RetroCirce/HTS-Audio-Transformer
The official code repo of "HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection"
PythonMIT
Issues
- 2
- 2
- 2
RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.
#61 opened by zhiziwy - 1
RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.
#59 opened by LiupengNew - 0
balanced_audioset pretrain
#60 opened by cxy0022 - 4
报错内容ValueError: The provided lr scheduler "<torch.optim.lr_scheduler.LambdaLR object at 0x7fe3d759bb50>" is invalid
#48 opened by kkkjjjj1 - 0
Where do I get the MD5 for the Audio Set
#58 opened by TFEI-Nagato - 1
SEDWrapper sed_model problem
#54 opened by Roon311 - 0
- 2
type of GPU
#50 opened by haloolahh - 0
- 2
FileNotFoundError: [Errno 2] No such file or directory: 'audio_32k/1-100032-A-0.wav'
#53 opened by Darcy0218 - 1
- 3
Audioset dataset for pretraining
#38 opened by youngwhite - 2
cannot pickle 'module' object
#52 opened by gillesmeyhi - 1
cyclic window shifting in the (256,256) tensor
#47 opened by tsw123tsw - 1
the size of the input spectrum
#49 opened by haloolahh - 1
- 0
Validation loss metric
#46 opened by OhadCohen97 - 2
Usage on Strongly labelled Dataset for SED
#45 opened by urvashi07 - 3
Key to checkpoints in drive
#44 opened by Sreyan88 - 5
Model Checkpoints
#42 opened by wonyangcho - 0
训练过程报错:段错误 (核心已转储)
#43 opened by ammerser - 3
reporduce training on esc-50 has an error
#41 opened by visionchan - 1
About shape of input wav
#40 opened by wangqian621 - 4
Getting started with a custom dataset
#34 opened by OhadCohen97 - 1
RuntimeError: Input and output sizes should be greater than 0, but got input (H: 0, W: 64) output (H: 1024, W: 64)
#39 opened by Mizuho32 - 3
- 1
- 0
audioset 训练中报错
#36 opened by fuguanyu - 1
upsample_bicubic2d_backward_out_cuda
#33 opened by DuQingChen - 2
How can run this project with one GPU?!
#32 opened by saeedmaroof - 2
How to finetune on strong label dataset?
#30 opened by wengstA - 1
- 2
- 1
How can i use my own dataset in this model:
#29 opened by yyssxxx - 2
How can I test model?
#26 opened by yyssxxx - 1
- 1
- 1
Questions about models.py
#24 opened by PrShi113 - 1
- 1
Unexpectedly high accuracy of 99 percent
#22 opened by the6thsense - 2
TypeError: cannot pickle 'module' object
#21 opened by JonathanFL - 1
- 3
Learning rate for small datasets
#18 opened by EmreOzkose - 1
预训练的.ckpt文件
#17 opened by dong-0412 - 1
Question about reshape log_mel to img size
#15 opened by lyc1993 - 1
- 6
Hello, I have the following problem when running the code, what is the reason?
#13 opened by ykingliu - 1
想利用这个网络训练一个用于sed的模型,训练数据都是强帧级的标注
#12 opened by sanli0