Music Generation Model

🚀 music_generation_model

This is a music generation model that merges multiple models using LazyMergekit, offering text - to - audio capabilities.

🚀 Quick Start

The music_generation_model is a merge of the following models using LazyMergekit:

• TheBloke/openchat_3.5-GPTQ
• asigalov61/Allegro-Music-Transformer

✨ Features

• Model Merging: Combines the strengths of TheBloke/openchat_3.5-GPTQ and asigalov61/Allegro-Music-Transformer.
• Text-to-Audio: Capable of generating audio content from text input.

📦 Installation

To use this model, you need to install the necessary libraries. You can do this using the following command:

!pip install -qU transformers accelerate

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer
import transformers
import torch

model = "nagayama0706/music_generation_model"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

📚 Documentation

🧩 Configuration

The following is the configuration used to merge the models:

slices:
  - sources:
      - model: TheBloke/openchat_3.5-GPTQ
        layer_range: [0, 32]
      - model: asigalov61/Allegro-Music-Transformer
        layer_range: [0, 32]
merge_method: slerp
base_model: TheBloke/openchat_3.5-GPTQ
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

📄 License

This project is licensed under the apache-2.0 license.