lyeoni/nlp-tutorial

Covers foundational NLP architectures through annotated PyTorch implementations spanning text classification (CBoW, LSTM, Transformers), sequence-to-sequence machine translation with attention mechanisms, and language model pretraining on real-world datasets like HuffPost and Wikipedia. Includes multilingual examples and demonstrates both classical approaches (TF-IDF, TextCNN) alongside modern transformer-based methods with SentencePiece tokenization.

1,375 stars. No commits in the last 6 months.