nTesting, 1)
#Predict
testPredictedOutputDataStandard = minMax.inverse_transform(
testingPredictedOutputData)
actual = minMax.inverse_transform(testActualOutputData)
standardError = errorFunction.compute(
actual.reshape((actual.shape[0], 1)),
testPredictedOutputDataStandard.reshape(
(testPredictedOutputDataStandard.shape[0], 1)))
testActualOutputData = minMax.inverse_transform(testActualOutputData[:nTesting,
0])
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot(
"Outputs/Prediction.html", "Darwin Sea Level Pressure Prediction",
"Comparison of Random Graph Topolgies - Standard Vs Erdos", "Time",
"Sea Level Pressure")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testActualOutputData)
outplot.setYSeries('Predicted Output_standard_ESN_with_parameters_tuned',
testPredictedOutputDataStandard)
outplot.setYSeries('Predicted Output_Erdoys_ESN_with_parameters_tuned',
testPredictedOutputDataErdos)
outplot.createOutput()
#Plotting of regression error
errPlot = errorPlot.ErrorPlot("Outputs/RegressionError.html",
"Comparison of standard vs deterministic",
"with parameters tuner", "ESN Configuration",
"Total Error")
errPlot.setXAxis(np.array(['Standard', 'Erdos Renyi']))
# Predict
predictedTargetVectors = model.predict(featureVectors)
outputFolderName = "Outputs/Outputs" + datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
for i in range(horizon):
# Plot the actual and predicted
actual = targetVectors[:, i]
predicted = predictedTargetVectors[:, i]
# Descale
actual = util.scalingFunction.inverse_transform(actual)
predicted = util.scalingFunction.inverse_transform(predicted)
outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction_horizon"+str(i+1)+".html", "Facebook Fans Change - Linear Regression", "Taylor Swift", "Time", "Output")
outplot.setXSeries(np.arange(1, targetVectors.shape[0]))
outplot.setYSeries('Actual Output', actual)
outplot.setYSeries('Predicted Output', predicted)
outplot.createOutput()
print("Done!")
testingPredictedOutputData = np.array(testingPredictedOutputData).reshape(
nTesting, 1)
#Predict
testPredictedOutputDataStandard = minMax.inverse_transform(
testingPredictedOutputData)
standardError = errorFunction.compute(
testActualOutputData.reshape((testActualOutputData.shape[0], 1)),
testPredictedOutputDataStandard.reshape(
(testPredictedOutputDataStandard.shape[0], 1)))
testActualOutputData = minMax.inverse_transform(testActualOutputData[:nTesting,
0])
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot("Outputs/Prediction.html",
"Darwin Sea Level Pressure Prediction",
"Comparison of standard vs deterministic",
"Time", "Sea Level Pressure")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testActualOutputData)
outplot.setYSeries('Predicted Output_standard_ESN',
testPredictedOutputDataStandard)
for i in range(len(topologyObjects)):
seriesName = 'Predicted Output_' + topologyNames[i]
seriesData = topologyTestOutput[i]
outplot.setYSeries(seriesName, seriesData)
outplot.createOutput()
#Plotting of regression error
topologyNames.append('Standard')
topologyError.append(standardError)
errPlot = errorPlot.ErrorPlot("Outputs/RegressionError.html",
topologyTestOutput.append(testPredictedOutputData)
#Error
topologyError.append(
errorFunction.compute(
testActualOutputData.reshape((testActualOutputData.shape[0], 1)),
testPredictedOutputData.reshape(
(testPredictedOutputData.shape[0], 1))))
#De-normalize
testActualOutputData = minMax.inverse_transform(testActualOutputData[:nTesting,
0])
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot("Outputs/Prediction.html",
"Darwin Sea Level Pressure Prediction",
"Deterministic echo state networks", "Time",
"Sea Level Pressure")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testActualOutputData)
for i in range(len(topologyObjects)):
seriesName = 'Predicted Output_' + topologyNames[i]
seriesData = topologyTestOutput[i]
outplot.setYSeries(seriesName, seriesData)
outplot.createOutput()
#Plotting of regression error
topologyNames.append('Standard')
errPlot = errorPlot.ErrorPlot("Outputs/RegressionError.html",
"Deterministic echo state networks",
"with different topologies", "Topology",
"Total Error")
# Initial seed
initialSeedForValidation = trainingData[-1]
predictedOutputData = util.tuneTrainPredictConnectivity(
trainingInputData=trainingInputData,
trainingOutputData=trainingOutputData,
validationOutputData=validationData,
initialInputSeedForValidation=initialSeedForValidation,
horizon=nTesting,
resTopology=util.Topology.ScaleFreeNetworks)
predictedOutputData = minMax.inverse_transform(predictedOutputData)
#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 - Scale Free Networks (Tuned)",
"Prediction on Validation + Testing Set", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', actualOutputData)
outplot.setYSeries('Predicted Output', predictedOutputData)
outplot.createOutput()
endTime = time()
run_time = endTime - startTime
print("The run time:" + str(run_time))
print("Done!")
# Initial seed
initialSeedForValidation = trainingData[-1]
predictedOutputData = util.tuneTrainPredictConnectivityNonBrute(
trainingInputData=trainingInputData,
trainingOutputData=trainingOutputData,
validationOutputData=validationData,
initialInputSeedForValidation=initialSeedForValidation,
horizon=nTesting,
resTopology=util.Topology.SmallWorldGraphs)
predictedOutputData = minMax.inverse_transform(predictedOutputData)
#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 - Small World Graphs (Tuned)",
"Prediction on Validation + Testing Set", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', actualOutputData)
outplot.setYSeries('Predicted Output', predictedOutputData)
outplot.createOutput()
endTime = time()
run_time = endTime - startTime
print("The run time:" + str(run_time))
print("Done!")
trainingInputData, trainingOutputData = util.formFeatureVectors(trainingData)
actualOutputData = minMax.inverse_transform(testingData)[:, 0]
# Initial seed
initialSeedForValidation = trainingData[-1]
predictedOutputData = utilGA.tuneTrainPredictGA(
trainingInputData=trainingInputData,
trainingOutputData=trainingOutputData,
validationOutputData=validationData,
initialInputSeedForValidation=initialSeedForValidation,
testingData=actualOutputData)
predictedOutputData = minMax.inverse_transform(predictedOutputData)
#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 - GA Optimization",
"Predicted vs Actual", "Time", "Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', actualOutputData)
outplot.setYSeries('Predicted Output', predictedOutputData)
outplot.createOutput()
endTime = time()
run_time = endTime - startTime
print("The run time:" + str(run_time))
print("Done!")
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 - Small World Graphs 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()
print("Done!")
)
#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!")
initialInputSeedForValidation=initialSeedForValidation,
horizon=nTesting,
noOfBest=noOfBest,
resTopology=utilityGA.Topology.Random,
size=networkSize,
popSize=populationSize,
maxGeneration=noOfGenerations)
predictedOutputData = minMax.inverse_transform(predictedOutputData)
#Plotting of the prediction output and error
outputFolderName = "Outputs/Random_Graph_Outputs" + datetime.now().strftime(
"%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
outplot = outputPlot.OutputPlot(outputFolderName + "/Prediction.html",
"Mackey-Glass Time Series - GA Optimization)",
"Prediction on Testing data", "Time", "Output")
outplot.setXSeries(np.arange(1, nValidation + nTesting + 1))
outplot.setYSeries('Actual Output', actualOutputData)
outplot.setYSeries('Predicted Output', predictedOutputData)
outplot.createOutput()
# Plotting of the best population details
utilityGA.plotNetworkPerformance(bestPopulation,
topology=utilityGA.Topology.Random,
fileName=outputFolderName +
"/NetworkPerformance.html",
networkSize=networkSize)
# Store the best population in a file (for later analysis)
popFileName = outputFolderName + "/population.pkl"
# Form feature vectors for training data
trainingInputData, trainingOutputData = util.formFeatureVectors(trainingData)
actualOutputData = minMax.inverse_transform(testingData)[:,0]
# Initial seed
initialSeedForValidation = trainingData[-1]
predictedOutputData = util.tuneTrainPredictConnectivity(trainingInputData=trainingInputData,
trainingOutputData=trainingOutputData,
validationOutputData=validationData,
initialInputSeedForValidation=initialSeedForValidation,
horizon = nTesting,
resTopology= util.Topology.ErdosRenyi)
predictedOutputData = minMax.inverse_transform(predictedOutputData)
#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 - Erdos Renyi (Tuned)", "Prediction on Validation + Testing Set", "Time", "Output")
outplot.setXSeries(np.arange(1, nValidation + nTesting + 1))
outplot.setYSeries('Actual Output', actualOutputData)
outplot.setYSeries('Predicted Output', predictedOutputData)
outplot.createOutput()
endTime = time()
run_time = endTime - startTime
print("The run time:"+str(run_time))
print("Done!")
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 - Scale Free Networks 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()
print("Done!")
testPredictedOutputDataStandard, standardError = runStandardESN()
testPredictedOutputDataErdos, erdosError = runErdosESN()
testPredictedOutputDataSmallWorld, smallWorldError = runSmallWorld()
testPredictedOutputDataScaleFree, scaleFreeError = runScaleFree()
testActualOutputData = minMax.inverse_transform(testActualOutputData[:nTesting,
0])
outputFolderName = "Outputs/Outputs" + datetime.now().strftime(
"%Y_%m_%d_%H_%M_%S")
os.mkdir(outputFolderName)
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot(
outputFolderName + "/Prediction.html", "Mackey Glass Prediction",
"Comparison of Random Graph Topolgies - Standard Vs Erdos", "Time",
"Sea Level Pressure")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries('Actual Output', testActualOutputData)
outplot.setYSeries('Predicted Output_standard_ESN_with_all_parameters_tuned',
testPredictedOutputDataStandard)
outplot.setYSeries('Predicted Output_Erdoys_ESN_with_parameters_tuned',
testPredictedOutputDataErdos)
outplot.setYSeries('Predicted Output_Small World_ESN_with_parameters_tuned',
testPredictedOutputDataSmallWorld)
outplot.setYSeries('Predicted Output_Scale_Free_with_parameters_tuned',
testPredictedOutputDataScaleFree)
outplot.createOutput()
#Plotting of regression error
errPlot = errorPlot.ErrorPlot(outputFolderName + "/RegressionError.html",
lastAvailableData = testInputData[0, 1]
for i in range(nTesting):
query = [1.0]
query.append(lastAvailableData)
#Predict the next point
nextPoint = res.predict(np.array(query).reshape((1, 2)))[0, 0]
predictedTestOutputData.append(nextPoint)
lastAvailableData = nextPoint
predictedTestOutputData = np.array(predictedTestOutputData).reshape(
(nTesting, 1))
#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",
"Prediction of future values", "Time",
"Output")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries(
'Actual Output',
minMax.inverse_transform(testActualOutputData[:nTesting, 0]))
outplot.setYSeries(
'Predicted Output',
minMax.inverse_transform(predictedTestOutputData[:nTesting, 0]))
outplot.createOutput()
print("Done!")
from utility import Utility
import sys
import numpy as np
from scipy.stats import pearsonr, spearmanr
from plotting import OutputPlot as plot
from datetime import datetime
import os
from sklearn import preprocessing as pp
folderName = "Outputs/Outputs_" + str(datetime.now())
os.mkdir(folderName)
output = plot.OutputPlot(folderName + "/Correlation.html",
"Correlation of output with depth", "",
"Depth (hours)", "Correlation coefficient")
#Get the commamd line arguments
profileList = ["AcerDE"]
directoryName = "Datasets/"
for i in range(len(profileList)):
profileName = profileList[i]
datasetFileName = directoryName + profileName + "_time_interaction.csv"
util = Utility.SeriesUtility()
series = util.convertDatasetsToSeries(datasetFileName)
resampledSeries = util.resampleSeriesSum(series, "H")
yearsOfData = 3
recentCount = yearsOfData * 365 * 24
filteredSeries = util.filterRecent(resampledSeries, recentCount)
depth = 365 * 24 #1 year
#Predict for future
lastAvailablePoint = predictedTrainingOutputData[nTraining - 1, 0]
testingPredictedOutputData = []
for i in range(nTesting):
#Compose the query
query = [1.0]
query.append(lastAvailablePoint)
#Predict the next point
nextPoint = network.predict(np.array(query).reshape(1, 2))[0, 0]
testingPredictedOutputData.append(nextPoint)
lastAvailablePoint = nextPoint
testingPredictedOutputData = np.array(testingPredictedOutputData).reshape(
nTesting, 1)
#Plotting of the prediction output and error
outplot = outputPlot.OutputPlot("Outputs/Prediction.html",
"Darwin Sea Level Pressure Prediction",
"Prediction of future values", "Time",
"Sea Level Pressure")
outplot.setXSeries(np.arange(1, nTesting + 1))
outplot.setYSeries(
'Actual Output',
minMax.inverse_transform(testActualOutputData[:nTesting, 0]))
outplot.setYSeries(
'Predicted Output',
minMax.inverse_transform(testingPredictedOutputData[:nTesting, 0]))
outplot.createOutput()
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!")
