Bytes igniting the literary passion (a beta phase model).
Poetica is an AI with Long Short Term Memory Network (LSTM) with over 200 cells using Recurrent Neural Networks (RNNs) that can compose Shakespeare-like poetry given the line starter and the word count needed. This artificial neural net has a vocabulary of more than 3800 words and an accuracy rate of over 80% on it’s n-grammed training dataset.
View notebook on Kaggle
- Overall LSTM Network Architecture -
- Neural Net Architecture -
The current version of poetica (v1.0) has accuracy of around 80% over 500 epochs and still has a scope of improvement. To contribute to this project, the notebook can be copied from kaggle and enhacements to the neural network can be done to raise the overall performance.
View dataset (Credits: Harshal Gadhe)
If you have any feedback or suggestions please reach out to the project admin sanidhyak or you can create an issue and mention there which new features can be added to make Poetica better.