100 Gecs

🚀 HuggingArtists Model: 100 gecs

This model is designed to generate lyrics in the style of 100 gecs. It offers a way to create text based on the unique lyrical style of this artist, providing an interesting tool for lyric generation and exploration.

🚀 Quick Start

You can use this model directly with a pipeline for text generation:

from transformers import pipeline
generator = pipeline('text-generation',
                     model='huggingartists/100-gecs')
generator("I am", num_return_sequences=5)

Or with Transformers library:

from transformers import AutoTokenizer, AutoModelWithLMHead
  
tokenizer = AutoTokenizer.from_pretrained("huggingartists/100-gecs")

model = AutoModelWithLMHead.from_pretrained("huggingartists/100-gecs")

✨ Features

• Artist-Specific Lyric Generation: Generate lyrics in the style of 100 gecs.
• Easy to Use: Can be used with a simple pipeline or the Transformers library.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import pipeline
generator = pipeline('text-generation',
                     model='huggingartists/100-gecs')
generator("I am", num_return_sequences=5)

Advanced Usage

from transformers import AutoTokenizer, AutoModelWithLMHead
  
tokenizer = AutoTokenizer.from_pretrained("huggingartists/100-gecs")

model = AutoModelWithLMHead.from_pretrained("huggingartists/100-gecs")

📚 Documentation

How does it work?

To understand how the model was developed, check the W&B report.

Training data

The model was trained on lyrics from 100 gecs. Dataset is available here. And can be used with:

from datasets import load_dataset

dataset = load_dataset("huggingartists/100-gecs")

Explore the data, which is tracked with W&B artifacts at every step of the pipeline.

Training procedure

The model is based on a pre - trained GPT-2 which is fine - tuned on 100 gecs's lyrics. Hyperparameters and metrics are recorded in the W&B training run for full transparency and reproducibility. At the end of training, the final model is logged and versioned.

Limitations and bias

The model suffers from the same limitations and bias as GPT-2. In addition, the data present in the user's tweets further affects the text generated by the model.

🔧 Technical Details

The model is built on a pre - trained GPT - 2 architecture and fine - tuned on the lyrics of 100 gecs. The training process and related metrics are tracked using Weights & Biases (W&B) for better transparency and reproducibility.

📄 License

No license information is provided in the original document.

About

Built by Aleksey Korshuk

For more details, visit the project repository.

I was made with huggingartists.

Create your own bot based on your favorite artist with the demo!