MID-FiLD: MIDI Dataset for Fine-Level Dynamics

Authors: Jesung Ryu, Seungyeon Rhyu, Hong-Gyu Yoon, Eunchong Kim, Ju Young Yang, Taehyun Kim

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Based on our dataset, we conduct the following analysis and experiments to demonstrate its excellence. We conducted an ablation study for objective evaluation, successively removing items in metadata as a model input.
Researcher Affiliation Collaboration Jesung Ryu1, Seungyeon Rhyu1, Hong-Gyu Yoon1, Eunchong Kim1, Ju Young Yang2, Taehyun Kim1* 1Pozalabs, Republic of Korea 2Duke University, United States
Pseudocode No The paper describes the input encoding approach and problem definition, but it does not include any explicitly labeled pseudocode blocks or algorithms.
Open Source Code No The paper does not provide any direct links to a code repository or an explicit statement about the public release of the source code for the methodology described.
Open Datasets No The paper introduces a new dataset, MID-Fi LD, but does not provide any specific link (URL, DOI, or repository name) or an explicit statement confirming its public availability for download.
Dataset Splits Yes We divided 4,422 MID-Fi LD samples into a ratio of 8:1:1 for training, validation and test sets respectively, leaving about 10% of the data for test performance measurements (i.e., 3, 547 : 443 : 432).
Hardware Specification No The paper does not provide any specific hardware details such as GPU models (e.g., NVIDIA A100, RTX 2080 Ti), CPU models, or memory configurations used for running the experiments.
Software Dependencies No The paper mentions using deep learning models and a support vector machine but does not specify any software versions for libraries, frameworks (e.g., PyTorch 1.x, TensorFlow 2.x), or programming languages (e.g., Python 3.x).
Experiment Setup Yes We used 9-fold cross-validation for training the SVM (C = 10, γ = 1) following the division ratio of the dataset.