This repository implements an end to end encoder transformer model completely from scratch in C

This repository implements an end to end encoder transformer model completely from scratch in C

PseudoCode2Cpp is a deep learning model that translates pseudocode into C++ using a Transformer-based Seq2Seq architecture. Implemented from scratch, it leverages attention mechanisms, positional encoding, and beam search for high-quality code generation. Trained on the SPOC dataset in Google Colab.