Skip to content

Latest commit

 

History

History
9 lines (7 loc) · 773 Bytes

README.md

File metadata and controls

9 lines (7 loc) · 773 Bytes

Deep-Learning-for-NLP

This repo contains code for:

  1. Using the Word2Vec package from the Gensim library [https://radimrehurek.com/gensim/index.html] for producing word2vec embeddings on your own dataset
  2. Using a bidirectional stacked RNN with LSTM cells for classifying whether a movie review of the IMDB dataset is positive or negative
  3. Using a Convolutional Neural Network for classifying whether a movie review of the IMDB dataset is positive or negative
  4. Using the Optuna framework for faster hyperparameter tuning of the above architectures ((2)(3))
  5. Finetuning a pretrained model from HugginFace on the IMDB Dataset for classifying whether a movie review of the IMDB dataset is positive or negative. Here DistilBERT is used for computational efficiency.