niaoyu

Results 5 comments of niaoyu

YoloV3 error requiring network without implicit batch dimensions

Just install tensorrt oss will fix the problem

nvJPEG_encode failed with image on size 8500*5667 on cuda10.2

Hi, did you fix the problem?

how to add soft-nms?

hi how's going?

I can't achieve your mAP (77.x)

77.29 for mine

Conversion from onnx hurts accuracy when model is using fp16

> I was able to recover our transformer model from a similar situation by setting the reduce and unary layers to fp32: > > ```python > for layer in network:...