greedyBOEM = [ge+["boem"] for ge in greedyMethods]
identBOEM = [ge+["boem"] for ge in identMethods]
identMethods += identBOEM
for solvers in greedyMethods + greedyBOEM + [["sdp2l", "pivot", "boem"],
["boem"]] + identMethods:
#exp = Experiment(solverCmd, solvers, matrix)
exp = Experiment(solverCmd, ["log"] + solvers, matrix)
args.append(exp)
ev = ChatEvaluation(exp, matrix.stripext()+".truth", golds,
matrix)
args.append(ev)
session = Session("Hogwash.Action", "action_runner", args, name=sessionName)
deps = {}
for expt,evalt in batch(session, 2):
deps[evalt] = expt
session.set_dependencies(deps)
if not sessionBinDir.exists():
sessionBinDir.mkdir()
copy("bin32/chainedSolvers", sessionBinDir)
session64BinDir = path(sessionName)/"bin64"
if not session64BinDir.exists():
session64BinDir.mkdir()
options)
except KeyError:
print("No such classifier as", options.classifier)
sys.exit(1)
print(sys.stderr, "training...")
classifier.train()
if options.check:
print(sys.stderr, "evaluating...")
classifier.check()
elif options.distributed:
print(sys.stderr, "distributed labeling...")
args = list(izip(repeat(classifier), batch(range(len(testPoints)),
30)))
ssn = Session("classify", "partLabel", args, name=options.distributed)
print(sys.stderr, "session created...")
if 0:
sty = Sty("workingSession")
sty.do_clear("all")
sty.do_runlocally("")
jobs_done = False
while not jobs_done:
stats = ssn.get_job_status_report()
queued = []
if "queued" in stats:
queued = stats["queued"]
started = []
if "started" in stats:
started = stats["started"]
["first"],
["best"],
["vote"],
["pivot"],
["boem"],
["first boem"],
["best boem"],
["vote boem"],
["pivot boem"],
["lazylp first"],
["lazylp best"],
["lazylp vote"],
["lazylp pivot"],
["lazylp first boem"],
["lazylp best boem"],
["lazylp vote boem"],
["lazylp pivot boem"],
["ilp"],
["sumsdp vote"],
]
args = []
for model in models:
for datadir in workdir.dirs():
if "bin" in datadir.basename():
continue
args.append((model, datadir.abspath()))
ssn = Session("timingRuns", "runtask", args,
name=("hog%s" % workdir.basename()))
trainPoints, testPoints, options)
except KeyError:
print("No such classifier as", options.classifier)
sys.exit(1)
print (sys.stderr, "training...")
classifier.train()
if options.check:
print(sys.stderr, "evaluating...")
classifier.check()
elif options.distributed:
print(sys.stderr, "distributed labeling...")
args = list(izip(repeat(classifier),
batch(range(len(testPoints)), 30)))
ssn = Session("classify", "partLabel", args, name=options.distributed)
print(sys.stderr, "session created...")
if 0:
sty = Sty("workingSession")
sty.do_clear("all")
sty.do_runlocally("")
jobs_done = False
while not jobs_done:
stats = ssn.get_job_status_report()
queued = []
if "queued" in stats:
queued = stats["queued"]
started = []
if "started" in stats:
started = stats["started"]
greedyBOEM = [ge + ["boem"] for ge in greedyMethods]
identBOEM = [ge + ["boem"] for ge in identMethods]
identMethods += identBOEM
for solvers in greedyMethods + greedyBOEM + [["sdp2l", "pivot", "boem"],
["boem"]] + identMethods:
#exp = Experiment(solverCmd, solvers, matrix)
exp = Experiment(solverCmd, ["log"] + solvers, matrix)
args.append(exp)
ev = ChatEvaluation(exp, matrix.stripext() + ".truth", golds, matrix)
args.append(ev)
session = Session("Hogwash.Action",
"action_runner",
args,
name=sessionName)
deps = {}
for expt, evalt in batch(session, 2):
deps[evalt] = expt
session.set_dependencies(deps)
if not sessionBinDir.exists():
sessionBinDir.mkdir()
copy("bin32/chainedSolvers", sessionBinDir)
session64BinDir = path(sessionName) / "bin64"
if not session64BinDir.exists():
session64BinDir.mkdir()
matrixFile = file(ddir / "matrix")
matrix = readClMat(matrixFile)
propFile = file(str(job.results))
prop = readDataFile(propFile)
res = Eval(gold=truth, prop=prop, clMat=matrix, filename=job.args[1])
res.stats["time"] = jobTime(job)
return res
if __name__ == "__main__":
workdir = path(sys.argv[1])
print "Working directory", workdir
wbase = workdir.dirname().basename()
ssnName = "hog" + wbase
print "Session", ssnName
session = Session(ssnName, read_only=True)
args = list(session)
evalSsn = Session("evaluateExperiments",
"evalTask",
args,
name="hogeval%s" % wbase)
depDict = {}
for evJob, expJob in izip(evalSsn, session):
depDict[evJob] = expJob
evalSsn.set_dependencies(depDict)
trainPts = readDataFile(goldFile)
truth = readDataFile(goldFile)
matrixFile = file(ddir/"matrix")
matrix = readClMat(matrixFile)
propFile = file(str(job.results))
prop = readDataFile(propFile)
res = Eval(gold=truth, prop=prop, clMat=matrix, filename=job.args[1])
res.stats["time"] = jobTime(job)
return res
if __name__ == "__main__":
workdir = path(sys.argv[1])
print "Working directory", workdir
wbase = workdir.dirname().basename()
ssnName = "hog" + wbase
print "Session", ssnName
session = Session(ssnName, read_only=True)
args = list(session)
evalSsn = Session("evaluateExperiments", "evalTask", args,
name="hogeval%s" % wbase)
depDict = {}
for evJob,expJob in izip(evalSsn,session):
depDict[evJob] = expJob
evalSsn.set_dependencies(depDict)
from AIMA import DefaultDict
from DistributedExperiment import Experiment, Evaluation
def getCorpus(obj):
return int(re.search("\d+", obj.exp.mfile.basename()).group(0))
def solverObj(solvers):
if "log" in solvers:
return "logObjective"
return "objective"
if __name__ == "__main__":
sessionName = sys.argv[1]
ssn = Session(sessionName, read_only=True)
evals = DefaultDict(DefaultDict([]))
for job in ssn:
#print job
if job.status != "finished":
continue
obj = job.args[0]
if isinstance(obj, Evaluation):
corpus = getCorpus(obj)
solvers = tuple(obj.exp.solvers)
evals[corpus][solvers].append(job.results)
avgBest = DefaultDict([])
avgAvg = DefaultDict([])
else:
color = "k"
if "boem" in modelStr:
sym = "."
else:
sym = ""
return color+sym+line
if __name__ == "__main__":
workdir = path(sys.argv[1])
print "Working directory", workdir
ssnName = "hog" + workdir.dirname().basename()
print "Session", ssnName
session = Session(ssnName, read_only=True)
evalSsnName = "hogeval" + workdir.dirname().basename()
print "Eval session", evalSsnName
evalSsn = Session(evalSsnName, read_only=True)
plotStat = "time"
modelSeqs = DefaultDict([])
for ct,job in enumerate(evalSsn):
if ct % 20 == 0:
print ct, "..."
if job.status == "finished":
res = job.results