rsen

If your still interested let me know and I can share some code

Do you want a webcam example or read video from file?

> Video from file would be great > […](#) > On Sat, Jun 6, 2020 at 4:36 PM robisen1 ***@***.***> wrote: Do you want a webcam example or read video...

Terrance, This file works. It ingests a video, reads it, performs detections, writes boxes on any detections, displays the frames (video) as it is processing, the writes to disk. The...

> 损失函数是有问题，也可能是权重初始化的参数有问题（那部分来源于yolov3），在加载预训练权重的前提下可以训练，这两部分都需要重写 Have you had a chance to fix the issues in train? thanks

I used Colabler but switched to VoTT which is, in general, easier to use and faster workflow.

> @robisen1 Do you able to convert yolo-custom model (which you got after training your dataset) to Keras model?. sorry, i did not reply. I have not sure your question...

> @Simon-liusheng 8G显存的话，只需要将batch_size = 2就能完成作者的训练吗？model size不要修改吧？比如将（608，608）改为（512， 512），或者（416，416）。 > 因为我看默认下载的yolov4.weights的model size是（608，608）。 > 谢谢 ohh i did not notice that. are we supposed to be using 608X608 in train? I sort of...

This usally means that data you are supplying is overwhelming the GPU. Can you tell us what GPU you have, watch batch settings you are using, if this happens during...

Try changing your unfrozen layers batch to just 4 and you are frozen to just 2 and let me know what happens. For some reason this implementation of Yolo 4...