search

LAION-AI / CLAP

Contrastive Language-Audio Pretraining
https://arxiv.org/abs/2211.06687
Creative Commons Zero v1.0 Universal
1.43k stars 137 forks source link

A bug for obtaining CLAP Embedding #147

Closed LittleFlyingSheep closed 6 months ago

LittleFlyingSheep commented 7 months ago

Hi~I find that in the 49 line of data/infer_clap.py, the parameter of the CLAP_module() has a wrong name.

The original verision:

model = CLAP_Module(enable_fusion=args.enable_fusion, aencoder=args.audio_encoder)

In this version, the parameter "aencoder" should be "amodel" according to the library of "laion_clap". Thus, the revised version should be:

model = CLAP_Module(enable_fusion=args.enable_fusion, amodel=args.audio_encoder)

I hope this suggestion can make this project better.

RetroCirce commented 6 months ago

Thank you for your finding! It is a typo for this version, we will correct it.