标签:

• 合并
• mergekit
• lazymergekit
• TheBloke/openchat_3.5-GPTQ
• asigalov61/Allegro-Music-Transformer 基础模型:
• TheBloke/openchat_3.5-GPTQ
• asigalov61/Allegro-Music-Transformer 许可证: apache-2.0 管道标签: 文本到音频

音乐生成模型

music_generation_model 是通过 LazyMergekit 合并以下模型而成:

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

🧩 配置

切片:
  - 来源:
      - 模型: TheBloke/openchat_3.5-GPTQ
        层范围: [0, 32]
      - 模型: asigalov61/Allegro-Music-Transformer
        层范围: [0, 32]
合并方法: slerp
基础模型: TheBloke/openchat_3.5-GPTQ
参数:
  t:
    - 过滤器: self_attn
      值: [0, 0.5, 0.3, 0.7, 1]
    - 过滤器: mlp
      值: [1, 0.5, 0.7, 0.3, 0]
    - 值: 0.5
数据类型: bfloat16

💻 使用方式

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "nagayama0706/music_generation_model"
messages = [{"role": "user", "content": "什么是大型语言模型？"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "文本生成",
    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"])