Predicts the response for sample(s)

outputs = model.predictProb(inputs)
[outputs,outputProbs,f] = model.predictProb(inputs)
[...] = model.predictProb(inputs, 'OptionName',optionValue, ...)

Input

• inputs The input samples, floating-point matrix. One or more vectors stored as rows of the matrix.

Output

• outputs The predicted class for each sample.
• outputProbs Matrix that contains the output probabilities corresponding to each element of result. (A matrix of size nsamples-by-nclasses). The probabilities are not normalized to the 0..1 range.
• f unused and returns 0.

Options

• Flags The optional predict flags, model-dependent. For convenience, you can set the individual flag options below, instead of directly setting bits here. default 0
• RawOutput makes the method return the raw results (class index without mapping to class labels). default false

The method estimates the most probable classes for input vectors. Input vectors (one or more) are stored as rows of the matrix inputs. In case of multiple input vectors, there should be one output vector outputs. The matrix outputProbs contains the output probabilities corresponding to each element of result.

The function is parallelized with the TBB library.