raphavlas/gdb13
收藏Hugging Face2025-05-20 更新2025-08-30 收录
下载链接：
https://hf-mirror.com/datasets/raphavlas/gdb13
下载链接
链接失效反馈官方服务：
资源简介：
GDB13数据集包含了大约20亿个由13个原子组成的分子，采用7种启发式方法生成。该数据集用于基于分子属性进行条件生成测试或属性预测。数据集移除了原子位置信息，保留了SMILES表示作为结构信息。数据集中的分子属性包括SMILES表示、质量、折射率、疏水性等，这些属性均通过rdkit工具计算得出。
The GDB13 dataset contains approximately 2 billion molecules composed of 13 atoms, generated using 7 heuristics. It is used for conditional generation tests based on molecular properties or for predicting these properties. The dataset has removed atomic position information, retaining only the SMILES representation as structural information. Molecular properties in the dataset include SMILES representation, molecular weight, refractive index, hydrophobicity, etc., all calculated using the rdkit tool.
提供机构：
raphavlas
