search

WongKinYiu / ScaledYOLOv4

Scaled-YOLOv4: Scaling Cross Stage Partial Network
GNU General Public License v3.0
2.02k stars 575 forks source link

NaN output from the first layer when training with GPU #263

Open hoangphucITJP opened 3 years ago

hoangphucITJP commented 3 years ago

As the title. When I train on the CPU, it works fine. But training on GPU produces nan after the first block.

Please advise. Thanks.

kXborg commented 2 years ago

@hoangphucITJP, any leads on this one?