🤗 CodeLlama 7B Finetuned Model (HF)
让Llama2执行、调试和保存代码，并访问互联网
本项目旨在赋予一个语言学习模型（LLM）生成、执行、调试代码以及回答查询的能力。 考虑一个用户请求："给我最后一个获得布克奖图书的亚马逊链接"。 这个任务涉及访问百科网站或搜索引擎并搜索亚马逊， 工作流程如下： 步骤1：用户提出请求。 步骤2：LLM生成代码以执行请求。 步骤3：如有必要，执行并调试代码。 步骤4：LLM根据结果回答用户的查询。 步骤5：记录并抽象该过程以备将来使用。
如果我们的LLM具有可重用的抽象化代码来执行此类任务，我们可以更有效地执行更复杂的任务。
The purpose and direction of the project
Run the Gradio App:
python3 chatbot.py --path Seungyoun/codellama-7b-instruct-pad- 🔥🔥🔥[2023/08/27] We're thrilled to announce that our 🤗 Llama2 Code Interpreter-7B (Finetuned from CodeLlama-7B-Instruct) model achieved a remarkable 70.12pass@1 on the HumanEval Benchmarks.
HumanEval
| Model | Score(pass@1) |
|---|---|
| Codellama instruct 7b | 34.8% |
| Codellama instruct 7b - finetuning | 70.12% |
GSM8K
| Model | Score |
|---|---|
| Code Llama 7B | 13% |
| Code Llama 13B | 20.8% |
| Codellama instruct 7b - finetuning | 28% |
- 🚀 Code Generation and Execution: Llama2 is capable of generating code, which it then automatically identifies and executes within its generated code blocks.
- Monitors and retains Python variables that were used in previously executed code blocks.
- 🌟 At the moment, my focus is on "Data development for GPT-4 code interpretation" and "Enhancing the model using this data". For more details, check out the feat/finetuning branch in our repository.
- 🌟 CodeLlama Support CodeLlama2
Llama2 in Action
In the GIF, Llama2 is seen in action. A user types in the request: Plot Nvidia 90 days chart. Llama2, an advanced code interpreter fine-tuned on a select dataset, swiftly queries Yahoo Finance. Moments later, it fetches the latest Nvidia stock prices from the past 90 days. Using Matplotlib, Llama2 then generates a clear and detailed stock price chart for Nvidia, showcasing its performance over the given period.
-
Clone the Repository (if you haven't already):
git clone https://github.com/SeungyounShin/Llama2-Code-Interpreter.git cd Llama2-Code-Interpreter -
Install the required dependencies:
pip install -r requirements.txt
To start interacting with Llama2 via the Gradio UI using codellama-7b-instruct-pad, follow the steps below:
- Run the Gradio App:
python3 chatbot.py --path Seungyoun/codellama-7b-instruct-pad
For those who want to use other models:
To start interacting with Llama2 via the Gradio UI using other models:
- Run the Command:
python3 chatbot.py --model_path <your-model-path>
Replace <your-model-path> with the path to the model file you wish to use. A recommended model for chat interactions is meta-llama/Llama-2-13b-chat.
Contributions, issues, and feature requests are welcome! Feel free to check issues page.
Distributed under the MIT License. See LICENSE for more information.
Seungyoun, Shin - [email protected]
Here are some relevant and related projects that have contributed to the development of this work:
- llama2 : GitHub Repository
- yet-another-gpt-tutorial : GitHub Repository
These projects have been instrumental in providing valuable insights and resources, and their contributions are highly appreciated.