Why not all properties in ClassifierMeasurements are available?

Question

I'm doing an experiment of objects classification on dataset CIFAR-10 and I'm doing this using a neural network. The basic code of this experiment is come from the demo on Mathematica website and I tried to add a part of evaluating the classifier I trained.

I used CNN as a classifier and this is what it look like:

    lenet = NetChain[ {ConvolutionLayer[20, 5], Ramp, PoolingLayer[2, 2],        ConvolutionLayer[50, 5], Ramp, PoolingLayer[2, 2], FlattenLayer[], 500, Ramp, 2, SoftmaxLayer[]}, "Output" -> NetDecoder[{"Class", classes}], "Input" -> NetEncoder[{"Image", {32, 32}}] ]

After training, I created a ClassifierMeasurementsObject using function ClassifierMeasurements:

    cm=ClassifierMeasurements[trained,testData]

where trained is the name of trained CNN module.

Then I use this code to see what I can do with the ClassifierMeasurementsObject:

    cm["Properties"]

and I got the output:

    {Accuracy, AccuracyRejectionPlot, ClassRejectionRate,   ConfusionFunction, ConfusionMatrix, ConfusionMatrixPlot, Error, FScore, InversePerplexity, Likelihood, LogLikelihood, Perplexity, Precision, Properties, Recall, RejectionRate}

what I got from the output is much less than the document for ClassifierMeasurements, something missed such as "AreaUnderROCCurve", "CorrectlyClassifiedExamples", "Examples", {"Examples",i->j}, "WorstClassifiedExamples", etc.

I don't know why would this happen. Is this because the algorithm I use or the dataset I use? Do I have any way to find and split out all the misclassified samples?

Thank you very much!