Fang Jie

Results 1 issues of Fang Jie

Thanks for your work! But I'm confuse why the int8 model performance worse on jetson tx2. The inference time of fp32 416 model is about 250ms and the inference time...