YOLO early stopping not running off epochs

19 次查看（过去 30 天）
Conner Carriere
Conner Carriere 2022-11-29
回答： Vidip Jain 2023-3-20
Does the ValidationPatience option in trainingOptions() go by epocs or iterations? I am trying to implement early stopping into my YOLO V4 learning, and it seems to be by iterations, and stopping at a selected number. For example, ValidationPatience = 800, it will stop at the 800th iteration, even though the validation loss is not increasing.
Any help on how to implement early stopping into YOLO
  1 个评论
Reece Pene
Reece Pene 2022-12-15
编辑：Reece Pene 2022-12-15
I have the same issue. Matlab isnt allowing the training to stop according to the validation patience. Instead it stops at a certain iteration.
Hope someone fixes this. It might be an error on the yolov4 training function.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Vidip Jain
Vidip Jain 2023-3-20
The “ValidationPatience” option in “tainingOptions()” goes by epochs, not iterations. The patience value determines the number of epochs to wait before stopping training when the validation loss has stopped improving. If the validation loss does not improve for the specified number of epochs, the training stops early.
In your case, it sounds like you are using a custom implementation of early stopping based on the number of iterations rather than epochs. You may want to modify your code to use a patience value based on epochs instead, which will give you more consistent results across different training configurations. You can calculate the number of iterations for a given epoch by multiplying the number of iterations per epoch by the epoch number.

类别

在 Help Center 和 File Exchange 中查找有关 Recognition, Object Detection, and Semantic Segmentation 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by