/// <summary>
///
/// </summary>
/// <param name="inputs"></param>
/// <param name="outputs"></param>
/// <returns></returns>
public double Train(float[][] inputs, float[][] outputs)
{
//ensure that data is destroyed after use
using (Value inputsValue = Value.CreateBatch(inputVariable.Shape, NP.ToOneDimensional(inputs), device))
using (Value outputsValue = Value.CreateBatch(outputVariable.Shape, NP.ToOneDimensional(outputs), device))
{
traindEpochs++;
var miniBatch = new Dictionary <Variable, Value>()
{
{ inputVariable, inputsValue }, { outputVariable, outputsValue }
};
trainer.TrainMinibatch(miniBatch, false, device);
return(trainer.PreviousMinibatchEvaluationAverage());
}
}
public float[][] Predicts(float[][] inputs)
{
using (Value inputsValue = Value.CreateBatch(inputVariable.Shape, NP.ToOneDimensional(inputs), device))
{
var inputDict = new Dictionary <Variable, Value>()
{
{ inputVariable, inputsValue }
};
var outputDict = new Dictionary <Variable, Value>()
{
{ classifierOutput.Output, null }
};
classifierOutput.Evaluate(inputDict, outputDict, device);
var prdict = outputDict[classifierOutput.Output].GetDenseData <float>(classifierOutput.Output);
float[][] outputs = new float[inputs.Length][];
for (int i = 0; i < inputs.Length; i++)
{
outputs[i] = prdict[i].ToArray();
}
return(outputs);
}
}
