trainNetwork: loss output vs. manual calculation

2 次查看（过去 30 天）
Roland Kruse
Roland Kruse 2021-1-18
编辑： Roland Kruse 2021-1-18
Dear Matlab community,
I have recently become a bit puzzled when it comes to the trainNetwork function, specifically the diagnostis printed.
I get the following values during the last epoch:
|======================================================================================================================|
| Epoch | Iteration | Time Elapsed | Mini-batch | Validation | Mini-batch | Validation | Base Learning |
| | | (hh:mm:ss) | RMSE | RMSE | Loss | Loss | Rate |
|======================================================================================================================|
| 1000 | 252750 | 12:58:24 | 0.04 | 0.17 | 0.0007 | 0.0136 | 6.2500e-05 |
| 1000 | 252800 | 12:58:34 | 0.03 | 0.16 | 0.0006 | 0.0120 | 6.2500e-05 |
| 1000 | 252850 | 12:58:43 | 0.04 | 0.16 | 0.0007 | 0.0133 | 6.2500e-05 |
| 1000 | 252900 | 12:58:52 | 0.03 | 0.17 | 0.0004 | 0.0143 | 6.2500e-05 |
| 1000 | 252950 | 12:59:01 | 0.03 | 0.16 | 0.0005 | 0.0121 | 6.2500e-05 |
| 1000 | 253000 | 12:59:11 | 0.03 | 0.17 | 0.0004 | 0.0150 | 6.2500e-05 |
One observes that the training loss is much lower than the validation loss, sign of overtraining but not the issue here.
If I now use the trained network, predict the responses and calculate the loss manually I receive:
training: 0.137 / validation: 0.149
This is systematic and leads me to wonder if the "Mini-batch Loss" is not the MSE of the training data.

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Custom Training Loops 的更多信息

标签

产品


版本

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by