Table 4.
Scenario | Training time (min) | Prediction time (min) |
---|---|---|
optA | 10 849 | 5283 (630) |
optB | 736 | 91 (95) |
optC | 351 | (71)a |
In brackets, we show the computational time that was needed to complete predictions on the same test sequences with the expert-trained parameters.
aTraining and prediction on the two different data sets were performed by one autoAug.pl run of the Training Web Service, i.e. prediction time is included in training time.