Question about the train and results

[WooKimm]

Hello, I attempted to replicate your work, but the accuracy I obtained was significantly lower than the model accuracy that you provided. My mAP was only around 23. I am wondering if the config file that I used differs from the one that you actually used, as I noticed that the code defaults to using the gt input of bbox for the gen_net, rather than the output of det_net as reported in the paper.

Additionally, while debugging the program, I encountered instances where the batch size of some of the input queries was 0, https://github.com/Mrmoore98/VectorMapNet_code/blob/47ed228de3b1441fb5ae31a573e87affcb01129e/plugin/models/heads/polylinegenerator.py#L167

which caused the program to crash. I have currently implemented a try statement to filter out this issue. Have you encountered this problem before?

[Mrmoore98]

I guess you didn't totally use the code we provided. We have updated the config two month ago, which can reproduce the results we listed in the readme. And we have code to prevent the situation of input with size zero. https://github.com/Mrmoore98/VectorMapNet_code/blob/47ed228de3b1441fb5ae31a573e87affcb01129e/plugin/models/mapers/vectormapnet.py#L102