Xudong Wang

我是否可以只用LR图像进行Train

1

Hi, Thank you for your work and sharing code. Can the network model only input LR images for training?

why don`t have objCompose_cycleGAN_baseline and objCompose_baseline

3

elif opt.dataset_mode == 'baseline': from .objCompose_baseline import objComposeBaselineModel model = objComposeBaselineModel() elif opt.dataset_mode == 'baseline_unaligned': from .objCompose_cycleGAN_baseline import objComposeBaselineCycleModel model = objComposeBaselineCycleModel() else:

about classifiction.yaml

2

How is the vector value in classification.yaml obtained? I found that the vector value was not accurate

ERROR: Insufficient count of input/output paths supplied! Expected 1 output paths and 2 input paths! Got sys.argv: []

3

The defaults.yaml file path is set as follows path: local paths: # local: /home/anne/Anne/data/airway_vida_test/A2019RR46/ local: G:\Airway\A2069BH51\ The file structure is  Error content prompts ERROR: Insufficient count of input/output paths...

pth2onnx和onnx2trt是没有问题，但是trt推理出现box坐标是特别大的值，标签和得分是正确的

5

pth转换为onnx，用官方的代码转，并且测试是正确。但是onnx转换为trt，用的是tensorRT8.6.1,转换没有问题，但是推理出现box坐标是特别大的值，标签和得分是正确的。 D:\tool\TensorRT-8.6.1.6.Windows10.x86_64.cuda-11.8\TensorRT-8.6.1.6\bin\trtexec.exe --onnx=model.onnx --workspace=4096 --avgRuns=100 --shapes=images:1x3x640x640 --saveEngine=model.trt 我的C++推理代码如下 #include #include #include #include #include #include #include #include "NvInfer.h" #include "NvInferRuntimeCommon.h" #include #include using namespace nvinfer1; using namespace std; #define CUDA_CHECK(call) \...