Public Data files for DeepCDM

Public data files for experiments in DeepCDM. merged_MoNA_ESI-MSMS_spectra.sdf is used for model pre-training, replicate_from_mona.sdf is used for validation in pre-training. fine_tuning_example.sdf is a demo training set used for model fine-tuning, while replicate_from_dansylation_example.sdf is a demo for validation. Larger training set is recommended for better performance. Moreover, the training set for fine-tuning can be changed for variable CDMs. example_molecule.sdf is a example of molecular structural information for spectrum prediction. All raw data of wastewater samples.用于DeepCDM实验的公开数据集文件。 merged_MoNA_ESI-MSMS_spectra.sdf 用于模型预训练，replicate_from_mona.sdf 用于预训练阶段的验证。 fine_tuning_example.sdf 为模型微调所用的演示训练集，replicate_from_dansylation_example.sdf 则为验证用演示集。为获得更优的模型性能，建议使用规模更大的训练集。此外，针对不同的CDM，可更换用于微调的训练集。 example_molecule.sdf 为用于光谱预测的分子结构信息示例。 所有废水样本的原始数据。