Reading Material

Speech and Language Processing (3rd ed. draft)
Dan Jurafsky and James H. Martin

Word embeddings:

• [Chapter 6]
• Surveys on conventional count-based models and the more recent representations (embeddings)
• The illustrated Word2vec
• Embeddings in NLP

Text classification and regression:

• [Chapter 4] and [Chapter 5]

Language models:

• N-gram [Chapter 3]
• RNN [Chapter 8]

Transformers:

• Stanford’s Foundation models
• The illustrated Transformer
• The annotated Transformer

Pytorch tutorials:

• Stanford’s introduction to Pytorch
• Deep learning with Pytorch, a 60-min blitz

NLP demos:

• AllenNLP demos: reading comprehension, NER, sentiment analysis, MLM, etc.
• Farsi demos: NER, colloquial to formal transfer, EL, sentiment analysis, and sentence similarity

Resources for Farsi:

• A list of datasets and tools for farsi: [nlpdataset.ir]
• DadmaTools NLP toolkit for Farsi