Skip to content

Physics Informed Neural Networks. A deep learning framework for solving problems involving partial differential equations. In this particular project, working on SMILE Dataset, which is representation of chemical compounds in string format.

Notifications You must be signed in to change notification settings

vedaant912/PINNs

Repository files navigation

PINNs

Data

Download the dataset and unzip in root folder.

Files

  • training_with_pinns_with_early_stopping.py - To train the model on the basis of smile combinations.
  • Evaluation/Evaluating_Model.ipynb - Uses three different methods to Evaluate the Models.

Smile Emeddings

  • clean_chembl25.smi - Consists of all the SMILE value ( > 1.7M )
  • SMILE_Embeddings.ipynb - To create the vocabulary using all the SMILES obtained
  • Download the data related to Smile Embeddings smile_embedding and unzip in ./Smile Embeddings folder.
  • Reference : https://github.com/XinhaoLi74/SmilesPE

About

Physics Informed Neural Networks. A deep learning framework for solving problems involving partial differential equations. In this particular project, working on SMILE Dataset, which is representation of chemical compounds in string format.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published