def plotHistogram(arrayValues, bins, ranges, fileName, title, subTitle, xAxisText, yAxisText, seriesName):
hist, edges = np.histogram(a=arrayValues, bins=bins, range=ranges)
histPlot = columnPlot.ErrorPlot(fileName, title, subTitle, xAxisText, yAxisText)
# Form the range of the bins
xAxisData = []
prev = edges[0]
for i in range(1, edges.shape[0]):
current = edges[i]
xAxis = str(prev) + "-" + str(current)
xAxisData.append(xAxis)
prev = current
xAxisData = np.array(xAxisData)
histPlot.setXAxis(xAxisData)
histPlot.setYAxis(seriesName, hist)
histPlot.createOutput()
# Get the number of minibatches
n_minibatches = data_sets.train.num_minibatches(batch_size)
#
# Plotting Setup
#
# Allow interactvie plotting
plt.allow_plot()
# Since we can't plot all the images in a batch, we pick some indices
# that we use throughout the training
train_recon_indices = choice(range(batch_size), 5, replace=False)
valid_recon_indices = choice(range(data_sets.validation.num_examples), 5, replace=False)
recept_indices = choice(range(n_hid), 20, replace=False)
# Initialize the plots
error_plot = ErrorPlot(1)
recon_valid_plot = ReconstructionPlot(2)
recept_fields_plot = ReceptiveFieldsPlot(3, width=5, height=4)
activation_plot = ActivationPlot(4)
#
# Training
#
# K Step Contrasting Divergence.
for i_epoch in range(n_epochs):
train_err = 0.0
# Step 4
error = errorFunction.compute(actualData.reshape(1, horizon),
deScaled.reshape(1, horizon))
regressionError.append(error)
xAxisError.append("Depth_" + str(depth))
# Plotting of the actual and prediction output
outputFolderName = "Outputs/Output" + str(
datetime.now()) + "_depth_comparison_" + "_horizon_" + str(horizon)
os.mkdir(outputFolderName)
outplot = outTimePlot.OutputTimeSeriesPlot(
outputFolderName + "/Prediction.html", "Comparison of prediction outputs",
"with varying depths", "Likes Count")
outplot.setSeries('Actual Output', np.array(xAxis), actualData)
for depth in depthList:
seriesName = 'Predicted_Output_depth' + str(depth)
predicted = predictedDict[str(depth)]
outplot.setSeries(seriesName, np.array(xAxis), predicted)
outplot.createOutput()
#Plotting of regression
errPlot = errorPlot.ErrorPlot(outputFolderName + "/Regression_Error.html",
"Comparison of regression error",
"with varying depths", "ESN Configuration",
"Total Error")
#X-axis
errPlot.setXAxis(np.array(xAxisError))
#Series data
errPlot.setYAxis('Total Regression Error', np.array(regressionError))
errPlot.createOutput()
# Get the number of minibatches
n_minibatches = data_sets.train.num_minibatches(batch_size)
#
# Plotting Setup
#
# Allow interactvie plotting
plt.allow_plot()
# Since we can't plot all the images in a batch, we pick some indices
# that we use throughout the training
train_recon_indices = choice(range(batch_size), 5, replace=False)
valid_recon_indices = choice(range(data_sets.validation.num_examples), 5, replace=False)
recept_indices = choice(range(n_hid), 20, replace=False)
# Initialize the plots
error_plot = ErrorPlot(1)
recon_valid_plot = ReconstructionPlot(2)
recept_fields_plot = ReceptiveFieldsPlot(3, width=5, height=4)
activation_plot = ActivationPlot(4)
#
# Training
#
# K Step Contrasting Divergence.
for i_epoch in range(n_epochs):
train_err = 0.0
# Step 4
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']))
errPlot.setYAxis('RMSE', np.array([standardError, erdosError]))
errPlot.createOutput()
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")
errPlot.setXAxis(np.array(topologyNames))
errPlot.setYAxis('RMSE', np.array(topologyError))
errPlot.createOutput()
outplot = outputPlot.OutputPlot(
outputFolderName + "/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_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",
"Comparison of different graph topologies",
"with parameters tuner", "ESN Configuration",
"Total Error")
errPlot.setXAxis(
np.array(
['Standard', 'Erdos Renyi', 'Small World Graph',
'Scale Free Network']))
errPlot.setYAxis(
'RMSE',
np.array([standardError, erdosError, smallWorldError, scaleFreeError]))
errPlot.createOutput()