---
tags:
- generated_from_keras_callback
- music
datasets:
- juancopi81/mutopia_guitar_dataset
widget:
- text: PIECE_START TIME_SIGNATURE=4_4 BPM=90 TRACK_START INST=0 DENSITY=2 BAR_START
    NOTE_ON=43
  example_title: Time signature 4/4, BPM=90, NOTE=G2
base_model: gpt2
model-index:
- name: juancopi81/mutopia_guitar_mmm
  results: []
---

# juancopi81/mutopia_guitar_mmm

Music generation could be approached similarly to language generation. There are many ways to represent music as text and then use a language model to create a model capable of music generation. For encoding MIDI files as text, I am using the excellent [implementation](https://github.com/AI-Guru/MMM-JSB) of Dr. Tristan Beheren of the paper: [MMM: Exploring Conditional Multi-Track Music Generation with the Transformer](https://arxiv.org/abs/2008.06048).

This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the [Mutopia Guitar Dataset](https://huggingface.co/datasets/juancopi81/mutopia_guitar_dataset). Use the widget to generate your piece, and then use [this notebook](https://colab.research.google.com/drive/14vlJwCvDmNH6SFfVuYY0Y18qTbaHEJCY?usp=sharing) to listen to the results (work in progress). 
I created the notebook as an adaptation of [the one created by Dr. Tristan Behrens](https://huggingface.co/TristanBehrens/js-fakes-4bars).

It achieves the following results on the evaluation set:
- Train Loss: 0.5365
- Validation Loss: 1.5482

## Model description

The model is GPT-2 loaded with the GPT2LMHeadModel architecture from Hugging Face. The context size is 256, and the vocabulary size is 588. The model uses a 
`WhitespaceSplit` pre-tokenizer. The [tokenizer](https://huggingface.co/juancopi81/mutopia_guitar_dataset_tokenizer) is also in the Hugging Face hub. 

## Intended uses & limitations

I built this model to learn more about how to use Hugging Face. I am implementing some of the parts of the [Hugging Face course](https://huggingface.co/course/chapter1/1) with a project that I find interesting. 
The main intention of this model is educational. I am creating a [series of notebooks](https://github.com/juancopi81/MMM_Mutopia_Guitar) where I show every step of the process:
- Collecting the data
- Pre-processing the data
- Training a tokenizer from scratch
- Fine-tuning a GPT-2 model
- Building a Gradio app for the model

I trained the model using the free version of Colab with a small dataset. Right now, it is heavily overfitting. My idea is to have a more extensive dataset of Guitar Music from Latinoamerica to train a new model similar to the Mutopia Guitar Model, using more GPU resources.

## Training and evaluation data

I am training the model with [Mutopia Guitar Dataset](https://huggingface.co/datasets/juancopi81/mutopia_guitar_dataset). This dataset consists of the soloist guitar pieces of the [Mutopia Project](https://www.mutopiaproject.org/). 
The dataset mainly contains guitar music from western classical composers, such as Sor, Aguado, Carcassi, and Giuliani.

For the first epochs of training, I transposed the notes by raising and lowering the pitches using the twelve semi-tones of an entire octave. Later, I trained the model without transposing the pieces so that generation shows better results of a real guitar piece.

### Training hyperparameters

<details>
<summary>Click to expand</summary>

The following hyperparameters were used during training (with transposition):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 5726, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition - first round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, '__passive_serialization__': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition - second round): 
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, '__passive_serialization__': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition, new tokenizer - third round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition, new tokenizer - fourth round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition, new tokenizer - fifth round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, '__passive_serialization__': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition, new tokenizer - sixth round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-07, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-07, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

The following hyperparameters were used during training (without transposition, new tokenizer - seventh round):
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0005, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0005, 'decay_steps': 1025, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}

- training_precision: mixed_float16
</details>

### Training results

<details>
<summary>Click to expand</summary>
Using transposition:
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 1.0705     | 1.3590          | 0     |
| 0.8889     | 1.3702          | 1     |
| 0.7588     | 1.3974          | 2     |
| 0.7294     | 1.4813          | 3     |
| 0.6263     | 1.5263          | 4     |
| 0.5841     | 1.5263          | 5     |
| 0.5844     | 1.5263          | 6     |
| 0.5837     | 1.5346         | 7     |
| 0.5798     | 1.5411         | 8     |
| 0.5773     | 1.5440         | 9     |

Without transposition (first round):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 0.5503     | 1.5436          | 0     |
| 0.5503     | 1.5425          | 1     |
| 0.5476     | 1.5425          | 2     |
| 0.5467     | 1.5425          | 3     |
| 0.5447     | 1.5431          | 4     |
| 0.5418     | 1.5447          | 5     |
| 0.5418     | 1.5451          | 6     |
| 0.5401     | 1.5472         | 7     |
| 0.5386     | 1.5479         | 8     |
| 0.5365     | 1.5482         | 9     |

Without transposition (second round):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 0.5368     | 1.5482          | 0     |
| 0.5355     | 1.5480          | 1     |
| 0.5326     | 1.5488          | 2     |
| 0.5363     | 1.5493          | 3     |
| 0.5346     | 1.5488          | 4     |
| 0.5329     | 1.5502          | 5     |
| 0.5329     | 1.5514          | 6     |
| 0.5308     | 1.5514         | 7     |
| 0.5292     | 1.5536         | 8     |
| 0.5272     | 1.5543         | 9     |

Without transposition (third round - new tokenizer):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 6.1361 | 6.4569 | 0 |
| 5.6383 | 5.8249 | 1 |
| 4.9125 | 4.8956 | 2 |
| 4.2013 | 4.2778 | 3 |
| 3.8665 | 4.0330 | 4 |
| 3.7106 | 3.8956 | 5 |
| 3.6041 | 3.7995 | 6 |
| 3.5301 | 3.7485 | 7 |
| 3.4973 | 3.7323 | 8 |
| 3.4909 | 3.7323 | 9 |

Without transposition (fourth round - new tokenizer):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 3.4879 | 3.7206 | 0 |
| 3.4667 | 3.6874 | 1 |
| 3.4229 | 3.6373 | 2 |
| 3.3680 | 3.5751 | 3 |
| 3.2998 | 3.5026 | 4 |
| 3.2208 | 3.4240 | 5 |
| 3.1385 | 3.3397 | 6 |
| 3.0580 | 3.2587 | 7 |
| 2.9949 | 3.2118 | 8 |
| 2.9646 | 3.1958 | 9 |

Without transposition (fifth round - new tokenizer):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 2.9562 | 3.1902 | 0 |
| 2.9457 | 3.1751 | 1 |
| 2.9266 | 3.1512 | 2 |
| 2.9039 | 3.1176 | 3 |
| 2.8705 | 3.0775 | 4 |
| 2.8291 | 3.0295 | 5 |
| 2.7872 | 2.9811 | 6 |
| 2.7394 | 2.9321 | 7 |
| 2.6996 | 2.9023 | 8 |
| 2.6819 | 2.8927 | 9 |

Without transposition (sixth round - new tokenizer):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 2.6769 | 2.8894 | 0 |
| 2.6719 | 2.8791 | 1 |
| 2.6612 | 2.8638 | 2 |
| 2.6465 | 2.8439 | 3 |
| 2.6242 | 2.8174 | 4 |
| 2.6006 | 2.7877 | 5 |
| 2.5679 | 2.7554 | 6 |
| 2.5387 | 2.7223 | 7 |
| 2.5115 | 2.7029 | 8 |
| 2.5011 | 2.6970 | 9 |

Without transposition (seventh round - new tokenizer):
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 2.2881 | 2.2059 | 0 |
| 1.7702 | 1.8533 | 1 |
| 1.4625 | 1.6948 | 2 |
| 1.2876 | 1.6865 | 3 |
| 1.1926 | 1.6414 | 4 |
| 1.1329 | 1.6360 | 5 |
| 1.1069 | 1.6448 | 6 |
| 1.0408 | 1.6207 | 7 |
| 0.8939 | 1.5837 | 8 |
| 0.7265 | 1.5901 | 9 |
| 0.5902 | 1.6261 | 10 |
| 0.4489 | 1.7007 | 11 |
| 0.3223 | 1.7940 | 12 |
| 0.2158 | 1.9032 | 13 |
| 0.1448 | 1.9892 | 14 |
</details>

### Framework versions
- Transformers 4.22.1
- TensorFlow 2.8.2
- Datasets 2.5.1
- Tokenizers 0.12.1