only r_b type can be classified

[liuke97]

I try to use tensorboard and find that room type image is completely black,only room boundary can be classified.I didn't modify the t code of train part.So anyone come to similar results like me ?I check the code and don't know what the problem is.Thank you .

[zcemycl]

@liuke97
It depends on your batch size, I trained it with bs=8, the room type will slowly appear. (rewrote part of the codes into python3 and tf2, view it in https://github.com/zcemycl/TF2DeepFloorplan)

[liuke97]

@liuke97
It depends on your batch size, I trained it with bs=8, the room type will slowly appear. (rewrote part of the codes into python3 and tf2, view it in https://github.com/zcemycl/TF2DeepFloorplan)

By using bs=8,only in first 5 steps I can see gray results of room predicts.After the whole training I test the model, still all black rooms.