Nguyen Viet Hung comments

Repositories
Issues
Comments

Results 1 comments of


                                            Nguyen Viet Hung

Yolov11-n 模型int8量化误差过大

标题: ncnn2int8量化时出现“Floating point exception (core dumped)”求助内容: 大家好，我在使用ncnn对YOLO模型进行量化时，执行以下命令： ./ncnn2int8 yolo11n.ncnn-opt.param yolo11n.ncnn-opt.bin yolo11n-int8.param yolo11n-int8.bin yolo11n.ncnn.table 运行到某些卷积层时，出现报错： quantize_convolution conv_80 quantize_convolutiondepthwise convdw_174 ... Floating point exception (core dumped) 我已经检查过table文件和param/bin文件都是有效的。不清楚为什么会出现floating point错误。请问有谁对ncnn2int8量化有经验，能帮我分析原因并给出解决方法吗？...