lansfair comments

Results 5 comments of


                                            lansfair

微信号：18280382462 题目：优化 ncnn int8 模型部署

So now how do I convert F.conv2d to tensorrt?

Shapely is too slow; one should first use bounding boxes to find all intersecting targets, and then use Shapely

因为GPU版会自动导出torchscript，但whisper-large-v3并不支持这么做

修改cpp里的代码逻辑，去掉导出torchscript的部分，重写加载模型的逻辑，再重新编译