PearlLeeStudio/TheArtist-MusicTransformer-ft-pop29

TheArtist Music Transformer — F4 (Pop 1K Mix, jazz-leaning)

Jazz-adapted chord model with a 1,000-sequence pop rehearsal buffer. The jazz-leaning endpoint of the mix-ratio sweep. Highest jazz top-1 in the collection (81.50%) at the cost of 1.22 pop points.

One of six checkpoints released alongside the paper Empirical Study of Pop and Jazz Mix Ratios for Genre-Adaptive Chord Generation (Lee, 2026). See the collection overview at PearlLeeStudio/TheArtist-MusicTransformer-pop-baseline.

Model summary

Field	Value
Architecture	Music Transformer with relative positional attention
Parameters	25,661,440
Vocabulary size	351 tokens
Max sequence length	256
d_model / heads / FFN / layers	512 / 8 / 2048 / 8
Fine-tune resumed from	Phase 0 pop baseline
Best epoch	6

Training data

All 1,513 jazz training sequences plus 1,000 pop rehearsal sequences (seed 42). Pop:jazz ≈ 0.66:1, that is, less pop than jazz in the mix.

Evaluation (held-out per-genre test sets)

Metric	Pop test	Jazz test
Top-1 accuracy	83.02%	81.50%
Top-5 accuracy	96.93%	92.59%
Perplexity	1.81	2.26
Δ vs. Phase 0 baseline	−1.22	+8.64

F4 is the jazz-leaning endpoint of the mix-ratio sweep. It produces the most jazz-flavoured continuations among the released checkpoints, with secondary dominants, tritone substitutions, modal interchange, and II-V chains across distant keys. The cost is roughly one point of pop top-1 accuracy. Qualitative samples (paper §6.4) on a minor ii-V prompt show the bebop-style harmonic motion that this checkpoint commits to more strongly than F3.

Intended use

Recommended for jazz-flavoured chord composition where the user is willing to trade some pop fluency for stronger jazz identity. F3 (ft-pop50) is the balanced alternative; F1 (ft-pop80) is the symmetric pop-leaning endpoint.

Out of scope: melody or audio generation; genres outside pop, rock, and jazz; real-time low-latency settings.

Usage

import torch
from huggingface_hub import hf_hub_download
from model import MusicTransformer
from tokenizer import ChordTokenizer

ckpt_path = hf_hub_download(
    repo_id="PearlLeeStudio/TheArtist-MusicTransformer-ft-pop29",
    filename="best.pt",
)
tokenizer = ChordTokenizer()
ckpt = torch.load(ckpt_path, map_location="cpu", weights_only=False)

model = MusicTransformer(
    vocab_size=tokenizer.vocab_size,
    d_model=512, n_heads=8, d_ff=2048, n_layers=8,
    max_seq_len=256, dropout=0.0, pad_id=tokenizer.pad_id,
)
model.load_state_dict(ckpt["model_state_dict"])
model.eval()

Training-data licenses

Dataset	License
Chordonomicon	Public (user-generated)
McGill Billboard	CC0
Jazz Harmony Treebank	Public
JazzStandards (iReal Pro)	Community redistribution
Weimar Jazz Database	ODbL
JAAH	Research-use public

Citation

Preprint: arXiv:2605.04998.

@misc{lee2026chordmix,
  title         = {Empirical Study of Pop and Jazz Mix Ratios for Genre-Adaptive Chord Generation},
  author        = {Lee, Jinju},
  year          = {2026},
  eprint        = {2605.04998},
  archivePrefix = {arXiv}
}