A Solubility Model Web Applciation

In this project, I train a solubility model and deploy it as a web app. I use data provided by Lowe et. al. (2023). Please follow the link to my GitHub Repository

So far, I have trained a very simple and crude classification model around 0.001 M (or mol/L) cutoff value, meaning that compounds with solubility < 0.001 are insoluble. Model returns ‘0’ for insoluble and ‘1’ for soluble compounds. I used Python for training the model and Streamlit for building the web app. I was inspired by the tutorial and code from Data Professor and shared the link to his YouTube video in the references section.

What’s next?

The next step is to deploy the app using Flask or Heroku. Stay tuned!