Even though chemistry has become a more data driven discipline in the recent years, the amount of data available for training deep learning models is limited when compared to their imaging counterparts (ImageNet, for example). Transfer learning is a strategy that aims to leverage representations learnt, by training deep learning models on larger datasets with available labels, and then using the trained network for fine-tuning on smaller, costly to label datasets.
This project is about porting a Transfer Learning Framework, ChemNet into DeepChem. This would involve reproducing the results of the paper and a blog-post, Jupyter notebook detailing its use. A ChemNet API will also be developed to allow any DeepChem TensorGraph model to use this framework.