scikit-fingerprints/MoleculeNet_HIV

MoleculeNet HIV数据集是MoleculeNet基准测试的一部分，主要用于预测分子抑制HIV复制的能力。该数据集包含41127个样本，任务类型为分类，推荐使用scaffold分割方法，评估指标为AUROC。

The HIV dataset is part of the MoleculeNet benchmark, intended to be used through the scikit-fingerprints library. The task is to predict the ability of molecules to inhibit HIV replication. The dataset contains 41127 samples, with a recommended scaffold split method and AUROC as the evaluation metric. It falls under the categories of tabular classification, graph machine learning, and text classification, and involves chemistry, biology, and medical fields.

搜集汇总

背景与挑战

背景概述

该数据集是一个用于预测分子抑制HIV病毒复制能力的分类数据集，包含41,127个样本，推荐使用AUROC评估模型性能。