pythonic-function-calling

# Pythonic Function Calling Dataset

This dataset contains synthetic data used for training Pythonic function calling models [Dria-Agent-a-3B](https://huggingface.co/driaforall/Dria-Agent-a-3B) and [Dria-Agent-a-7B](https://huggingface.co/driaforall/Dria-Agent-a-7B).

[Dria](docs.dria.co) is a python framework to generate synthetic data on globally connected edge devices with 50+ models. See the network [here](https://dria.co/edge-ai)

## Dataset Summary

The dataset includes various examples of function calling scenarios, ranging from simple to complex multi-turn interactions.
It was generated synthetically using the Dria network on edge devices with various models.

### Categories:
- Simple function calls
- Parallel function executions
- Multiple function interactions
- Chained function calls
- Multi-turn conversations

## Data

Here's the distribution of example types in the dataset as a Markdown table:

| Type | Description | Percentage |
|------|-------------|------------|
| simple | Single function schema, single function call | 27.6% |
| parallel | Single function schema, multiple function calls | 27.5% |
| multiple | Multiple function schemas, single function call | 12.2% |
| step_by_step | Multiple function schemas, multiple function calls, with step by step reasoning | 21.5% |
| multi_turn | Multiple function schemas, multiple function calls, multi-turn | 11.1% |

Each entry contains:

| Feature Name | Type | Description |
|-------------|------|-------------|
| id | string | Unique identifier for each example |
| domain | string | Main category of the function calling scenario |
| subdomain | string | Specific subcategory within the domain |
| tools | string | Available functions/tools for the scenario in as python functions |
| conversations | list | List of conversation turns |
| - content | string | The actual content of each conversation turn |
| - role | string | The role of the speaker (e.g., user, assistant) |
| type | string | Type of function calling scenario |

Note: The dataset includes step-by-step thinking examples that show the reasoning process. All chained entries contain these detailed reasoning steps, while other categories have partial coverage. This helps models learn both direct execution and detailed reasoning approaches.

## License

Apache 2.0

## Additional Information

The dataset emphasizes real-world use cases, particularly developer-centric scenarios.

For more detailed information about the methodology and approach, visit the [model documentation](https://huggingface.co/driaforall/Dria-Agent-a-7B).

#### Citation

```
@misc{Dria-Agent-a,
url={https://huggingface.co/blog/andthattoo/dria-agent-a},
title={Pythonic Function Calling Dataset},
author={"andthattoo", "Atakan Tekparmak"}
}
```

# Python风格函数调用数据集（Pythonic Function Calling Dataset）

本数据集为合成生成数据集，用于训练Python风格函数调用模型[Dria-Agent-a-3B](https://huggingface.co/driaforall/Dria-Agent-a-3B)与[Dria-Agent-a-7B](https://huggingface.co/driaforall/Dria-Agent-a-7B)。

[Dria](docs.dria.co)是一款Python框架，可在全球互联的边缘设备上基于50余种模型生成合成数据。可在此处查看其网络架构：https://dria.co/edge-ai

## 数据集概述

本数据集涵盖各类函数调用场景示例，范围从简单场景到复杂的多轮交互。其通过Dria网络在边缘设备上借助多种模型合成生成。

### 分类：
- 简单函数调用
- 并行函数执行
- 多函数交互
- 链式函数调用
- 多轮对话

## 数据

本数据集的示例类型分布如下Markdown表格所示：

| 示例类型 | 描述 | 占比 |
|------|-------------|------------|
| simple | 单函数模式，单次函数调用 | 27.6% |
| parallel | 单函数模式，多次函数调用 | 27.5% |
| multiple | 多函数模式，单次函数调用 | 12.2% |
| step_by_step | 多函数模式，多次函数调用，附带分步推理 | 21.5% |
| multi_turn | 多函数模式，多次函数调用，多轮对话 | 11.1% |

每条数据包含以下字段：

| 字段名 | 类型 | 描述 |
|-------------|------|-------------|
| id | 字符串 | 每条示例的唯一标识符 |
| domain | 字符串 | 函数调用场景的主分类 |
| subdomain | 字符串 | 领域内的具体子分类 |
| tools | 字符串 | 场景中可用的函数/工具，以Python函数形式呈现 |
| conversations | 列表 | 对话轮次列表 |
| - content | 字符串 | 单条对话轮次的实际内容 |
| - role | 字符串 | 发言者角色（如用户、助手） |
| type | 字符串 | 函数调用场景的类型 |

注意：本数据集包含展示推理过程的分步思考示例。所有链式调用条目均包含此类详细推理步骤，其余分类也有部分覆盖。这有助于模型同时学习直接执行与详细推理两种路径。

## 许可证

Apache 2.0

## 补充说明

本数据集侧重真实应用场景，尤其是面向开发者的使用案例。
如需了解方法与实现细节的更多信息，请参阅[模型文档](https://huggingface.co/driaforall/Dria-Agent-a-7B)。

### 引用格式

@misc{Dria-Agent-a,
url={https://huggingface.co/blog/andthattoo/dria-agent-a},
title={"Pythonic Function Calling Dataset"},
author={"andthattoo", "Atakan Tekparmak"}
}