Artificial Neural Network Modeling of Drug Solubility in Cosolvent Systems

This notebook implements an artificial neural network (ANN) to predict drug solubility in cosolvent systems. It includes data preprocessing, model architecture definition, training, and performance evaluation using regression metrics. The goal is to explore the non-linear relationships between solubility and experimental variables such as temperature and cosolvent composition.

本Jupyter笔记本实现了人工神经网络（Artificial Neural Network，ANN），用于预测助溶剂体系中的药物溶解度。该实现涵盖数据预处理、模型架构定义、模型训练以及基于回归指标的性能评估环节。本研究的核心目标为探究药物溶解度与温度、助溶剂组成等实验变量之间的非线性关联。