorter, default=False)
def __getNextStarter(self):
# Get the distances
minDistances = {}
percentile = pct([region.metrics[self.metricID] for region in self.__getUnplacedRegions()], 50)
for region in filter(lambda region: region.metrics[self.metricID] >= percentile, self.__getUnplacedRegions()):
reachablePlacedRegions = { inRegion: distanceMatrix[region][inRegion] for inRegion in self.placedRegions if distanceMatrix[region][inRegion] > 0 }
minDistances[region] = min(reachablePlacedRegions.items(), key=lambda item: item[1], default=("", float("inf")))
# If nothing is reachable, just get the biggest unused region
if all(distance[1] == float("inf") for distance in minDistances.values()):
return self.__getLargestUnplacedFor()
# If we can reach some items, get those items!
else:
return max(minDistances, key=lambda code: minDistances[code][1], default=False)
def __getUnusedRegionFor(self, criteria):
return next((region for region in self.__getUnusedRegionsFor(criteria)), False)
def __getLargestUnusedRegionFor(self, criteria):
return max(self.__getUnusedRegionsFor(criteria),
key=lambda region: region.metrics[self.metricID],
default=False)
def __getLargestUnusedRegionForDistrict(self, district, criteria):
return max(self.__getUnusedRegionsFor(lambda region: region.code in district.adj and criteria(region)),
key=lambda region: (district.adj[region.code], region.metrics[self.metricID]),
default=False)
def __getLargestUnusedNeighborlessRegionFor(self, criteria):
disconnectedRegions = [region for district in self.__getUnusedDistricts() if len(district.adj) == 0 for region in district.regions]
return max(self.__getUnusedRegionsFor(lambda region: region in disconnectedRegions and criteria(region)),
key=lambda region: region.metrics[self.metricID],
default=False)
def __getUnusedRegionsFor(self, criteria):
# filter to get the regions matching this criteria
return filter(criteria, self.__getUnplacedRegions())
def __getUnusedDistricts(self):
# Group unused regions into districts
regionsToBePlaced = sorted(self.__getUnplacedRegions(), key=lambda region: len(region.adj))
while seedRegion := next(iter(regionsToBePlaced), False):
unusedDistrict = District(0)
unusedDistrict.addRegion(seedRegion, self.metricID)
regionsToBePlaced.remove(seedRegion)
while adjRegion := next(filter(lambda region: region.code in unusedDistrict.adj, regionsToBePlaced), False):
unusedDistrict.addRegion(adjRegion, self.metricID)
regionsToBePlaced.remove(adjRegion)
yield unusedDistrict
def __getUnplacedRegions(self):
for region in filter(lambda region: self.placements[region] == 0, self.placements):
yield region
# Solve it ------------------------------------------------------------------------------------
def doStep(self, doStatus = False):
self.__updateTime()
# Don't double-dip, and don't perform this if it's solved
if self.inProgress or self.isSolved():
return
self.__updateTime("checkSolved")
self.inProgress = True
# get the smallest district
district = min(self.districts)
self.__updateTime("getMinDistrict")
if all(len(district.regions) == 0 for district in self.districts):
self.__seedStarters(doStatus)
self.__updateTime("seed")
# if this district has no neighbors (either empty or only containing neighborless regions), append the largest unused Region
elif len(district.adj) == 0 and (region := self.__getLargestUnusedRegionFor(lambda region: self.__canAddToDistrict(region, district))):
self.__updateTime("selectMax")
self.__place(region, district)
# else if there are adjacent regions, add the biggest one
elif region := self.__getLargestUnusedRegionForDistrict(district, lambda region: self.__canAddToDistrict(region, district)):
self.__updateTime("selectAdj")
self.__place(region, district)
# else if there are neighborless regions, add the biggest one
elif region := self.__getLargestUnusedNeighborlessRegionFor(lambda region: self.__canAddToDistrict(region, district)):
self.__updateTime("selectNoNb")
self.__place(region, district)
# else step backwards until we find something we can add to this district
else:
self.__updateTime("selectFailed")
# Whatever led us to this point failed us - record the failure
self.__addToFailures()
while region := self.__popLastPlaced(district):
if region in district.adj and self.__canAddToDistrict(region, district):
self.__updateTime("backtrack")
self.__place(region, district)
break
else:
self.__addToFailures()
self.__updateTime("add region")
if all(len(district.regions) > 0 for district in self.districts) and not self.isSolved():
if not self.__addUnusedDistricts():
# Whatever led us to this point failed us - record the failure
self.__addToFailures()
self.__updateTime("checkUnused")
self.inProgress = False
# Do the logging for this step
if doStatus: self.__doStepLogging()
def solve(self, doStatus = False, doLogging = False):
Logger.doLogging = doLogging
# Don't show the progress bar if logging is enabled
if doLogging: doStatus = False
while not self.isSolved():
self.doStep(doStatus)
if doStatus: print()
return self
# Solve and unit test
import cProfile
import pstats
def profile(string):
filename = "stats.profile"
cProfile.run(string, filename)
stats = pstats.Stats(filename)
stats.strip_dirs()
stats.sort_stats("tottime")
stats.print_stats(10)
def unitTest(count, doStatus=True):
solver = Solver(0, 1)
result = {new_list: [] for new_list in allowed}
for count in range(1, count + 1):
if doStatus: print("{} districts".format(count))
for metric in allowed:
tick = dt.now()
solver.reset(metric, count)
solver.solve(doStatus).printSummary()
result[metric].append((dt.now() - tick).total_seconds())
# Write to file
i = 0
while os.path.exists("logs/log{}.txt".format(i)):
i += 1
metricFmt = "{:>13} |"
intFmt = " | ".join(["{:^9}"]*count)+"\n"
floatFmt = " | ".join(["{:^9.3f}"]*count)+"\n"
with open("logs/log{}.txt".format(i), "w", encoding='utf8') as log:
log.write((metricFmt.format("") + intFmt.format(*range(1,count + 1))))
for metric, values in result.items():
log.write((metricFmt.format(metric) + floatFmt.format(*values)))
def lightUnitTest(start=1, end=6):
for count in range(start, end+1):
for metric in allowed:
Solver(metric, count).solve(True).printSummary()
#lightUnitTest()
#unitTest(6)
#profile("Solver(4, 3).solve().printConcise()")
#Solver(1, 4).solve(doLogging=True).printResult().printConcise()
#Solver(0, 10).getStarters().printConcise()