More than 32 Gb RAM is required to train the entire dataset. Code for training the entire dataset: https://www.kaggle.com/mmelahi/organic-chemistry-reaction-prediction?scriptVersionId=7438604
Thanks. Now it works as expected. Any reason you trained on only 10,000 reactions? How can I re-train the model on the entire dataset?
Spaces should be added. "N C @ @ H C ( = O ) O . S = C = S". Also, I've not trained the model on the entire dataset, which has nearly 1 million reactions. I've used only 10,000 reactions of small length.
Thanks for the reply. I had to change the keras backend on my machine. Now the program opens up. However, no matter what input I give I get the same output now. For example, when I enter : NC@@HC(=O)O.S=C=S I get the following as the predicted smile: OCCCOCCOCC This does not make sense as it does not fllow basic principles of atom conservation. Is there a step which I am missing for running the software?
I have used tensorflow as the backend to keras. So, try installing tensorflow.
Theano not found issue