timegoby
timegoby
你好,我看了流程,但是说的很笼统 比如,IP的预分段和各分段的同步升级有没有具体的参考文档呢? 感谢!
https://modelscope.cn/models/iic/cv_cspnet_image-object-detection_yolox/summary 使用他们的推理代码报错
数据集中有部分图片后缀名是.JPG,需要手动修改为.jpg
I have refered the code from [https://github.com/NVIDIA-AI-IOT/torch2trt/issues/612](url), and it works for me. @tensorrt_converter('torch.nn.functional.pixel_shuffle') def convert_PixelShuffle(ctx): input = ctx.method_args[0] scale_factor = ctx.method_args[1] input_trt = add_missing_trt_tensors(ctx.network, [input])[0] output = ctx.method_return batch_size, in_channels,...