| --- | |
| license: cc-by-nc-sa-4.0 | |
| language: | |
| - fr | |
| tags: | |
| - legal | |
| - finance | |
| pretty_name: MicRou | |
| size_categories: | |
| - n<1K | |
| --- | |
| # MicRou | |
| ## Introduction | |
| The documents that constitute the dataset were gathered for a RAG project in memory of [Michel Rouger](https://www.pitti.io/articles/michel-rouger) : the documents were part of his personal archives and include his own work as well as work produced by other authors during projects he ran. | |
| ## Datasets | |
| The [MicRou repository](https://github.com/pappitti/MicRou) includes 2 datasets in French: | |
| 1. microu | |
| This dataset includes approximately 850 documents in French (books, articles, minutes of debates) produced between 1998 and 2020. It covers justice and law, finance and economics, management, healthcare, education, sports, history and geopolitics... Overall it represents between 1.5m and 2m tokens depending on the tokenizer you use. | |
| In many cases, the documents stem from a larger source that was broken down as parts could be considered independently (e.g. different chapters of a book or different articles of a newsletter). It is nonetheless possible to recombine the entire source : within a "dossier", you can group by date and, within each group, order by index. Documents that do not come from a larger source have an index of 0 by default. | |
| 2. microu-chunked | |
| As part of the RAG projet, we used an embeddings model, [Solon](https://huggingface.co/OrdalieTech/Solon-embeddings-large-0.1), with a context window of 512 tokens so we had to split the MicRou dataset into chunks. This is the resulting dataset. | |
| ## License (CC-BY-NC-SA-4.0) | |
| The dataset is currently under restrictive license. We plan to convert it to an open license once we have finalized the review of the right holders. Some documents may be excluded following the review, but we also plan to add others over time. | |