PraxySante/Qwen3-ASR-PostTrain-Complete-Medical-French-Full
收藏资源简介：
该数据集是一个多源的法语医学文本集合，包含来自多个不同来源的文本数据，例如药物警戒数据库（bdpm）、医疗转录（nemo_transcriptions）、患者病历（parcomed、parcomed_research）、医学文献（hitz_medical）、法语维基百科（wikipedia_fr）、网络爬取语料（c4_fr、fineweb2_fr）、对话（claire_dialogue）、字幕（open_subtitles_fr）等。每个样本包含文本内容和来源标签，总样本数超过千万，主要用于训练和评估法语自然语言处理模型，尤其是医学领域。
This dataset is a multi-source collection of French medical texts, comprising data from various sources such as pharmacovigilance database (bdpm), medical transcriptions (nemo_transcriptions), patient records (parcomed, parcomed_research), medical literature (hitz_medical), French Wikipedia (wikipedia_fr), web-crawled corpora (c4_fr, fineweb2_fr), dialogues (claire_dialogue), subtitles (open_subtitles_fr), and more. Each sample consists of a text field and a source label, with over 10 million examples in total, intended for training and evaluating French NLP models, particularly in the medical domain.
