def main():
n = 20200
trainN = 20000
seqLength = 500
data = np.expand_dims(StandardGenerator('long_term').generate(n), axis=1)
trainData, testData = Utility.trainTestSplit(data, trainN)
trainSequences = Utility.breakTrainSeq(trainData, None, seqLength)
forecastHorizon = 1
lag = 30
model = DeepNN(
forecastHorizon=forecastHorizon,
lag=lag,
numUnitsPerLayer=10,
numLayers=2,
numTargetVariables=1,
numExoVariables=0
)
loss = model.train(
trainSequences=trainSequences,
numIterations=20,
optimizer=tf.keras.optimizers.Adam(
learning_rate=tf.keras.optimizers.schedules.ExponentialDecay(
0.1,
25,
0.97
)
),
verboseLevel=2,
returnLosses=True
)
Plot.plotLoss(loss)
evalLoss, Ypred = model.evaluate(
testData,
returnPred=True
)
Ytrue = testData[lag + forecastHorizon:, :]
print(f'Eval Loss: {evalLoss}')
Plot.plotPredTrue(Ypred, Ytrue)
def evaluateAndPlot(loss, model, trainData, testData, numTrainSeqPlot,
trainPlotSeqLength, plotPrefix, plotDir):
plotPath = plotDir + f'/{plotPrefix}_loss.png'
Plot.plotLoss(loss, savePath=plotPath, saveOnly=True)
for i in range(numTrainSeqPlot):
idx = np.random.randint(0, trainData.shape[0] - trainPlotSeqLength)
seq = trainData[idx:idx + trainPlotSeqLength]
evalLoss, Ypred = model.evaluate(seq, returnPred=True)
Ytrue = seq[1:]
print(f'Train Eval Loss: {evalLoss}')
plotPath = plotDir + f'/{plotPrefix}_train{i}.png'
Plot.plotPredTrue(Ypred,
Ytrue,
'Train Data',
savePath=plotPath,
saveOnly=True)
testLoss, Ypred = model.evaluate(testData, returnPred=True)
Ytrue = testData[1:]
print(f'Test Eval Loss: {testLoss}')
plotPath = plotDir + f'/{plotPrefix}_test.png'
Plot.plotPredTrue(Ypred,
Ytrue,
'Test Data',
savePath=plotPath,
saveOnly=True)
lossPath = plotDir + f'/{plotPrefix}_loss'
fl = open(lossPath, 'w')
fl.write(f'Final Train Loss: {loss[-1]}\n' + f'Test Loss: {testLoss}')
fl.close()
def tryModelMultiSeq(trainSequences, testData, plotPrefix, plotDir):
model = LstmForecast(
forecastHorizon=1,
stateSize=10,
activation='tanh',
numRnnLayers=1
)
loss = model.train(
trainSequences=trainSequences,
numIterations=15,
optimizer=tf.keras.optimizers.Adam(
learning_rate=tf.keras.optimizers.schedules.ExponentialDecay(
0.08,
40,
0.98
)
)
)
plotPath = plotDir + f'/{plotPrefix}_loss.png'
Plot.plotLoss(loss, savePath=plotPath, saveOnly=True)
trainPlot = 5
for i, idx in enumerate(list(np.random.randint(0, len(trainSequences), size=(trainPlot,)))):
seq = trainSequences[idx]
evalLoss, Ypred = model.evaluate(seq, returnPred=True)
Ytrue = seq[1:]
print(f'Training Eval Loss: {evalLoss}')
plotPath = plotDir + f'/{plotPrefix}_train{i}.png'
Plot.plotPredTrue(
Ypred,
Ytrue,
'Train Data',
savePath=plotPath,
saveOnly=True
)
testLoss, Ypred = model.evaluate(testData, returnPred=True)
Ytrue = testData[1:]
print(f'Test Eval Loss: {testLoss}')
plotPath = plotDir + f'/{plotPrefix}_test.png'
Plot.plotPredTrue(Ypred, Ytrue, 'Test Data', savePath=plotPath, saveOnly=True)
lossPath = plotDir + f'/{plotPrefix}_loss'
fl = open(lossPath, 'w')
fl.write(
f'Final Train Loss: {loss[-1]}\n'
+ f'Test Loss: {testLoss}'
)
fl.close()
def main():
n = 21500
trainN = 21000
seqLength = 500
numSeqPlot = 5
trainData, testData = Utility.trainTestSplit(
StandardGenerator('long_term').generate(n),
trainN
)
trainSequences = Utility.breakSeq(trainData, seqLength=seqLength)
# for i in range(numSeqPlot):
# Plot.plotDataCols(trainSequences[
# np.random.randint(0, len(trainSequences))
# ])
model = LstmForecast(
forecastHorizon=1,
stateSize=50,
activation='tanh',
numRnnLayers=3
)
model.model.summary()
loss = model.train(
trainSequences=trainSequences,
numIterations=15,
optimizer=tf.keras.optimizers.Adam(
learning_rate=tf.keras.optimizers.schedules.ExponentialDecay(
0.01,
20,
0.96
)
)
)
Plot.plotLoss(loss)
for i in range(numSeqPlot):
idx = np.random.randint(0, len(trainSequences))
evalLoss, Ypred = model.evaluate(trainSequences[idx], returnPred=True)
Ytrue = trainSequences[idx][1:]
Plot.plotPredTrue(Ypred, Ytrue, 'On Train')
evalLoss, Ypred = model.evaluate(testData, returnPred=True)
Ytrue = testData[1:]
Plot.plotPredTrue(Ypred, Ytrue, 'On Test')
def main():
n = 21500
trainN = 21000
seqLength = 500
numSeqPlot = 5
trainData, testData = Utility.trainTestSplit(
StandardGenerator('long_term').generate(n), trainN)
model = ExtremeTime2(forecastHorizon=1,
memorySize=20,
windowSize=10,
embeddingSize=10,
contextSize=10)
loss = model.train(
targetSeries=trainData,
sequenceLength=seqLength,
numIterations=10,
optimizer=tf.keras.optimizers.Adam(
learning_rate=tf.keras.optimizers.schedules.ExponentialDecay(
0.01, 50, 0.99)),
verboseLevel=1,
returnLosses=True)
Plot.plotLoss(loss)
for i in range(numSeqPlot):
idx = np.random.randint(0, trainN - seqLength)
seq = trainData[idx:idx + seqLength]
evalLoss, Ypred = model.evaluate(seq, returnPred=True)
Ytrue = seq[1:]
print(f'Train Eval Loss: {evalLoss}')
Plot.plotPredTrue(Ypred, Ytrue, 'On Train')
evalLoss, Ypred = model.evaluate(testData, returnPred=True)
Ytrue = testData[1:]
print(f'Test Eval Loss: {evalLoss}')
Plot.plotPredTrue(Ypred, Ytrue, 'On Test')