PearlLeeStudio/TheArtist-MusicTransformer-pop-baseline

TheArtist Music Transformer — Phase 0 (Pop Baseline)

Pop-pretrained chord generation model. The starting point that all five jazz fine-tunes resume from. Released as the no-jazz reference for measuring catastrophic forgetting in the companion paper.

This checkpoint is one of six released alongside the paper Empirical Study of Pop and Jazz Mix Ratios for Genre-Adaptive Chord Generation (Lee, 2026). The collection landing page is at PearlLeeStudio/TheArtist-MusicTransformer-pop-baseline (this repo) and at the per-checkpoint repositories listed in the paper. See the paper for the experimental design that motivates this set of checkpoints.

Demo

Watch TheArtist in action on YouTube — interactive staff editor, MIDI input, AI generation with live progress, and per-genre LoRA playback across the 13-genre vocabulary.

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
Training framework	PyTorch 2.5+ (CUDA 12.1)

Training data

Trained from scratch on the pop training split: Chordonomicon (679K songs) and McGill Billboard (890 songs), deduplicated and twelve-key-augmented. Three epochs at peak learning rate 3 × 10⁻⁴ with one-epoch warmup and cosine decay. Wall-clock time ≈27 hours on a single NVIDIA GeForce RTX 4070 Laptop GPU.

Evaluation (held-out per-genre test sets)

Metric	Pop test	Jazz test
Top-1 accuracy	84.24%	72.86%
Top-5 accuracy	97.10%	86.51%
Perplexity	1.73	4.01

The 72.86% jazz top-1 from a pop-only model reflects the substantial token overlap between the two genres. Pop and jazz share most of their chord tokens; jazz-specific gains in our fine-tuned checkpoints come from learning the transition statistics over those tokens, not from learning new tokens.

Intended use and limitations

This checkpoint is the unmodified pop baseline. It is the most pop-fluent model in the collection and the most jazz-naive. Use it when pop output is the only target. For balanced pop and jazz output, use F3 (ft-pop50). For pop-leaning output that still includes occasional jazz coloration, use F1 (ft-pop80). For jazz-leaning output, use F4 (ft-pop29).

Out of scope: melody or audio generation (symbolic chord-only model); genres outside pop, rock, and jazz (out-of-distribution behavior not characterized); real-time low-latency settings (use batched inference).

Usage

The repo bundles the project's model.py and tokenizer.py at the repo root, so external users can load the checkpoint end-to-end without cloning anything from GitHub. snapshot_download materializes the full repo on disk; sys.path makes the bundled model.py / tokenizer.py importable.

Required dependencies: torch, huggingface_hub.

import sys
import torch
from huggingface_hub import snapshot_download

# Download the full repo (model.py, tokenizer.py, best.pt, config.json).
ckpt_dir = snapshot_download(repo_id="PearlLeeStudio/TheArtist-MusicTransformer-pop-baseline")
sys.path.insert(0, ckpt_dir)  # so the next two imports resolve

from model import MusicTransformer
from tokenizer import ChordTokenizer

tokenizer = ChordTokenizer()
ckpt = torch.load(f"{ckpt_dir}/best.pt", 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()

# Prompt = ii-V-I in C major; ask for a pop-flavoured continuation.
song = {
    "key": "Cmaj", "time_signature": "4/4", "genre": "pop",
    "bars": [["Dm7", "G7"], ["Cmaj7"]],
}
prompt_ids = tokenizer.encode_sequence(song)[:-1]
ids = torch.tensor([prompt_ids])
with torch.no_grad():
    for _ in range(32):
        logits = model(ids)
        next_id = torch.multinomial(
            torch.softmax(logits[:, -1, :] / 0.8, dim=-1), 1,
        )
        ids = torch.cat([ids, next_id], dim=-1)
        if next_id.item() == tokenizer.eos_id:
            break
print(tokenizer.decode(ids[0].tolist()))

For per-genre adaptation beyond pop and jazz, see the 11 LoRA adapter repos at PearlLeeStudio — they chain on top of this base.

Per-genre real-song eval (held-out 130-song set, 2026-05)

First per-genre evaluation of pop-baseline beyond the pop/jazz split that the original paper reports.

Eval results

Genre	n_songs	Top-1 (%)	Top-5 (%)	val_loss
pop	10	86.68	96.01	0.5734
rock	10	86.69	97.48	0.4578
jazz	10	64.96	81.16	1.8958
blues	10	81.52	93.91	0.8410
bossa	10	81.43	95.47	0.7825
classical	10	49.55	81.17	2.2389
country	10	85.89	98.44	0.5152
electronic	10	87.39	98.45	0.5072
folk	10	85.04	98.92	0.5244
funk	10	83.85	96.03	0.6811
gospel	10	79.79	96.85	0.7367
hip_hop	10	90.66	98.59	0.3957
rnb_soul	10	85.11	97.07	0.5877

On this eval set Phase0 peaks on hip_hop (90.66%) and struggles most on classical (49.55%). This is auxiliary signal — the 11 per-genre LoRAs (sister lora-* repos) are the recommended path for production use on the 9 non-pop, non-jazz genres. F-series cells on those genres show what the base model produces under [GENRE:none] conditioning (the model's [GENRE:X] token does not exist for the 9 new genres in the F-series vocab=351).

Eval dataset composition

130 songs total, 10 per genre × 13 genres. Drawn from the same splits/val.jsonl + splits/test.jsonl partitions every F-series model was held out from during training — no train-set leakage. Built by ai/training/build_eval_real_songs.py --seed 42 --per-genre 10 (deterministic).

Genre	n	Source(s)	Bar range	Avg duration · named
pop	10	billboard	58–116	189s · 10/10 named
rock	10	chordonomicon_rock	52–87	127s · 0/10 named
jazz	10	choco:jazz-corpus, choco:real-book, jazzstandards, jht	16–89	72s · 10/10 named
blues	10	chordonomicon_blues	24–46	93s · 0/10 named
bossa	10	chordonomicon_bossa	24–78	88s · 0/10 named
classical	10	chordonomicon_classical	11–40	60s · 10/10 named
country	10	chordonomicon_country	30–81	110s · 0/10 named
electronic	10	chordonomicon_electronic	25–84	89s · 0/10 named
folk	10	chordonomicon_folk	33–82	114s · 0/10 named
funk	10	chordonomicon_funk	30–60	92s · 0/10 named
gospel	10	chordonomicon_gospel	24–85	98s · 0/10 named
hip_hop	10	chordonomicon_hip_hop	24–81	136s · 0/10 named
rnb_soul	10	chordonomicon_rnb_soul	34–82	128s · 0/10 named

Source license summary: McGill Billboard (CC0, named pop songs), Jazz Harmony Treebank / JazzStandards / WJazzD (Public / community-redistributed, named jazz standards), Bach chorales via music21 (public domain, named pieces), Chordonomicon per-genre subsets (CC BY-NC 4.0; titles are Spotify track IDs by upstream dataset policy — progressions are real songs). See docs/EVAL.md for full breakdown.

Methodology

Teacher-forced next-token cross-entropy / top-1 / top-5 over each song's token sequence (BOS + key + time_sig + genre + bars + EOS, truncated to max_seq_len=256). Same evaluate() call as ai/results/f1_per_genre_baseline.csv, just narrowed to the curated 130-song subset. Token-level metrics; not a generation-quality eval (free-generation comparison with R1 Sethares + R2 theory RAG rerank is documented separately in ai/results/eval_report.md).

Caveats:

• classical val partition is intrinsically small (37 sequences in full eval); the 10-song subset here has even narrower confidence bands. Directional finding (LoRA helps a lot on Bach harmony) is robust, exact pp deltas are noisy.
• F-series numbers on the 9 LoRA-only genres are conditioned without genre tag (vocab=351 has no [GENRE:country] token etc.). This is the realistic "F-series alone" condition, not a controlled ablation.

Source CSV: ai/results/real_song_eval.csv (17 models × 130 songs, long format).

Training-data licenses

Dataset	License
Chordonomicon	Public (user-generated)
McGill Billboard	CC0

Citation

Cite the original mix-ratio paper. The companion per-genre LoRA paper (chord-symbol time-series adaptation) is in preparation; its arXiv ID will be added here once posted.

@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}
}

@misc{lee2026chordtimeseries,
  title         = {How Far Can Chord-Symbol Time-Series Adaptation Carry Genre Identity?},
  author        = {Lee, Jinju},
  year          = {2026},
  note          = {arXiv preprint, ID TBD},
}