def __reservoirTrain__(self, x):
#Extract the parameters
spectralRadius = x[0]
inputScaling = x[1]
reservoirScaling = x[2]
leakingRate = x[3]
#Create the reservoir
res = classicESN.Reservoir(size=self.size,
spectralRadius=spectralRadius,
inputScaling=inputScaling,
reservoirScaling=reservoirScaling,
leakingRate=leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=self.inputWeightRandom,
reservoirWeightRandom=self.reservoirWeightRandom)
#Train the reservoir
res.trainReservoir()
#Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed, self.horizon)
#Calcuate the regression error
errorFunction = rmse.MeanSquareError()
regressionError = errorFunction.compute(self.validationOutputData, predictedOutputData)
#Return the error
print("\nThe Parameters: "+str(x)+" Regression error:"+str(regressionError))
return regressionError
def __reservoirTrain__(self, x):
#Extract the parameters
meanDegree, beta = x
meanDegree = int(meanDegree)
# To get rid off the randomness in assigning weights, run it 10 times and take the average error
times = 100
cumulativeError = 0
for i in range(times):
# Input and weight connectivity Matrix
inputWeightMatrix = topology.ClassicInputTopology(
self.inputD, self.size).generateWeightMatrix()
reservoirWeightMatrix = topology.SmallWorldGraphs(
size=self.size, meanDegree=meanDegree,
beta=beta).generateWeightMatrix()
#Create the reservoir
res = classicESN.Reservoir(
size=self.size,
spectralRadius=self.spectralRadius,
inputScaling=self.inputScaling,
reservoirScaling=self.reservoirScaling,
leakingRate=self.leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=inputWeightMatrix,
reservoirWeightRandom=reservoirWeightMatrix)
#Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(
self.trainingInputData[-self.initialTransient:])
#Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed,
self.horizon)
gc.collect()
#Calcuate the regression error
errorFunction = metrics.MeanSquareError()
error = errorFunction.compute(self.validationOutputData,
predictedOutputData)
cumulativeError += error
regressionError = cumulativeError / times
#Return the error
print("\nThe Parameters: " + str(x) + " Regression error:" +
str(regressionError))
return regressionError
def evaluate(self, x):
# Extract the parameters
attachment = int(x[0, 0])
# To get rid off the randomness in assigning weights, run it 10 times and take the average error
times = 1
cumulativeError = 0
for i in range(times):
# Input and weight connectivity Matrix
inputWeightMatrix = topology.ClassicInputTopology(
self.inputD, self.size).generateWeightMatrix()
network = topology.ScaleFreeNetworks(size=self.size,
attachmentCount=attachment)
reservoirWeightMatrix = network.generateWeightMatrix()
# Create the reservoir
res = esn.Reservoir(size=self.size,
spectralRadius=self.spectralRadius,
inputScaling=self.inputScaling,
reservoirScaling=self.reservoirScaling,
leakingRate=self.leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=inputWeightMatrix,
reservoirWeightRandom=reservoirWeightMatrix)
# Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(
self.trainingInputData[-self.initialTransient:])
# Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed,
self.horizon)
gc.collect()
# Calculate the regression error
errorFunction = metrics.MeanSquareError()
error = errorFunction.compute(self.validationOutputData,
predictedOutputData)
cumulativeError += error
regressionError = cumulativeError / times
# Return the error
#print("Attachment: "+str(attachment) + "Error: "+str(regressionError))
return regressionError
def evaluate(self, x):
# Extract the parameters
meanDegree = int(x[0,0])
beta = x[0,1]
# To get rid off the randomness in assigning weights, run it 10 times and take the average error
times = 1
cumulativeError = 0
for i in range(times):
# Input and weight connectivity Matrix
inputWeightMatrix = topology.ClassicInputTopology(self.inputD, self.size).generateWeightMatrix()
network = topology.SmallWorldGraphs(size=self.size, meanDegree=meanDegree, beta=beta)
reservoirWeightMatrix = network.generateWeightMatrix()
# Create the reservoir
res = esn.Reservoir(size=self.size,
spectralRadius=self.spectralRadius,
inputScaling=self.inputScaling,
reservoirScaling=self.reservoirScaling,
leakingRate=self.leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=inputWeightMatrix,
reservoirWeightRandom=reservoirWeightMatrix)
# Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(self.trainingInputData[-self.initialTransient:])
# Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed, self.horizon)
gc.collect()
# Calculate the regression error
errorFunction = metrics.MeanSquareError()
error = errorFunction.compute(self.validationOutputData, predictedOutputData)
cumulativeError += error
regressionError = cumulativeError/times
# Return the error
#print("SMG parameters: "+str(x) + "Error: "+str(regressionError))
return regressionError
def __reservoirTrain__(self, x):
#Extract the parameters
attachment = int(x)
# To get rid off the randomness in assigning weights, run it 10 times and take the average error
times = 100
cumulativeError = 0
for i in range(times):
# Input and weight connectivity Matrix
inputWeightMatrix = topology.ClassicInputTopology(self.inputD, self.size).generateWeightMatrix()
reservoirWeightMatrix = topology.ScaleFreeNetworks(size=self.size, attachmentCount=attachment).generateWeightMatrix()
#Create the reservoir
res = classicESN.Reservoir(size=self.size,
spectralRadius=self.spectralRadius,
inputScaling=self.inputScaling,
reservoirScaling=self.reservoirScaling,
leakingRate=self.leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=inputWeightMatrix,
reservoirWeightRandom=reservoirWeightMatrix)
#Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(self.trainingInputData[-self.initialTransient:])
#Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed, self.horizon)
gc.collect()
#Calcuate the regression error
errorFunction = metrics.MeanSquareError()
error = errorFunction.compute(self.validationOutputData, predictedOutputData)
cumulativeError += error
regressionError = cumulativeError/times
#Return the error
print("\nThe Parameters: "+str(x)+" Regression error:"+str(regressionError))
return regressionError
def __reservoirTrain__(self, x):
#Extract the parameters
spectralRadius = x[0]
inputScaling = x[1]
reservoirScaling = x[2]
leakingRate = x[3]
#Create the reservoir
res = classicESN.Reservoir(
size=self.size,
spectralRadius=spectralRadius,
inputScaling=inputScaling,
reservoirScaling=reservoirScaling,
leakingRate=leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=self.inputWeightRandom,
reservoirWeightRandom=self.reservoirWeightRandom)
#Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(
self.trainingInputData[-self.initialTransient:])
#Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed,
self.horizon)
gc.collect()
#Calcuate the regression error
errorFunction = metrics.MeanSquareError()
regressionError = errorFunction.compute(self.validationOutputData,
predictedOutputData)
#Return the error
print("\nThe Parameters: " + str(x) + " Regression error:" +
str(regressionError))
return regressionError
def evaluate(self, x):
#Extract the parameters
spectralRadius = x[0, 0]
inputScaling = x[0, 1]
reservoirScaling = x[0, 2]
leakingRate = x[0, 3]
#Create the reservoir
res = esn.Reservoir(size=self.size,
spectralRadius=spectralRadius,
inputScaling=inputScaling,
reservoirScaling=reservoirScaling,
leakingRate=leakingRate,
initialTransient=self.initialTransient,
inputData=self.trainingInputData,
outputData=self.trainingOutputData,
inputWeightRandom=self.inputWeightRandom,
reservoirWeightRandom=self.reservoirWeightRandom)
#Train the reservoir
res.trainReservoir()
# Warm up
predictedTrainingOutputData = res.predict(self.trainingInputData[-self.initialTransient:])
#Predict for the validation data
predictedOutputData = util.predictFuture(res, self.initialSeed, self.horizon)
gc.collect()
#Calcuate the regression error
errorFunction = metrics.MeanSquareError()
regressionError = errorFunction.compute(self.validationOutputData, predictedOutputData)
#Return the error
return regressionError
inputData=inputTrainingData,
outputData=outputTrainingData,
spectralRadius=0.79,
inputScaling=0.5,
reservoirScaling=0.5,
leakingRate=0.3,
initialTransient=initialTransient,
inputWeightRandom=inputWeight,
reservoirWeightRandom=reservoirWeight)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(inputTrainingData)
#Predict future values
predictedTestOutputData = util.predictFuture(res, trainingData[-1], nTesting)
#Calculate the error
errorFunction = rmse.MeanSquareError()
error = errorFunction.compute(testingData, predictedTestOutputData)
print("Regression error:" + str(error))
#Plotting of the prediction output and error
outputFolderName = "Outputs/Outputs" + datetime.now().strftime(
"%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html",
"Mackey-Glass Time Series - Classic ESN",
"Prediction of future values", "Time",
"Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
inputScalingBound=inputScalingBound,
reservoirScalingBound=reservoirScalingBound,
leakingRateBound=leakingRateBound)
spectralRadiusOptimum, inputScalingOptimum, reservoirScalingOptimum, leakingRateOptimum, inputWeightOptimum, reservoirWeightOptimum = resTuner.getOptimalParameters()
#Train
res = reservoir.Reservoir(size=size,
spectralRadius=spectralRadiusOptimum,
inputScaling=inputScalingOptimum,
reservoirScaling=reservoirScalingOptimum,
leakingRate=leakingRateOptimum,
initialTransient=initialTransient,
inputData=trainingInputData,
outputData=trainingOutputData,
inputWeightRandom=inputWeightOptimum,
reservoirWeightRandom=reservoirWeightOptimum)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(validationInputData)
# Predict
predictedTestOutputData = util.predictFuture(res, validationData[-1], nTesting)
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot("Outputs/Prediction.html", "Mackey-Glass Time Series", "Prediction of future values", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testingData[:nTesting, 0])
outplot.setYSeries('Predicted Output', predictedTestOutputData[:nTesting, 0])
outplot.createOutput()
print("Done!")
inputData=inputTrainingData,
outputData=outputTrainingData,
spectralRadius=0.79,
inputScaling=0.5,
reservoirScaling=0.5,
leakingRate=0.3,
initialTransient=initialTransient,
inputWeightRandom=inputWeight,
reservoirWeightRandom=reservoirWeight)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(inputTrainingData)
#Predict future values
predictedTestOutputData = util.predictFuture(res, trainingData[-1], nTesting)
#Calculate the error
errorFunction = rmse.MeanSquareError()
error = errorFunction.compute(testingData, predictedTestOutputData)
print("Regression error:"+str(error))
#Plotting of the prediction output and error
outputFolderName = "Outputs/Outputs" + datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html", "Mackey-Glass Time Series - Classic ESN", "Prediction of future values", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', minMax.inverse_transform(testingData[:nTesting, 0]))
outplot.setYSeries('Predicted Output', minMax.inverse_transform(predictedTestOutputData[:nTesting, 0]))
outplot.createOutput()
)
#Train
res = reservoir.Reservoir(size=size,
spectralRadius=spectralRadiusOptimum,
inputScaling=inputScalingOptimum,
reservoirScaling=reservoirScalingOptimum,
leakingRate=leakingRateOptimum,
initialTransient=initialTransient,
inputData=trainingInputData,
outputData=trainingOutputData,
inputWeightRandom=inputWeightOptimum,
reservoirWeightRandom=reservoirWeightOptimum)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(validationInputData)
# Predict
predictedTestOutputData = util.predictFuture(res, validationData[-1], nTesting)
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot("Outputs/Prediction.html",
"Mackey-Glass Time Series",
"Prediction of future values", "Time",
"Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testingData[:nTesting, 0])
outplot.setYSeries('Predicted Output', predictedTestOutputData[:nTesting, 0])
outplot.createOutput()
print("Done!")
reservoirScaling=0.5,
leakingRate=0.3,
initialTransient=initialTransient,
inputWeightRandom=inputWeight,
reservoirWeightRandom=reservoirWeight)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(
inputTrainingData[-initialTransient:])
#Future - Non-accumulative
#predictedTestOutputData = res.predict(inputTestingData)
#Predict future values
predictedTestOutputData = util.predictFuture(res, testingData[0], nTesting)
#Calculate the error
# errorFunction = rmse.MeanSquareError()
# error = errorFunction.compute(testingData, predictedTestOutputData)
# print("Regression error:"+str(error))
#Plotting of the prediction output and error
# nTraining = inputTrainingData.shape[0]
# outputFolderName = "Outputs/Outputs" + datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
# os.mkdir(outputFolderName)
# outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html", "Mackey-Glass Time Series - Classic ESN", "Prediction of future values", "Time", "Output")
# outplot.setXSeries(np.arange(1, nTraining + 1))
# outplot.setYSeries('Actual Output', minMax.inverse_transform(outputTrainingData[:nTraining, 0]))
# outplot.setYSeries('Predicted Output', minMax.inverse_transform(predictedTrainingOutputData[:nTraining, 0]))
# outplot.createOutput()
inputScaling=0.5,
reservoirScaling=0.5,
leakingRate=0.3,
initialTransient=initialTransient,
inputWeightRandom=inputWeight,
reservoirWeightRandom=reservoirWeight)
res.trainReservoir()
#Warm up
predictedTrainingOutputData = res.predict(inputTrainingData[-initialTransient:])
#Future - Non-accumulative
#predictedTestOutputData = res.predict(inputTestingData)
#Predict future values
predictedTestOutputData = util.predictFuture(res, testingData[0], nTesting)
#Calculate the error
# errorFunction = rmse.MeanSquareError()
# error = errorFunction.compute(testingData, predictedTestOutputData)
# print("Regression error:"+str(error))
#Plotting of the prediction output and error
# nTraining = inputTrainingData.shape[0]
# outputFolderName = "Outputs/Outputs" + datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
# os.mkdir(outputFolderName)
# outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html", "Mackey-Glass Time Series - Classic ESN", "Prediction of future values", "Time", "Output")
# outplot.setXSeries(np.arange(1, nTraining + 1))
# outplot.setYSeries('Actual Output', minMax.inverse_transform(outputTrainingData[:nTraining, 0]))
# outplot.setYSeries('Predicted Output', minMax.inverse_transform(predictedTrainingOutputData[:nTraining, 0]))
resOnline = onlineESN.Reservoir(size=size,
inputData=inputTrainingData,
outputData=outputTrainingData,
spectralRadius=0.79,
inputScaling=0.5,
reservoirScaling=0.5,
leakingRate=0.3,
initialTransient=initialTransient,
inputWeightRandom=inputWeight,
reservoirWeightRandom=reservoirWeight)
resOnline.trainReservoir()
#Warm up
predictedTrainingOutputData = resBatch.predict(inputTrainingData[-initialTransient:])
predictedTrainingOutputData = resOnline.predict(inputTrainingData[-initialTransient:])
#Predict future values
predictedTestOutputDataBatch = util.predictFuture(resBatch, trainingData[-1], nTesting)
predictedTestOutputDataOnline = util.predictFuture(resOnline, trainingData[-1], nTesting)
#Plotting of the prediction output and error
outputFolderName = "Outputs/Outputs" + datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html", "Mackey-Glass Time Series - Comparison of Batch vs Online", "Prediction of future values", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', minMax.inverse_transform(testingData[:nTesting, 0]))
outplot.setYSeries('Predicted Batch learning Output', minMax.inverse_transform(predictedTestOutputDataBatch[:nTesting, 0]))
outplot.setYSeries('Predicted Online RLS learning Output', minMax.inverse_transform(predictedTestOutputDataOnline[:nTesting, 0]))
outplot.createOutput()
print("Done!")
