TucanoBR/wikipedia-PT

Wikipedia-PT数据集是从维基百科的葡萄牙语部分提取的，主要用于语言建模任务。数据集的每个实例包含一个文本字段，表示文章的内容。数据集仅包含一个训练集，数据来源于维基百科的dump文件，并使用`mwparserfromhell`工具进行解析。数据集的文本内容遵循GNU自由文档许可证和Creative Commons Attribution-Share-Alike 3.0许可证。

The Portuguese portion of the Wikipedia dataset is built from the Wikipedia dumps and contains text articles in Portuguese. The dataset is primarily used for Language Modeling tasks and includes a single train split with 1,103,446 samples. The text content of the dataset is parsed using the `mwparserfromhell` tool. The dataset is licensed under the GNU Free Documentation License and the Creative Commons Attribution-Share-Alike 3.0 License.