Next Word Prediction using LSTM

[snehapm04]

Closes: #97

• Title : Next Word Prediction using LSTM
• Your Name: Sneha P M
• Identify yourself: SWOC Contributor

Describe the add-ons or changes you've made 📃

I have added the following enhancements to the project

1. Imported and structured the necessary libraries for text preprocessing, sequence creation, and model development using TensorFlow and Keras.
2. Loaded and preprocessed the dataset by cleaning text, tokenizing sentences, and generating input-output word sequences.
3. Visualized word distributions and sequence lengths to understand the dataset's structure and potential challenges.
4. Designed and trained an LSTM-based neural network for next-word prediction, using embedding layers and sequential modeling techniques.
5. Evaluated model performance using metrics like categorical cross-entropy loss and accuracy, and visualized training and validation loss trends.

Checklist: ☑️

• My code follows the Contributing Guidelines & Code of Conduct of this project.
• This PR does not contain plagiarized content.
• I have performed a self-review of my own code.
• I have commented my code, particularly wherever it was hard to understand.
• My changes generate no new warnings.

Screenshots 📷

Working Video 🎥

Next Word Prediction Recording

Happy Coding 🎉

[github-actions]

Thank you for submitting your pull request! We'll review it as soon as possible. For further communication, join our discord server https://discord.gg/tSqtvHUJzE.

[Avdhesh-Varshney]

Really Appreaciable Work!

[11PRIMUS]

@snehapm04 use Bidirectional LSTM so model can get contextual of past and future and improve the accuracy