Chemical-Compound-Classification

Binary Classification of Chemical Compounds on the basis of taste using Graph Neural Networks

This is a machine learning solution for the classification of chemical compounds. We have extracted and preprocessed data from a CSV file containing chemical information. Then trained a neural network to predict whether a compound is bitter or sweet based on its atomic structure.

Tools and Technologies: Python, NumPy, pandas, scikit-learn, TensorFlow, PyTorch Model: Used the SchNet model in PyTorch for the implementation.