polyBERT: A chemical language model to enable fully machine-driven ultrafast polymer informatics

This repository contains the code and data set of the polyBERT paper, which can be accessed on arXiv. polyBERT is a large chemical language machine learning model that has learned the PSMILES chemical language of polymers. It generates polymer fingerprints (PSMILES-based) quickly, facilitating rapid polymer informatics and machine-driven property predictions. polyBERT is compatible with GPUs and CPUs, and can easily scale on cloud systems.