def main():
inputs = ReadCSV('./data/input.csv')
outputs = ReadCSV('./data/output.csv')
ids = make_list(outputs)
x = []
y = {}
for i in range(3, 24, 3):
x.append(i)
for hidden_layers in range(
1, 7): #change loop for plot also if you change here
print 'Number of hidden layers', hidden_layers
y[hidden_layers] = []
for i in x:
print 'Number of training tuples', i
y[hidden_layers].append(
find_accuracy(i, ids, hidden_layers, inputs, outputs))
#print x
#print y
#Now plot the graph
for i in range(1, 7):
t = pylab.plot(x,
y[i],
label="Number of hidden layers " + str(i),
linewidth=4,
linestyle='-')
t = pylab.xlabel("Number of training tuples")
t = pylab.ylabel("Accuracy %")
t = pylab.title(
"Random Subset Sampling (Number of training samples vs Accuracy)")
t = pylab.legend(loc='lower right')
t = pylab.grid()
pylab.show()
def main():
inputs = ReadCSV('./data/input.csv')
outputs = ReadCSV('./data/output.csv')
test_set = test.keys()
train_set = []
for k in inputs.keys():
if k not in test_set:
train_set.append(k)
print "Number of training samples", len(train_set)
print "Number of testing samples", len(test_set)
net = buildNetwork(178, 6, 5)
ds=SupervisedDataSet(178,5)
for id in train_set:
ds.addSample(inputs[id],outputs[id])
trainer = BackpropTrainer(net, ds, learningrate=0.001, momentum = 0.001)
trainer.trainUntilConvergence(maxEpochs=1000, validationProportion = 0.5)
for id in test_set:
predicted = net.activate(inputs[id])
actual = outputs[id]
print '-----------------------------'
print test[id]
print '-----------------------------'
print 'Trait\t\tPredicted\tActual\tError'
for i in range(0,5):
error = abs(predicted[i] - actual[i])*100/4.0
print traits[i], '\t', predicted[i], '\t', actual[i], '\t', error,"%"
def main():
inputs = ReadCSV('./data/input.csv')
outputs = ReadCSV('./data/output.csv')
ids = make_list(outputs)
x = []
y = {}
for i in range(2, 11, 2):
x.append(i)
for hidden_layers in range(
1, 6): #change loop for plot also if you change here
print 'Number of hidden layers', hidden_layers
y[hidden_layers] = []
for i in x:
print 'Number of folds', i
y[hidden_layers].append(
find_accuracy(i, ids, hidden_layers, inputs, outputs))
#print x
#print y
#Now plot the graph
for i in range(1, 6):
t = pylab.plot(x,
y[i],
label="Number of hidden layers " + str(i),
linewidth=4,
linestyle='-')
t = pylab.xlabel("Number of folds")
t = pylab.ylabel("Accuracy %")
t = pylab.title("K-Fold cross validation (Number of folds vs Accuracy)")
t = pylab.legend(loc='lower right')
t = pylab.grid()
pylab.show()
def main():
inputs = ReadCSV('./data/input.csv')
outputs = ReadCSV('./data/output.csv')
ids = make_list(outputs)
net = constructNet(ids, inputs, outputs)
pickle.dump(net, open('neuralNet.sl', 'w'))
def connect(self):
if (self.csv_loaded == False):
self.plotWidget.clear()
line = 1
rfile = ReadCSV.read("test.csv")
for row in rfile:
if (line == 1):
line = line + 1
continue
self.upload_data.append(row)
Fixer.fixTimeAndAlt(self.upload_data)
buff = Buffer()
if os.path.exists('data.txt'):
os.remove('data.txt')
for i in self.upload_data:
buff.sendToBuffer(i)
buff.sendData()
self.data = (DataLoader.read('data.txt'))
self.csv_loaded = True
self.infocsv.setText(_fromUtf8("Zaladowano"))
self.info.setText("")
self.wys = self.kat = self.dyst = self.pred = self.odchyl = 0
def connect(self):
if (self.csv_loaded == False):
self.plotWidget.clear()
line = 1
rfile = ReadCSV.read("test.csv")
for row in rfile:
if(line == 1):
line = line+1
continue
self.upload_data.append(row)
Fixer.fixTimeAndAlt(self.upload_data)
buff = Buffer()
if os.path.exists('data.txt'):
os.remove('data.txt')
for i in self.upload_data:
buff.sendToBuffer(i)
buff.sendData()
self.data = (DataLoader.read('data.txt'))
self.csv_loaded = True
self.infocsv.setText(_fromUtf8("Zaladowano"))
self.info.setText("")
self.wys = self.kat = self.dyst = self.pred = self.odchyl = 0
def model_train(Save = False, modelname = None):
X1_left = ReadCSV('datasets_csv/feature_0630_1700/X1_offline_0630_1700.csv','float64')
X1_midle = ReadCSV('datasets_csv/feature_0701_1100/X1_offline_0701_1100.csv','float64')
X1_right = ReadCSV('datasets_csv/feature_0701_1400/X1_offline_0701_1400.csv','float64')
Y1 = ReadCSV('datasets_csv/feature_0701_1100/Y1_offline_0701_1100.csv','float64')
X1 = np.c_[X1_left,X1_midle,X1_right]
# X1 = np.c_[X1_left]
X1 = list(X1)
Y1 = list(Y1)
X3_left = ReadCSV('datasets_csv/feature_0630_1700/X3_offline_0630_1700.csv','float64')
X3_midle = ReadCSV('datasets_csv/feature_0701_1100/X3_offline_0701_1100.csv','float64')
X3_right = ReadCSV('datasets_csv/feature_0701_1400/X3_offline_0701_1400.csv','float64')
Y3 = ReadCSV('datasets_csv/feature_0701_1100/Y3_offline_0701_1100.csv','float64')
X3 = np.c_[X3_left,X3_midle,X3_right]
# X3 = np.c_[X3_left]
X3 = list(X3)
Y3 = list(Y3)
X1.extend(X3)
Y1.extend(Y3)
del X3,Y3
X4_left = ReadCSV('datasets_csv/feature_0630_1700/X4_offline_0630_1700.csv','float64')
X4_midle = ReadCSV('datasets_csv/feature_0701_1100/X4_offline_0701_1100.csv','float64')
X4_right = ReadCSV('datasets_csv/feature_0701_1400/X4_offline_0701_1400.csv','float64')
Y4 = ReadCSV('datasets_csv/feature_0701_1100/Y4_offline_0701_1100.csv','float64')
X4 = np.c_[X4_left,X4_midle,X4_right]
# X4 = np.c_[X4_left]
X4 = list(X4)
Y4 = list(Y4)
X1.extend(X4)
Y1.extend(Y4)
del X4,Y4
X5_left = ReadCSV('datasets_csv/feature_0630_1700/X5_offline_0630_1700.csv','float64')
X5_midle = ReadCSV('datasets_csv/feature_0701_1100/X5_offline_0701_1100.csv','float64')
X5_right = ReadCSV('datasets_csv/feature_0701_1400/X5_offline_0701_1400.csv','float64')
Y5 = ReadCSV('datasets_csv/feature_0701_1100/Y5_offline_0701_1100.csv','float64')
X5 = np.c_[X5_left,X5_midle,X5_right]
X5 = list(X5)
Y5 = list(Y5)
X1.extend(X5)
Y1.extend(Y5)
del X5,Y5
X6_left = ReadCSV('datasets_csv/feature_0630_1700/X6_offline_0630_1700.csv','float64')
X6_midle = ReadCSV('datasets_csv/feature_0701_1100/X6_offline_0701_1100.csv','float64')
X6_right = ReadCSV('datasets_csv/feature_0701_1400/X6_offline_0701_1400.csv','float64')
Y6 = ReadCSV('datasets_csv/feature_0701_1100/Y6_offline_0701_1100.csv','float64')
X6 = np.c_[X6_left,X6_midle,X6_right]
X6 = list(X6)
Y6 = list(Y6)
X1.extend(X6)
Y1.extend(Y6)
del X6,Y6
X7_left = ReadCSV('datasets_csv/feature_0630_1700/X7_offline_0630_1700.csv','float64')
X7_midle = ReadCSV('datasets_csv/feature_0701_1100/X7_offline_0701_1100.csv','float64')
X7_right = ReadCSV('datasets_csv/feature_0701_1400/X7_offline_0701_1400.csv','float64')
Y7 = ReadCSV('datasets_csv/feature_0701_1100/Y7_offline_0701_1100.csv','float64')
X7 = np.c_[X7_left,X7_midle,X7_right]
X7 = list(X7)
Y7 = list(Y7)
X1.extend(X7)
Y1.extend(Y7)
del X7,Y7
X8_left = ReadCSV('datasets_csv/feature_0630_1700/X8_offline_0630_1700.csv','float64')
X8_midle = ReadCSV('datasets_csv/feature_0701_1100/X8_offline_0701_1100.csv','float64')
X8_right = ReadCSV('datasets_csv/feature_0701_1400/X8_offline_0701_1400.csv','float64')
Y8 = ReadCSV('datasets_csv/feature_0701_1100/Y8_offline_0701_1100.csv','float64')
X8 = np.c_[X8_left,X8_midle,X8_right]
X8 = list(X8)
Y8 = list(Y8)
X1.extend(X8)
Y1.extend(Y8)
del X8,Y8
X1 = np.array(X1)
Y1 = np.array(Y1)
X1,Y1 = DowmSample(X1,Y1,1)
model = RandomForestClassifier(n_estimators=100,random_state=1)
# model = GradientBoostingClassifier(n_estimators=100,max_leaf_nodes=5, subsample=0.8, random_state=1)
# model = LogisticRegression('l2')
# model = RVC()
y0 = []
for y in Y1:
y0.append(y[0])
y1 = np.array(y0)
# y1 = y0.reshape(1,len(y0))
print 'size of X1',np.shape(X1)
print 'size of y1',np.shape(y1)
model.fit(X1, y1.ravel())
model.fit(X1, y1)
# 保存模型
if Save == True:
f = open(modelname,'w')
pickle.dump(model, f)
f.close()
print '\n -------------- Training is over ----------------------'
return model
import sys
sys.path.append('./lib')
from ReadCSV import ReadCSV
print ReadCSV("./data/input.csv")
from ReadCSV import ReadCSV
rs = ReadCSV(
'/home/dgrfi/MEGA/supersymmetry/7TeVxyz.csv').read_matrix_in_dataframe()
from pybrain.tools.shortcuts import buildNetwork
from pybrain.datasets import SupervisedDataSet
from pybrain.structure import TanhLayer
from pybrain.supervised.trainers import BackpropTrainer
import pickle
import sys
sys.path.append('./lib')
from ReadCSV import ReadCSV
new_input = ReadCSV("newinput.csv")
net = pickle.load(open("neuralNet.sl", "r"))
input = new_input[new_input.keys()[0]]
traits = net.activate(input)
text = ""
for trait in traits:
text = text + str(trait) + " "
print text
from pybrain.tools.shortcuts import buildNetwork
from pybrain.datasets import SupervisedDataSet
from pybrain.structure import TanhLayer
from pybrain.supervised.trainers import BackpropTrainer
import pickle
import sys
sys.path.append('./lib')
from ReadCSV import ReadCSV
new_input = ReadCSV("newinput.csv")
net = pickle.load(open("neuralNet.sl", "r"))
input = new_input[new_input.keys()[0]]
traits = net.activate(input)
text = ""
for trait in traits:
text = text + str(trait) + " "
print text
def createSample(self):
sum_w8 = self.w.sum_weights(self.w.weights)
#print(sum_w8)
rd = ReadCSV()
sample = rd.create_sampleList(self.sample_limit, self.db, sum_w8)
return sample
from ReadCSV import ReadCSV
from Buffer import Buffer
from DataLoader import DataLoader
from Fixer import Fixer
from Plots import Plots
import os
if os.path.exists('data.txt'):
os.remove('data.txt')
upload_data = []
data = []
line = 1
rfile = ReadCSV.read('test.csv')
for row in rfile:
if(line == 1):
line = line+1
continue
upload_data.append(row)
Fixer.fixTimeAndAlt(upload_data)
buff = Buffer()
for i in upload_data:
buff.sendToBuffer(i)
buff.sendData()
data = (DataLoader.read('data.txt'))
plots = Plots()
plots.altitude(data)
plots.pitch(data)
plots.distance(data)
plots.speed(data)
def model_train(Save = False, modelname = None):
X1_left = ReadCSV('datasets_csv/feature_0630_1700/X1_offline_0630_1700.csv','float64')
X1_midle = ReadCSV('datasets_csv/feature_0701_1100/X1_offline_0701_1100.csv','float64')
X1_right = ReadCSV('datasets_csv/feature_0701_1400/X1_offline_0701_1400.csv','float64')
Y1 = ReadCSV('datasets_csv/feature_0701_1100/Y1_offline_0701_1100.csv','float64')
X1 = np.c_[X1_left,X1_midle,X1_right]
# X1 = np.c_[X1_left]
X1 = list(X1)
Y1 = list(Y1)
X2_left = ReadCSV('datasets_csv/feature_0630_1700/X2_offline_0630_1700.csv','float64')
X2_midle = ReadCSV('datasets_csv/feature_0701_1100/X2_offline_0701_1100.csv','float64')
X2_right = ReadCSV('datasets_csv/feature_0701_1400/X2_offline_0701_1400.csv','float64')
Y2 = ReadCSV('datasets_csv/feature_0701_1100/Y2_offline_0701_1100.csv','float64')
X2 = np.c_[X2_left,X2_midle,X2_right]
# X2 = np.c_[X2_left]
X2 = list(X2)
Y2 = list(Y2)
X1.extend(X2)
Y1.extend(Y2)
del X2,Y2
X3_left = ReadCSV('datasets_csv/feature_0630_1700/X3_offline_0630_1700.csv','float64')
X3_midle = ReadCSV('datasets_csv/feature_0701_1100/X3_offline_0701_1100.csv','float64')
X3_right = ReadCSV('datasets_csv/feature_0701_1400/X3_offline_0701_1400.csv','float64')
Y3 = ReadCSV('datasets_csv/feature_0701_1100/Y3_offline_0701_1100.csv','float64')
X3 = np.c_[X3_left,X3_midle,X3_right]
# X3 = np.c_[X3_left]
X3 = list(X3)
Y3 = list(Y3)
X1.extend(X3)
Y1.extend(Y3)
del X3,Y3
X4_left = ReadCSV('datasets_csv/feature_0630_1700/X4_offline_0630_1700.csv','float64')
X4_midle = ReadCSV('datasets_csv/feature_0701_1100/X4_offline_0701_1100.csv','float64')
X4_right = ReadCSV('datasets_csv/feature_0701_1400/X4_offline_0701_1400.csv','float64')
Y4 = ReadCSV('datasets_csv/feature_0701_1100/Y4_offline_0701_1100.csv','float64')
X4 = np.c_[X4_left,X4_midle,X4_right]
# X4 = np.c_[X4_left]
X4 = list(X4)
Y4 = list(Y4)
X1.extend(X4)
Y1.extend(Y4)
del X4,Y4
X5_left = ReadCSV('datasets_csv/feature_0630_1700/X5_offline_0630_1700.csv','float64')
X5_midle = ReadCSV('datasets_csv/feature_0701_1100/X5_offline_0701_1100.csv','float64')
X5_right = ReadCSV('datasets_csv/feature_0701_1400/X5_offline_0701_1400.csv','float64')
Y5 = ReadCSV('datasets_csv/feature_0701_1100/Y5_offline_0701_1100.csv','float64')
X5 = np.c_[X5_left,X5_midle,X5_right]
X5 = list(X5)
Y5 = list(Y5)
X1.extend(X5)
Y1.extend(Y5)
del X5,Y5
X6_left = ReadCSV('datasets_csv/feature_0630_1700/X6_offline_0630_1700.csv','float64')
X6_midle = ReadCSV('datasets_csv/feature_0701_1100/X6_offline_0701_1100.csv','float64')
X6_right = ReadCSV('datasets_csv/feature_0701_1400/X6_offline_0701_1400.csv','float64')
Y6 = ReadCSV('datasets_csv/feature_0701_1100/Y6_offline_0701_1100.csv','float64')
X6 = np.c_[X6_left,X6_midle,X6_right]
X6 = list(X6)
Y6 = list(Y6)
X1.extend(X6)
Y1.extend(Y6)
del X6,Y6
X7_left = ReadCSV('datasets_csv/feature_0630_1700/X7_offline_0630_1700.csv','float64')
X7_midle = ReadCSV('datasets_csv/feature_0701_1100/X7_offline_0701_1100.csv','float64')
X7_right = ReadCSV('datasets_csv/feature_0701_1400/X7_offline_0701_1400.csv','float64')
Y7 = ReadCSV('datasets_csv/feature_0701_1100/Y7_offline_0701_1100.csv','float64')
X7 = np.c_[X7_left,X7_midle,X7_right]
X7 = list(X7)
Y7 = list(Y7)
X1.extend(X7)
Y1.extend(Y7)
del X7,Y7
X8_left = ReadCSV('datasets_csv/feature_0630_1700/X8_offline_0630_1700.csv','float64')
X8_midle = ReadCSV('datasets_csv/feature_0701_1100/X8_offline_0701_1100.csv','float64')
X8_right = ReadCSV('datasets_csv/feature_0701_1400/X8_offline_0701_1400.csv','float64')
Y8 = ReadCSV('datasets_csv/feature_0701_1100/Y8_offline_0701_1100.csv','float64')
X8 = np.c_[X8_left,X8_midle,X8_right]
X8 = list(X8)
Y8 = list(Y8)
X1.extend(X8)
Y1.extend(Y8)
del X8,Y8
X1 = np.array(X1)
Y1 = np.array(Y1)
X1,Y1 = DowmSample(X1,Y1,8)
model = GradientBoostingClassifier(n_estimators=100,max_leaf_nodes=5, learning_rate=0.1, subsample=0.8, random_state=1)
model.fit(X1, Y1.ravel())
# 保存模型
if Save == True:
f = open(modelname,'w')
pickle.dump(model, f)
f.close()
print '\n ------------- Training is over ----------------------'
return model
accuracies = []
for i in range(self.turns):
print(" turn ", i)
validationSet = self.create_validation_set(div_db, i)
#print(len(validationSet))
adaboost_set = []
for j in range(self.turns):
if (j != i):
adaboost_set += div_db[j]
adaboost_set = [self.attr_list] + adaboost_set
fscr, accuracy_per_turn = self.fscr_per_turn(
adaboost_set, validationSet)
#print(fscr)
f1Scores.append(fscr)
accuracies.append(accuracy_per_turn)
#print(len(f1Scores))
fscore = sum(f1Scores) / float(self.turns)
acc_score = sum(accuracies) / float(self.turns)
print("the f1 score of simulation: ", fscore)
print("the accuracy of simulation: ", acc_score)
##################################################################################
rd = ReadCSV()
db_1 = rd.produceDB()
k = KFoldCross(db_1, 5)
#print(k.block_size)
k.validation()
