Aya Nasser Salama comments

Results 8 comments of


                                            Aya Nasser Salama

Change number of classes in the model output layer

Hi @zhaoyue-zephyrus , thank u, but Does this modification guarantee that all layers above the classifier layers are frozen?

Are all the models here working on single object tracking not MOT?

Thank you @MasterBin-IIAU ^^ I have a question regarding the model demo I made, Why the tracking box is not vanishing when the tracked object is become outside the frame?...

I get the following error when trying to load the model for inferencing on a custom video - I have tested on all the three models provided - Is this affecting the accuracy ?

@Sidharth-Gulati @Leo63963 How could you make inference with a custom video ?

Inference on a custom video

I have the same inquiry

single image test

@617778859 Could you please upload the code-test.py file that you're talking about ?

Convert FairMot to onnx

I could convert it as I convert any other pytorch model,but I dnt know should I set the conversion command after or before the tracker update code line

make.sh error

I am working with gcc 8.4.0 as mentioned in the documentation but the make give me this error fatal error: cublas_v2.h: No such file or directory #include compilation terminated. error:...

I find a bug in this code

The error will be solved by adding this function: def randomize_parameters(self): pass At this class: CenterCrop