遇见数据集

MsAlEhR/scorpio-gene-taxa

收藏
Hugging Face2025-01-28 更新2025-02-15 收录
官方服务:

资源简介:

Scorpio-Gene-Taxa数据集是一个专门为评估DNA和蛋白质语言模型在基因和分类预测方面的性能而创建的 curated 基因序列数据集。它包含4634个基因组,每个属由一个基因组代表,重点放在细菌和古菌上。数据集分为训练集、测试集、Taxa_out和Gene_out,用于不同层次的模型评估。

The Scorpio-Gene-Taxa Dataset is a curated set of gene sequences created for evaluating the performance of DNA and protein language models in gene and taxonomy prediction. It contains 4,634 genomes, with each genus represented by a single genome, focusing on bacteria and archaea. The dataset is split into training, test, Taxa_out, and Gene_out sets for different levels of model evaluation.

提供机构:
MsAlEhR
二维码
社区交流群
二维码
科研交流群
商业服务