#Compute ideal penalties and error on training data
meanIdealPenalities += learner.parallelPenaltyGrid(trainX, trainY, testX, testY, paramDict)
for i in range(len(paramDict["setGamma"])):
allError = 0
learner.setGamma(paramDict["setGamma"][i])
for trainInds, testInds in idx:
validX = trainX[trainInds, :]
validY = trainY[trainInds]
learner.learnModel(validX, validY)
predY = learner.predict(trainX)
allError += learner.getMetricMethod()(predY, trainY)
meanAllErrors[i] += allError/float(len(idx))
#Compute true error grid
methodInd = 4
cvGrid = learner.parallelSplitGrid(trainX, trainY, testX, testY, paramDict)
meanCvGrid[methodInd, :] += cvGrid
bestLearner.setGamma(paramDict["setGamma"][numpy.argmin(cvGrid)])
bestLearner.learnModel(trainX, trainY)
predY = bestLearner.predict(testX)
meanErrors[methodInd] += bestLearner.getMetricMethod()(testY, predY)
meanDepths[methodInd] += bestLearner.tree.depth()
meanSizes[methodInd] += bestLearner.tree.getNumVertices()
#Compute true error grid using only training data
methodInd = 5
cvGrid = learner.parallelSplitGrid(trainX, trainY, trainX, trainY, paramDict)
meanCvGrid[methodInd, :] += cvGrid
bestLearner.setGamma(paramDict["setGamma"][numpy.argmin(cvGrid)])
bestLearner.learnModel(trainX, trainY)
predY = bestLearner.predict(testX)
