Hi, I'm trying to train deeplabv3 with xception backbone on voc + SBD dataset. You provided the voc pretrained model deeplabv3_xception.h5. But If I want to reproduce your training result, I should not use it as pretrained model, right? So I comment out the line in train.py to not loading pretrained weights. But after 100 epochs, my model accuracy is poor, compared with your model. Did I miss something, do I need something like an ImageNet pretrained model or COCO pretrained model? Thanks!
Hi, I have problem exporting a .pb file with the current produced .h5 files. I do not think you have provided the export method in the project file. Could you please give me some advice on that? @bubbliiiing