def test():
DS = loadPybrainData()
train, test = DS.splitWithProportion(0.1)
fnn = joblib.load(PKL)
# 预测test情况
output = fnn.activateOnDataset(test)
# ann.activate(onedata)可以只对一个数据进行预测
outputs = []
target = []
count = 0
for out in output:
outs = out.argmax()
outputs.append(outs)
for tar in test['target']:
ta = tar.argmax()
target.append(ta)
for i in range(0, len(target)):
if outputs[i] == target[i]:
count += 1
right = count / len(target) #单个字符正确率
rate = (right**4)
print("分类正确率是：%.4f%%" % (rate * 100))
v = Validator()
print(u'均方和差为：', v.MSE(output,
test['target'])) #计算test的原始值和预测值的均方差和,两者格式必须相等
def dataset_eval(dataset):
"""Return dataset hit rate and MSE"""
# Transform output values to bit vectors, similar to the targets
predicted = bit_array_transform(ff_network.activate(x)
for x in dataset['input'])
target = dataset['target']
# Lists of positions holding predicted and target classes to compare
predicted_pos = [list(x).index(1) for x in predicted]
target_pos = [list(x).index(1) for x in target]
hits = Validator.classificationPerformance(predicted_pos, target_pos)
mse = Validator.MSE(predicted, target)
return hits, mse
def nn():
DS = ClassificationDataSet(28, 1, nb_classes=4)
train = pickle.load(open('train_extracted_df.pkl', 'r'))
y = train["median_relevance"]
kfold_train_test = pickle.load(open('kfold_train_test.pkl', 'r'))
features = ['query_tokens_in_title', 'query_tokens_in_description', 'percent_query_tokens_in_description', 'percent_query_tokens_in_title', 'query_length', 'description_length', 'title_length', 'two_grams_in_q_and_t', 'two_grams_in_q_and_d', 'q_mean_of_training_relevance', 'q_median_of_training_relevance', 'avg_relevance_variance', 'average_title_1gram_similarity_1', 'average_title_2gram_similarity_1', 'average_title_1gram_similarity_2', 'average_title_2gram_similarity_2', 'average_title_1gram_similarity_3', 'average_title_2gram_similarity_3', 'average_title_1gram_similarity_4', 'average_title_2gram_similarity_4', 'average_description_1gram_similarity_1', 'average_description_2gram_similarity_1', 'average_description_2gram_similarity_2', 'average_description_1gram_similarity_2', 'average_description_1gram_similarity_3', 'average_description_2gram_similarity_3', 'average_description_1gram_similarity_4', 'average_description_2gram_similarity_4']
train = train[features]
for i in range(len(y)):
DS.addSample(train.values[i],y[i])
X = DS['input']
Y = DS['target']
dataTrain, dataTest = DS.splitWithProportion(0.8)
xTrain, yTrain = dataTrain['input'], dataTrain['target']
xTest, yTest = dataTest['input'], dataTest['target']
#fnn = RecurrentNetwork()
fnn = FeedForwardNetwork()
#fnn=buildNetwork(1,40,1,hiddenclass=TanhLayer, bias=True, outclass=SoftmaxLayer)
#fnn=buildNetwork(1,40,1,hiddenclass=LSTMLayer, bias=True, outclass=SoftmaxLayer)
inLayer = LinearLayer(28, name='inLayer')
hiddenLayer = SigmoidLayer(40, name='hiddenLayer0')
outLayer =LinearLayer(4, name='outLayer')
fnn.addInputModule(inLayer)
fnn.addModule(hiddenLayer)
fnn.addOutputModule(outLayer)
in_to_hidden = FullConnection(inLayer, hiddenLayer)
hidden_to_out = FullConnection(hiddenLayer, outLayer)
fnn.addConnection(in_to_hidden)
fnn.addConnection(hidden_to_out)
fnn.sortModules()
trainer = BackpropTrainer(fnn, DS, verbose = True, learningrate=0.01)
#trainer.trainUntilConvergence(maxEpochs=1000)
trainer.trainEpochs(epochs=5)
prediction = fnn.activateOnDataset(dataTest)
out=[]
total_score = 0
for i in prediction:
class_index = max(xrange(len(i)), key=i.__getitem__)
out.append(class_index+1)
print str((class_index+1-yTest[class_index+1])/yTest[class_index+1])
df=pd.DataFrame(out,columns=['predict'])
df['real']=dataTest['target']
coun = 0
for i,row in df.iterrows():
if row[0]== row[1]:
coun+=1
print coun
print "df['real']", df['real'],type(df['real'][0])
print "df['predict']",df['predict'],type(df['predict'][0])
print df
v=Validator()
#v.MSE(out,dataTest['target'])
print "out",out
print "dataTest['target']",dataTest['target']
def evalRNNOnSeqClassificationDataset(net, testing_dataset, verbose = False, silent = False):
# Fetch targets and calculate the modules output on dataset.
# Output and target are in one-of-many format. The class for each sequence is
# determined by first summing the probabilities for each individual sample over
# the sequence, and then finding its maximum.
target = testing_dataset.getField("target")
outputs = []
# print net
for seq in testing_dataset._provideSequences():
net.reset()
# print 'seq:'
# print seq
for i in xrange(len(seq)):
output = net.activate(seq[i][0])
outputs.append(output.copy())
outputs = array(outputs)
# determine last indices of the sequences inside dataset
ends = SequenceHelper.getSequenceEnds(testing_dataset)
##format = "%d"*len(ends)
summed_output = zeros(testing_dataset.outdim)
# class_output and class_target will store class labels instead of
# one-of-many values
class_output = []
class_target = []
for j in xrange(len(outputs)):
# sum up the output values of one sequence
# print outputs[j]
summed_output += outputs[j]
# print j, output[j], " --> ", summed_output
# if we reached the end of the sequence
if j in ends:
# print '------------------------------------------'
# convert summed_output and target to class labels
class_output.append(argmax(summed_output))
class_target.append(argmax(target[j]))
# reset the summed_output to zeros
summed_output = zeros(testing_dataset.outdim)
##print format % tuple(class_output)
##print format % tuple(class_target)
class_output = array(class_output)
class_target = array(class_target)
# print class_target
# print class_output
accuracy = Validator.classificationPerformance(class_output, class_target)
return (class_output, class_target, accuracy)
def testOnSequenceData(module, dataset):
"""
Fetch targets and calculate the modules output on dataset.
Output and target are in one-of-many format. The class for each sequence is
determined by argmax OF THE LAST ITEM IN THE SEQUENCE.
"""
target = dataset.getField("target")
output = ModuleValidator.calculateModuleOutput(module, dataset)
# determine last indices of the sequences inside dataset
ends = SequenceHelper.getSequenceEnds(dataset)
class_output = array([argmax(output[end]) for end in ends])
class_target = array([argmax(target[end]) for end in ends])
return Validator.classificationPerformance(class_output, class_target)
def __init__(self, evolino_network, dataset, **kwargs):
""" @param evolino_network: an instance of NetworkWrapper()
@param dataset: The evaluation dataset
@param evalfunc: Compares output to target values and returns a scalar, denoting the fitness.
Defaults to -mse(output, target).
@param wtRatio: Float array of two values denoting the ratio between washout and training length.
Defaults to [1,2]
@param verbosity: Verbosity level. Defaults to 0
"""
Filter.__init__(self)
ap = KWArgsProcessor(self, kwargs)
ap.add( 'verbosity', default=0 )
ap.add( 'evalfunc', default=lambda output, target: -Validator.MSE(output, target) )
ap.add( 'wtRatio', default=array([1,2], float) )
self.network = evolino_network
self.dataset = dataset
self.max_fitness = -Infinity
def runTraining(self, convergence=0, **kwargs):
""" Trains the network on the stored dataset. If convergence is >0, check after that many epoch increments
whether test error is going down again, and stop training accordingly. """
assert isinstance(self.Trainer, Trainer)
if self.Graph is not None:
self.Graph.setLabels(x='epoch', y='% classification error')
self.Graph.setLegend(['training','test'],loc='lower right')
epoch = 0
inc = self.epoinc
best_error = 100.0
best_epoch = 0
learncurve_x = [0]
learncurve_y = [0.0]
valcurve_y = [0.0]
converged = False
convtest = 0
if convergence>0:
logging.info("Convergence criterion: %d batches of %d epochs w/o improvement" % (convergence, inc))
while epoch<=self.maxepochs and not converged:
self.Trainer.trainEpochs(inc)
epoch+=inc
learncurve_x.append(epoch)
# calculate errors on TRAINING data
if isinstance(self.DS, SequentialDataSet):
r_trn = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.DS))
else:
# FIXME: messy - validation does not belong into the Trainer...
out, trueclass = self.Trainer.testOnClassData(return_targets=True)
r_trn = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
learncurve_y.append(r_trn)
if self.TDS is None:
logging.info("epoch: %6d, err_trn: %5.2f%%" % (epoch, r_trn))
else:
# calculate errors on TEST data
if isinstance(self.DS, SequentialDataSet):
r_tst = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.TDS))
else:
# FIXME: messy - validation does not belong into the Trainer...
out, trueclass = self.Trainer.testOnClassData(return_targets=True, dataset=self.TDS)
r_tst = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
valcurve_y.append(r_tst)
if r_tst < best_error:
best_epoch = epoch
best_error = r_tst
bestweights = self.Trainer.module.params.copy()
convtest = 0
else:
convtest += 1
logging.info("epoch: %6d, err_trn: %5.2f%%, err_tst: %5.2f%%, best_tst: %5.2f%%" % (epoch, r_trn, r_tst, best_error))
if self.Graph is not None:
self.Graph.addData(1, epoch, r_tst)
# check if convegence criterion is fulfilled (no improvement after N epoincs)
if convtest >= convergence:
converged = True
if self.Graph is not None:
self.Graph.addData(0, epoch, r_trn)
self.Graph.update()
logging.info("Best epoch: %6d, with error: %5.2f%%" % (best_epoch, best_error))
if self.VDS is not None:
# calculate errors on VALIDATION data
self.Trainer.module.params[:] = bestweights.copy()
if isinstance(self.DS, SequentialDataSet):
r_val = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.VDS))
else:
out, trueclass = self.Trainer.testOnClassData(return_targets=True, dataset=self.VDS)
r_val = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
logging.info("Result on evaluation data: %5.2f%%" % r_val)
self.trainCurve = (learncurve_x, learncurve_y, valcurve_y)
def runTraining(self, convergence=0, **kwargs):
""" Trains the network on the stored dataset. If convergence is >0, check after that many epoch increments
whether test error is going down again, and stop training accordingly. """
assert isinstance(self.Trainer, Trainer)
if self.Graph is not None:
self.Graph.setLabels(x='epoch', y='% classification error')
self.Graph.setLegend(['training','test'],loc='lower right')
epoch = 0
inc = self.epoinc
best_error = 100.0
best_epoch = 0
learncurve_x = [0]
learncurve_y = [0.0]
valcurve_y = [0.0]
converged = False
convtest = 0
if convergence>0:
logging.info("Convergence criterion: %d batches of %d epochs w/o improvement" % (convergence, inc))
while epoch<=self.maxepochs and not converged:
self.Trainer.trainEpochs(inc)
epoch+=inc
learncurve_x.append(epoch)
# calculate errors on TRAINING data
if isinstance(self.DS, SequentialDataSet):
r_trn = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.DS))
else:
# FIXME: messy - validation does not belong into the Trainer...
out, trueclass = self.Trainer.testOnClassData(return_targets=True)
r_trn = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
learncurve_y.append(r_trn)
if self.TDS is None:
logging.info("epoch: %6d, err_trn: %5.2f%%" % (epoch, r_trn))
else:
# calculate errors on TEST data
if isinstance(self.DS, SequentialDataSet):
r_tst = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.TDS))
else:
# FIXME: messy - validation does not belong into the Trainer...
out, trueclass = self.Trainer.testOnClassData(return_targets=True, dataset=self.TDS)
r_tst = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
valcurve_y.append(r_tst)
if r_tst < best_error:
best_epoch = epoch
best_error = r_tst
bestweights = self.Trainer.module.params.copy()
convtest = 0
else:
convtest += 1
logging.info("epoch: %6d, err_trn: %5.2f%%, err_tst: %5.2f%%, best_tst: %5.2f%%" % (epoch, r_trn, r_tst, best_error))
if self.Graph is not None:
self.Graph.addData(1, epoch, r_tst)
# check if convegence criterion is fulfilled (no improvement after N epoincs)
if convtest >= convergence:
converged = True
if self.Graph is not None:
self.Graph.addData(0, epoch, r_trn)
self.Graph.update()
logging.info("Best epoch: %6d, with error: %5.2f%%" % (best_epoch, best_error))
if self.VDS is not None:
# calculate errors on VALIDATION data
self.Trainer.module.params[:] = bestweights.copy()
if isinstance(self.DS, SequentialDataSet):
r_val = 100. * (1.0-testOnSequenceData(self.Trainer.module, self.VDS))
else:
out, trueclass = self.Trainer.testOnClassData(return_targets=True, dataset=self.VDS)
r_val = 100. * (1.0-Validator.classificationPerformance(out, trueclass))
logging.info("Result on evaluation data: %5.2f%%" % r_val)
self.trainCurve = (learncurve_x, learncurve_y, valcurve_y)
pylab.plot(tsts['input'], ctsts, c='g')
pylab.xlabel('x')
pylab.ylabel('y')
pylab.title('Neuron Number:' + str(nneuron))
pylab.grid(True)
plotname = os.path.join(plotdir, ('jpq2layers_plot' + str(iter)))
pylab.savefig(plotname)
# set-up the neural network
nneuron = 5
mom = 0.98
netname = "LSL-" + str(nneuron) + "-" + str(mom)
mv = ModuleValidator()
v = Validator()
#create the test DataSet
x = numpy.arange(0.0, 1.0 + 0.01, 0.01)
s = 0.5 + 0.4 * numpy.sin(2 * numpy.pi * x)
tsts = SupervisedDataSet(1, 1)
tsts.setField('input', x.reshape(len(x), 1))
tsts.setField('target', s.reshape(len(s), 1))
#read the train DataSet from file
trndata = SupervisedDataSet.loadFromFile(os.path.join(os.getcwd(), 'trndata'))
myneuralnet = os.path.join(os.getcwd(), 'myneuralnet.xml')
if os.path.isfile(myneuralnet):
n = NetworkReader.readFrom(myneuralnet, name=netname)
#calculate the test DataSet based on the trained Neural Network
ctsts = mv.calculateModuleOutput(n, tsts)
pylab.plot(tsts['input'], ctsts, c='g')
pylab.xlabel('x')
pylab.ylabel('y')
pylab.title('Neuron Number:' + str(nneuron))
pylab.grid(True)
plotname = os.path.join(plotdir, ('jpq2layers_plot' + str(iter)))
pylab.savefig(plotname)
# set-up the neural network
nneuron = 5
mom = 0.98
netname = "LSL-" + str(nneuron) + "-" + str(mom)
mv = ModuleValidator()
v = Validator()
n = FeedForwardNetwork(name=netname)
inLayer = LinearLayer(1, name='in')
hiddenLayer = SigmoidLayer(nneuron, name='hidden0')
outLayer = LinearLayer(1, name='out')
biasinUnit = BiasUnit(name="bhidden0")
biasoutUnit = BiasUnit(name="bout")
n.addInputModule(inLayer)
n.addModule(hiddenLayer)
n.addModule(biasinUnit)
n.addModule(biasoutUnit)
n.addOutputModule(outLayer)
in_to_hidden = FullConnection(inLayer, hiddenLayer)
bias_to_hidden = FullConnection(biasinUnit, hiddenLayer)
bias_to_out = FullConnection(biasoutUnit, outLayer)
hidden_to_out = FullConnection(hiddenLayer, outLayer)