jinbo-hu/l3das22-task2

A Track-Wise Ensemble Event Independent Network for 3D Polyphonic Sound Event Localization and Detection

l3das22-task2's Issues

When the "Data augmentation" and "Track-wise ensemble model" parts can be open-source

Hello, thank you for your openness. As a first-year graduate student, it has been greatly helpful to me. However, I would like to ask when the "Data augmentation" and "Track-wise ensemble model" parts can be open-sourced. Thank you once again!

ValueError: need at least one array to concatenate

Hi,Jinbo-Hu
I tried to reproduce the code,but ouccr this error.And the logs is here.
Dataset l3das22 is being developed......

Extracting scalar......

0it [00:02, ?it/s]
Traceback (most recent call last):
File "code/main.py", line 77, in
sys.exit(main(args, cfg))
File "code/main.py", line 51, in main
preprocessor.extract_scalar()
File "D:\sound source localization\L3DAS22-TASK2-master\code\learning\preprocess.py", line 271, in extract_scalar
features_A = np.concatenate(features_A, axis=1)
File "<array_function internals>", line 6, in concatenate
ValueError: need at least one array to concatenate