pageman/drugrl-btk-benchmark-v1.1
收藏Hugging Face2025-12-11 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/pageman/drugrl-btk-benchmark-v1.1
下载链接
链接失效反馈官方服务:
资源简介:
这是一个针对BTK(Brutons Tyrosine Kinase)靶点的药物优化强化学习基准数据集,提供了可重复的方法论,用于选择高质量的候选药物,并在随机基线基础上实现了164%的改进。数据集内容包括一个经过验证的工作笔记本和文档,但不包括训练好的模型文件。数据集的目标是展示如何使用表格Q学习来优化药物化合物的选择,平衡三个目标:疗效(40%)、安全性(40%)和选择性(20%)。
A proven reinforcement learning benchmark for drug compound optimization targeting BTK (Brutons Tyrosine Kinase). This dataset contains the reproducible methodology that achieved +164% improvement over random baseline in selecting high-quality drug candidates. The dataset includes a proven working notebook and documentation, but not the trained model file. The benchmark demonstrates using tabular Q-learning to optimize drug compound selection by balancing three objectives: efficacy (40%), safety (40%), and selectivity (20%).
提供机构:
pageman



