SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.00751 · 2020

Evaluation of CNN-based Automatic Music Tagging Models

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 2 failures of our instrument; 4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2006.00751.json · badge: /api/ran/2006.00751.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
minzwon/sota-music-tagging-models canonical (classifier label)
classifier basis: census
4 ran · 3 no recorded run
HephaestusProject/pytorch-FCN reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 1 no recorded run
FunctionStatusWhere it lives
gelu Ran
ran, honoured the contract we drafted
minzwon/sota-music-tagging-models/training/attention_modules.py
pointer only (licence: MIT); the stored copy get_code serves, from tanmay1618/text_to_sql_bert, is pointer only (licence: NONE) · get_code("fdc64f4c72036ae4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from tanmay1618/text_to_sql_bert, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_config Ran
ran; our contract draft was wrong, not the code
HephaestusProject/pytorch-FCN/evaluate.py
code served (permissive licence) · get_code("3315c75383bc102e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned DictConfig. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os
from pathlib import Path
from argparse import Namespace
from omegaconf import OmegaConf

# Create temporary directory structure with config files
temp_dir = tempfile.mkdtemp()
conf_dir = Path(temp_dir) / 'conf'
conf_dir.mkdir(exist_ok=True)

dataset_name = 'test_dataset'
dataset_dir = conf_dir / dataset_name
dataset_dir.mkdir(exist_ok=True)

model_dir = dataset_dir / 'model'
model_dir.mkdir(exist_ok=True)

pipeline_dir = dataset_dir / 'pipeline'
pipeline_dir.mkdir(exist_ok=True)

runner_dir = dataset_dir / 'runner'
runner_dir.mkdir(exist_ok=True)

# Create sample YAML config files
model_config = OmegaConf.create({'model_name': 'test_model', 'hidden_dim': 128})
OmegaConf.save(model_config, model_dir / 'test_model.yaml')

pipeline_config = OmegaConf.create({'pipeline_name': 'test_pipeline', 'batch_size': 32})
OmegaConf.save(pipeline_config, pipeline_dir / 'test_pipeline.yaml')

runner_config = OmegaConf.create({'runner_name': 'test_runner', 'num_epochs': 10})
OmegaConf.save(runner_config, runner_dir / 'test_runner.yaml')

# Change to the temp directory so the relative path 'conf' works
original_cwd = os.getcwd()
os.chdir(temp_dir)

# Create the args Namespace with required attributes
args = Namespace(
    dataset=dataset_name,
    model='test_model',
    pipeline='test_pipeline',
    runner='test_runner'
)

ARGS = {'args': args}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
omegaconf — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DictConfig — a DictConfig output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DictConfig output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
hz_to_midi Ran
ran; our contract draft was wrong, not the code
minzwon/sota-music-tagging-models/training/modules.py
code served (permissive licence) · get_code("2ee009ed5eb70422")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (4,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch

torch.manual_seed(0)

# hz_to_midi takes a single parameter 'hz' which should be a tensor
# The function applies torch.log2 to it, so it needs to support that operation
# Let's create a small tensor of frequency values in Hz

hz = torch.tensor([440.0, 880.0, 220.0, 110.0], dtype=torch.float32)

ARGS = {
    "hz": hz
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (4,) — shape (4,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
midi_to_hz Ran
ran, honoured the contract we drafted
minzwon/sota-music-tagging-models/training/modules.py
pointer only (licence: MIT); the stored copy get_code serves, from serkansulun/emsync, is pointer only (licence: NOASSERTION) · get_code("c1c8cd1745fa5a3e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from serkansulun/emsync, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_file Ran
ran on a synthesized input
minzwon/sota-music-tagging-models/training/robustness.py
code served (permissive licence) · get_code("8aa1dec7ce5bdeda")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import csv
import tempfile
import os

# Create a temporary TSV file with sample data
temp_dir = tempfile.mkdtemp()
tsv_file = os.path.join(temp_dir, 'test_tracks.tsv')

# Write sample TSV data
with open(tsv_file, 'w', newline='') as f:
    writer = csv.writer(f, delimiter='\t')
    # Header row
    writer.writerow(['track_id', 'col1', 'col2', 'audio_path', 'col4', 'col5', 'tag1', 'tag2', 'tag3'])
    # Data rows
    writer.writerow(['track_001', 'data1', 'data2', 'path/to/song1.mp3', 'data4', 'data5', 'electronic', 'dance', 'upbeat'])
    writer.writerow(['track_002', 'data1', 'data2', 'path/to/song2.mp3', 'data4', 'data5', 'rock', 'guitar', 'loud'])

ARGS = {
    'tsv_file': tsv_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
crop_audio Not yet run
unverified (harvested, no recorded run)
HephaestusProject/pytorch-FCN/infer.py
code served (permissive licence) · get_code("39b25bbed46ffd73")
How it ran

no invocation recorded.

get_audio_loader Not yet run
unverified (harvested, no recorded run)
minzwon/sota-music-tagging-models/training/data_loader/jamendo_loader.py
code served (permissive licence) · get_code("a2b6a8d6296c3474")
How it ran

no invocation recorded.

get_audio_loader Not yet run
unverified (harvested, no recorded run)
minzwon/sota-music-tagging-models/training/data_loader/msd_loader.py
code served (permissive licence) · get_code("e5021099627a13b1")
How it ran

no invocation recorded.

get_audio_loader Not yet run
unverified (harvested, no recorded run)
minzwon/sota-music-tagging-models/training/data_loader/mtat_loader.py
code served (permissive licence) · get_code("3665d5f930e76b94")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 3 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
gelu 198 other papers
53 dated earlier, 1 the same month, 142 later, 2 same year, order not determinable · 32 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.10125 (2019-01)
Glyce: Glyph-vectors for Chinese Character Representations
and 195 more not named here
hz_to_midi 1 other paper
1 later
2602.07063 (2026-02)
Video-based Music Generation
midi_to_hz 1 other paper
1 later
2602.07063 (2026-02)
Video-based Music Generation

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent advances in deep learning accelerated the development of content-based automatic music tagging systems. Music information retrieval (MIR) researchers proposed various architecture designs, mainly based on convolutional neural networks (CNNs), that achieve state-of-the-art results in this multi-label binary classification task. However, due to the differences in experimental setups followed by researchers, such as using different dataset splits and software versions for evaluation, it is difficult to compare the proposed architectures directly with each other. To facilitate further research, in this paper we conduct a consistent evaluation of different music tagging models on three datasets (MagnaTagATune, Million Song Dataset, and MTG-Jamendo) and provide reference results using common evaluation metrics (ROC-AUC and PR-AUC). Furthermore, all the models are evaluated with perturbed inputs to investigate the generalization capabilities concerning time stretch, pitch shift, dynamic range compression, and addition of white noise. For reproducibility, we provide the PyTorch implementations with the pre-trained models.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2006.00751")
get_code_for_paper("2006.00751")
have("2006.00751")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2006.00751.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2006.00751.svg)](https://syntology.ai/paper/2006.00751)

Connect an agent — have() is free.