Instructions to use m-a-p/YuE2-Vae with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use m-a-p/YuE2-Vae with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="m-a-p/YuE2-Vae", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("m-a-p/YuE2-Vae", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add preprocessor_config.json for Transformers
#1
by bezzam HF Staff - opened
Hello I'm from the Transformers team at Hugging Face! We're working on adding native Yue2 support to Transformers π€
This PR adds the feature extractor config used by the native Transformers integration of the YuE2 VAE (https://github.com/huggingface/transformers/pull/50107), so that AutoFeatureExtractor.from_pretrained("m-a-p/YuE2-Vae") works.
It only zero-pads batches of 48 kHz stereo audio to the same length (EncodecFeatureExtractor). It is not read by modeling_vae.py, so it does not change the behavior of the existing remote code.
NOTE: merging can wait until the Transformers PR is done π€