scikit-fingerprints/MoleculeNet_Tox21
收藏Hugging Face2025-02-02 更新2024-07-22 收录
下载链接:
https://hf-mirror.com/datasets/scikit-fingerprints/MoleculeNet_Tox21
下载链接
链接失效反馈官方服务:
资源简介:
MoleculeNet Tox21数据集是MoleculeNet基准测试的一部分,主要用于预测12种毒性目标,包括核受体和应激反应途径。所有任务都是二分类任务。数据集包含7831个样本,推荐使用scaffold分割方法,并使用AUROC作为评估指标。需要注意的是,目标值存在缺失,算法应仅在存在的标签上进行评估,训练数据可能需要用零值进行填补。
The MoleculeNet Tox21 dataset is part of the MoleculeNet benchmark, primarily used in the fields of chemistry and biomedicine. It contains 7831 samples for predicting 12 toxicity targets, including nuclear receptors and stress response pathways. All tasks are binary classification tasks with missing values. The recommended evaluation metric is AUROC, and scaffold splitting is suggested for data splitting.
提供机构:
scikit-fingerprints



