DzungDo

icon indicating a website link https://www.linkedin.com/in/do-huu-dung-4a7167193/ icon indicating an email [email protected]

icon company Vision-in Global R&D Center icon location Seoul, South Korea icon location Blockchain | Computer Vision | Machine Learning.

Results 5 comments of DzungDo

Training so, so slow

same as my problem! I trained on RTX 2070, with the dataset of 85K images and 17 classes, It takes ~5 hours for one epoch! Is there any solution to...

Training so, so slow

> Try to set `paste_in: 0.00` in `hyp.scratch.yaml` file Or repace `compute_loss_ota` to `compute_loss` in this line https://github.com/WongKinYiu/yolov7/blob/main/train.py#L362 > > Does it help? I tried it since last week and...

I cann't compile compile the DCNv2

> I rolled back to torch 1.4.0 and torchvision 0.5.0 and solved the issue Thanks, it solved my problem!

YOLOv3-tiny、YOLOv4可以转，YOLOv4-tiny转不了

hi, did you guys solve this problem? thanks

YOLOv3-tiny、YOLOv4可以转，YOLOv4-tiny转不了

> > hi, did you guys solve this problem? thanks > > yolo4有mish应该转不了caffe吧，v4tiny可以转 Chen has provided the 'mish' layer for caffe, just remake the caffe with new source and convert...