Wang Qinyu

**This program will throw the following error** ``` File "encode.py", line 39, in encode x, y = range( height / 2 ), range(width) TypeError: 'float' object cannot be interpreted as...

ValueError: cannot convert float NaN to integer

4

``` Traceback (most recent call last): File "tools/train_net.py", line 243, in args=(args,), File "/home/wangqinyu/detectron2/detectron2/engine/launch.py", line 59, in launch daemon=False, File "/home/wangqinyu/anaconda3/envs/Bezier/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 200, in spawn return start_processes(fn, args, nprocs, join,...

I can't find the detect head.

1

Why I can't find the detect head,are you fix it to backbone?

When I finetune in totalText(other dasets all like this), it throw "RuntimeError: cuda runtime error (101)"... Have you ever meet this? Noted that the config file is accurate，and test program...

It's a amazing work. Could you tell me how to train it on my own datasets?

Why the predicted mask image is not binarized?

1

When I tested the DAGM dataset after training the model，I found that the predicted mask image of visualized image is **NOT BINARIZED** as in the example,as shown in the figure...