Suggestion and invention of recipes using bi-directional LSTMs-based frameworks

Abstract Choosing which recipe to eat and avoid isn’t that simple for anyone. It takes strenuous efforts a lot of time people calculate the number calories P.H level dish. In this paper, we propose an ensemble neural network architecture suggests recipes based on taste person, calorie content recipes. We also bi-directional LSTMs-based variational autoencoder generating new have ensembled three LSTM-based recurrent networks can classify recipe. The proposed model predicts ratings custom loss function gave better results than standard functions after being trained with are fit person generates from existing After training testing autoencoder, tested 20 got overwhelming in experimentation, autoencoders generated couple recipes, healthy specific will be liked by person.