def get_defense_data(opponents):
import parse
defense_info=map(lambda team: parse.run(filename,team)[-1],opponents)
print defense_info
by_category=[
('Portcullis','Cheval'),
('Moat','Ramparts'),
('Drawbridge','Sally Port'),
('Rock Wall','Rough Terrain')
]
print sorted(defense_info[0])
def find_better(opts):
m=map(
lambda def_type: sum(map(lambda x: x[def_type],defense_info)),
opts
)
if m[0]<m[1]: return (opts[0],m[1]-m[0])
return (opts[1],m[0]-m[1])
return map(find_better,by_category)
def bayesianClassifier(inputSeq_train, inputSeq_test, labels_train,
labels_test):
""" An experimental implementation using orthogonalization of the input.
Gaussian Naive Bayesian Classifier is used.
"""
print 'Gaussian Naive Bayesian Classifier.'
feats_train = RealFeatures(inputSeq_train)
feats_test = RealFeatures(inputSeq_test)
labels = MulticlassLabels(labels_train)
print 'Initializing kernel..'
classifier = GaussianNaiveBayes(feats_train, labels)
print 'Training svm..'
classifier_train = classifier.train()
print 'Running test data..'
label_pred = classifier.apply(feats_test)
out = label_pred.get_labels()
# Evaluating the accuracy.
if label_test is not None:
labels_test_set = MulticlassLabels(numpy.asarray(label_test))
evaluator = MulticlassAccuracy()
acc = evaluator.evaluate(label_pred, labels_test_set)
# print 'Accuracy : %.4f' % (acc * 100)
# Giving the new input.
inp = parse.run("513_distribute/7cata.all")
inpMatrix = inputParser.createFeatureMatrix(inp)
inp_feats = RealFeatures(numpy.transpose(numpy.asarray(inpMatrix)))
# Running svm on the input.
label_out = classifier.apply(inp_feats)
out_new = label_out.get_labels()
return out_new
def bayesianClassifier(inputSeq_train, inputSeq_test, labels_train, labels_test ):
""" An experimental implementation using orthogonalization of the input.
Gaussian Naive Bayesian Classifier is used.
"""
print 'Gaussian Naive Bayesian Classifier.'
feats_train = RealFeatures(inputSeq_train)
feats_test = RealFeatures(inputSeq_test)
labels = MulticlassLabels(labels_train)
print 'Initializing kernel..'
classifier = GaussianNaiveBayes(feats_train, labels)
print 'Training svm..'
classifier_train = classifier.train()
print 'Running test data..'
label_pred = classifier.apply(feats_test)
out = label_pred.get_labels()
# Evaluating the accuracy.
if label_test is not None:
labels_test_set = MulticlassLabels(numpy.asarray(label_test))
evaluator = MulticlassAccuracy()
acc = evaluator.evaluate(label_pred, labels_test_set)
# print 'Accuracy : %.4f' % (acc * 100)
# Giving the new input.
inp = parse.run("513_distribute/7cata.all")
inpMatrix = inputParser.createFeatureMatrix(inp)
inp_feats = RealFeatures(numpy.transpose(numpy.asarray(inpMatrix)))
# Running svm on the input.
label_out = classifier.apply(inp_feats)
out_new = label_out.get_labels()
return out_new
import sys
sys.path.append(".")
from parse import run
if __name__ == '__main__':
run()
import parse
import token
# To take Multiline input from user.
print("Enter/Paste your code.Ctrl-Z ( windows ) to save it.")
contents = ""
while True:
try:
contents += input(">> ") + "\t"
except EOFError:
break
result = parse.run(contents)
print(result)
#!/usr/bin/env python import parse import scraper scraper.run() parse.run()
def get_data(team): p=parse.run(filename,team) print 'parsed for',team,':',p ball_time,defense_time,challenge,climb,defense_info=p return ball_time/2,defense_time,ball_time/2,challenge,climb
