def fullTest():
res = []
allHitRefs = set()
allRefs = set()
#TEST
totalCount = 0
keep = []
for paperName in os.listdir(DATA_DIR):
paper = None
try:
#TEST
totalCount += 1
paper = parser.parseFullDataset(DATA_DIR + '/' + paperName)
if not paper:
print "Error parsing " + paperName
continue
#TEST
keep.append(paperName)
paperRes = normalRun(paper)
if paperRes:
allHitRefs.update(paperRes[2])
allRefs.update([ ref.title.strip().upper() for ref in paper.references.values()])
res.append((paperRes[0], paperRes[1]))
except OSError:
print "Error parsing paper: " + paperName
continue
#TEST
print "\n\n------------------------\n\n"
print "{} / {}".format(len(keep), totalCount)
res = sorted(res, reverse = True)
total = 0
for paper in res:
print paper[1]
total += paper[0]
print "Total Papers: {}".format(len(res))
print "Average: {}".format(total / len(res))
if len(res) % 2 == 0:
print "Median: {}".format((res[len(res) / 2][0] + res[len(res) / 2 - 1][0]) / 2.0)
else:
print "Median: {}".format(res[len(res) / 2][0])
print "Hit Refs: {}, All Refs: {}".format(len(allHitRefs), len(allRefs))
print "Foaad Metric: {}".format(float(len(allHitRefs)) / len(allRefs))
def main():
# full test mode
if '-f' in sys.argv:
fullTest()
return
#paper = parser.parseFullDataset('data/dynamo')
#paperName = '../data/A Bayesian Inference Framework to Reconstruct Transmission Trees Using Epidemiological and Genetic Data'
paperName = '../data/A Graphical Modelling Approach to the Dissection of Highly Correlated Transcription Factor Binding Site Profiles'
if '-p' in sys.argv:
paperName = sys.argv[sys.argv.index('-p') + 1]
paper = parser.parseFullDataset(paperName)
if not paper:
print "Error parsing paper"
return
print paper
# Ordering mode
if '-o' in sys.argv:
orderingRun(paper)
else:
debugRun(paper)
