Anticipatory Music Transformer
Paper • 2306.08620 • Published • 10
How to use stanford-crfm/music-small-ar-inter-100k with Transformers:
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("stanford-crfm/music-small-ar-inter-100k", dtype="auto")Configuration Parsing Warning:In config.json: "architectures" must be an array
This is a Small (112M parameter) Transformer trained for 100k steps on interarrival-time encoded music from the Lakh MIDI dataset.
The Anticipatory Music Transformer paper is available on ArXiv.
The full model card is available here.
Code for using this model is available on GitHub.
See the accompanying blog post for additional discussion of this model.