leonardo-blanger / refinedet_tensorflow Goto Github PK

when I run train_refinedet_voc.py,The error is as follows:

File "D:\anac\envs\s190131016\lib\site-packages\tensorflow_core\python\ops\ragged\ragged_tensor.py", line 2027, in _from_compatible_tensor_list
"ragged_rank must be non-negative; got %s." % self._ragged_rank)

ValueError: ragged_rank must be non-negative; got 0.

tensorflow 2.0.0
python3.6

Thank you for your sharing Refinedet code.
https://github.com/Leonardo-Blanger/RefineDet_TensorFlow

I tried to find out the license, because you didn't indicate it in your source code.
Could you teach me if there is the best description among the following?

ex.1) I want simple and permissive. → MIT License
ex.2) I'm concerned about patents. → Apache License
ex.3) Others

If you don't know about that, the following URL will be help you.
https://choosealicense.com/

Thanks

I tried to train my own dataset, but met this error: 'Currently, you cannot build your model if it has positional or keyword arguments that are not inputs to the model, but are required for its call method. Instead, in order to instantiate and build your model, call your model on real tensor data with all expected call arguments'.
TF version: 1.12
Python: 3.5
I didn't changed anything, and could not find any solution from Google, please help me! Thanks!

is it right?

Thank you for your share!

Is there any implementation of tensorflow 1.x version?