Sequence-to-sequence framework with a focus on Neural Machine Translation based on PyTorch

Implementations for a family of attention mechanisms, suitable for all kinds of natural language processing tasks and compatible with TensorFlow 2.0 and Keras.

Code for the paper "STConvS2S: Spatiotemporal Convolutional Sequence to Sequence Network for Weather Forecasting" (Neurocomputing, Elsevier)

An Implementation of Transformer (Attention Is All You Need) in DyNet

Encoder-Decoder model for Semantic Role Labeling

Handwriting Trajectory Recovery using End-to-End Deep Encoder-Decoder Network, ICPR 2018.

Established a deep learning model which can translate English words/sentences into their corresponding French translations.

Successfully developed an encoder-decoder based sequence to sequence (Seq2Seq) model which can summarize the entire text of an Indian news summary into a short paragraph with limited number of words.

The proposed framework to retrieve the continuous chunk-level emotions via emo-rankers for Seq2Seq SER

Sequence to Sequence Transformer implementation in order to train a model to translate over Cap-verdian criole to English.

A deep learning-based AI tool for correcting grammatical and spelling errors using TensorFlow. Features LSTM-based sequence-to-sequence modeling, data preprocessing, and real-time correction capabilities.

Grounded Sequence-to-Sequence Transduction Team at JSALT 2018

This repository shows the implementation of the paper Neural Machine Translation by Jointly Learning to Align and Translate

A concise summary generator for Amazon product reviews built using Transformers which maintains the original semantic essence and user sentiment

Successfully established a neural machine translation model using sequence to sequence modeling which can successfully translate English sentences to their corresponding German translations.

Sequence to sequence encoder-decoder model with Attention for Neural Machine Translation

Sentiment analysis on the IMDB dataset using Bag of Words models (Unigram, Bigram, Trigram, Bigram with TF-IDF) and Sequence to Sequence models (one-hot vectors, word embeddings, pretrained embeddings like GloVe, and transformers with positional embeddings).