codegemma

智能代码补全和生成：无论您是在本地工作还是使用 Google Cloud 资源，都能完成行、函数，甚至生成完整的代码块。
增强精度：CodeGemma 模型经过 5000 亿个令牌的训练，这些令牌主要来自网络文档、数学和代码的英文语言数据，生成不仅语法上更正确，而且语义上更有意义的代码，从而减少错误和调试时间。
多语言能力：支持 Python、JavaScript、Java、Kotlin、C++、C#、Rust、Go 等语言。
简化工作流程：将 CodeGemma 模型集成到您的开发环境中，以便更快地编写更少的样板代码，并将重点放在有趣且差异化的代码上。

中间填充

CodeGemma 模型支持中间填充 (FIM)，用于自动完成或编码辅助工具。以下是用 Ollama Python 库的示例

response = generate(
  model='codegemma:2b-code',
  prompt=f'<|fim_prefix|>{prefix}<|fim_suffix|>{suffix}<|fim_middle|>',
  options={
    'num_predict': 128,
    'temperature': 0,
    'top_p': 0.9,
    'stop': ['<|file_separator|>'],
  },
)

参考资料

Hugging Face

报告

CodeGemma is a collection of powerful, lightweight models that can perform a variety of coding tasks like fill-in-the-middle code completion, code generation, natural language understanding, mathematical reasoning, and instruction following.

### Variants:

* `instruct` a 7b instruction-tuned variant for natural language-to-code chat and instruction following
* `code` a 7b pretrained variant that specializes in code completion and generation from code prefixes and/or suffixes
* `2b` a state of the art 2B pretrained variant that provides up to 2x faster code completion

### Advantages:

* **Intelligent code completion and generation**: Complete lines, functions, and even generate entire blocks of code, whether you're working locally or using Google Cloud resources.

* **Enhanced accuracy**: Trained on 500 billion tokens of primarily English language data from web documents, mathematics, and code, CodeGemma models generate code that's not only more syntactically correct but also semantically meaningful, reducing errors and debugging time.

* **Multi-language proficiency**: Supports Python, JavaScript, Java, Kotlin, C++, C#, Rust, Go, and other languages.

* **Streamlined workflows**: Integrate a CodeGemma model into your development environment to write less boilerplate and focus on interesting and differentiated code that matters, faster.

![benchmarks](https://github.com/ollama/ollama/assets/251292/0d8473cb-bcee-4bd0-9214-c527ce367d88)

### Fill-in-the-middle

CodeGemma models support fill-in-the-middle (FIM), for use in autocomplete or coding assistant tooling. Below is an example using the Ollama [Python](https://github.com/ollama/ollama-python) library:

```python
response = generate(
  model='codegemma:2b-code',
  prompt=f'<|fim_prefix|>{prefix}<|fim_suffix|>{suffix}<|fim_middle|>',
  options={
    'num_predict': 128,
    'temperature': 0,
    'top_p': 0.9,
    'stop': ['<|file_separator|>'],
  },
)
```

### References

[Hugging Face](https://hugging-face.cn/collections/google/codegemma-release-66152ac7b683e2667abdee11)

[Report](https://storage.googleapis.com/deepmind-media/gemma/codegemma_report.pdf)

粘贴、拖放或单击以上传图片（.png、.jpeg、.jpg、.svg、.gif）