rasta3050/aiguru_jazz

Jazz MIDI Model

Model Overview

This repository contains a model trained from scratch on a curated Jazz MIDI Dataset. The dataset consists of high-quality, genre-specific MIDI compositions, focusing on the intricate and dynamic elements of jazz music. This model is designed to generate jazz-style MIDI sequences and analyze jazz music patterns.

Training Details

• Dataset: Jazz MIDI Dataset (curated 20-second segments).
• Size: 549 compositions.
• Epochs: 4 epochs.
• Purpose: To train a model capable of generating and understanding jazz-specific musical structures, leveraging the most musically rich and dynamic portions of jazz compositions.

The model was trained to specialize in the unique characteristics of jazz, such as complex harmonies, improvisational motifs, and rhythmic diversity.

Files in the Repository

The repository includes the following files:

1. config.json:

   • Contains the configuration of the model architecture. This includes details such as the number of layers, hidden dimensions, attention heads, and other parameters used to define the model.

2. generation_config.json:

   • Contains generation-specific settings, such as maximum sequence length, temperature, top-k, and top-p sampling parameters. These configurations are crucial for controlling the behavior of the MIDI sequence generation process.

3. model.safetensors:

   • The model weights saved in the safetensors format for efficient and secure loading. This format ensures safe deserialization of model weights.

4. training_args.bin:

   • Stores the training arguments and hyperparameters used during training. This file can be useful for reproducing the training setup or understanding the specifics of the training process.

Dataset Details

Jazz MIDI Dataset

• Focus: Jazz genre-specific MIDI compositions.
• Size: 549 compositions (20-second segments from the middle of each composition).
• Preparation:
  • Removed duplicates.
  • Corrected formatting issues.
  • Extracted musically rich and dynamic segments from full compositions.

This dataset was specifically curated to provide a high-quality training set for jazz music generation.

Usage

This model is suitable for:

• Jazz MIDI music generation.
• Analysis of jazz-specific musical patterns.
• Creative applications in jazz-style music composition.

License

This model follows the licensing terms of the dataset sources. Please review the license for more details.