Vikhrmodels/tool-plannings-v0.2

这是一个用于函数调用任务的英俄合成数据集，通过OpenAI API和环境中的可执行函数获得（行为接近现实）。数据集包括full_dataset、multiple_request、rejected_by_required、rejected_by_unavailable和single_request等不同配置。每个配置都有特定的功能，如id、tools和conversation，并且包含训练分割的数量和字节。数据集适用于英语和俄语，归类于文本生成任务。它还提到了数据集的大小类别，并提供了样本数据结构和使用说明。该数据集遵循Apache-2.0许可。

This is a synthetic dataset for function calling tasks in English and Russian, obtained using the OpenAI API and executable functions in the environment (behavior close to reality). The dataset includes configurations such as full_dataset, multiple_request, rejected_by_required, rejected_by_unavailable, and single_request. Each configuration has specific features like id, tools, and conversation, and includes train splits with the number of examples and bytes. The dataset is designed for English-Russian languages and is categorized under text-generation tasks. It also mentions the datasets size category and provides sample data structure and usage instructions. The dataset is licensed under Apache-2.0.