LSTM-based Seq2Seq Translation Model: English ↔ Ewe
This repository contains an LSTM-based Seq2Seq translation model trained to translate between English and Ewe. The model uses a sequence-to-sequence architecture with LSTM layers to perform translation tasks with high accuracy.
📖 Overview
The LSTM-based Seq2Seq model is designed to translate sentences between English and Ewe. It leverages a recurrent neural network (RNN) architecture with LSTM layers to encode input sentences and decode the translations.
🛠️ Model Architecture
Encoder: Embedding layer followed by LSTM layers that encode the input sentence into a context vector.
Decoder: Embedding layer followed by LSTM layers that generate the translated output sentence.
Output Layer: A fully connected layer to predict the next word in the sequence.
Model Hyperparameters:
Embedding Dimension: 256
Hidden Dimension: 512
Number of LSTM Layers: 2
Batch Size: 32
Learning Rate: 0.001
Epochs: 20
📊 Evaluation Metrics
The model was evaluated using the following metrics:
BLEU Score: Measures the overlap between predicted and reference translations.
CHRF Score: Measures character-level n-gram F-scores between predicted and reference translations.
