Research Article
Comparing the Selected Transfer Functions and Local Optimization Methods for Neural Network Flood Runoff Forecast
Table 4
The selected quantiles of empirical distribution functions on runoff and rainfall depths for training, testing, and validation data sets.
| | Minimum | 1st Quartile | Median | Mean | 2nd Quartile | Maximum | St. dev. |
| Runoff depth [mm·hour−1] | | | | | | | | Training runoff | 0.000 | 0.0154 | 0.0194 | 0.1031 | 0.0592 | 3.828 | 0.326 | Testing runoff | 0.00380 | 0.00760 | 0.0231 | 0.09395 | 0.0622 | 4.107 | 0.249 | Validation runoff | 0.000 | 0.0040 | 0.0270 | 0.1008 | 0.0591 | 3.250 | 0.307 |
| Rainfall depth [mm·hour−1] | | | | | | | | Training rainfall | 0.2 | 0.4 | 0.8 | 1.365 | 1.700 | 12.400 | 1.658 | Testing rainfall | 0.2 | 0.2 | 0.6 | 1.439 | 1.400 | 73.9 | 4.485 | Validation rainfall | 0.2 | 0.2 | 0.8 | 1.575 | 2.200 | 15.800 | 1.900 |
|
|