About the cuda memory

[medicineqian]

As shown in the figure below, I have a memory explosion problem when running the code, there is no problem with the model training, but when running to val_epoch there will be memory explosion, that is, when running to the line of detect_object() in the inference.py there will be not enough memory.

[socome]

@medicineqian

I'm not sure. It seems like the model isn't trained enough for evaluation. (That is, the trained model predicts too many bounding boxes, leading to a memory explosion) To solve the problem, try to evaluate the model after 5 epochs.

if epoch > 5 : 
     val_epoch()