private DataSource PredictionSource(int step)
{
double[] prediction = trainer.GetPrediction(train_data, step);
DataSource prediction_source = new DataSource();
prediction_source.Length = data.Length;
float acc = 0;
for (int i = 0; i < data.Length; i++)
{
if (i < network.InputCount() + step - 1)
{
// prediction_source.Samples[i].x = i;
// prediction_source.Samples[i].y = 0.5f;
prediction_source.Samples[i].x = 0;
prediction_source.Samples[i].y = 0;
}
else if (i < network.InputCount() + prediction.Length + step - 1)
{
prediction_source.Samples[i].x = i;
prediction_source.Samples[i].y = (float)prediction[i - network.InputCount() - step + 1];
// acc += (float) normalizer.Denormalize(prediction[i - network.InputCount()]);
// prediction_source.Samples[i].y = acc;
}
else
{
prediction_source.Samples[i].x = 0;
prediction_source.Samples[i].y = 0;
}
}
return(prediction_source);
}