10.7K ·GitHub
What is this?
This is the Yandex School of Data Analysis (YSDA) Natural Language Processing course: a full set of lecture slides, seminar notebooks, and homework assignments organized by week, covering NLP from word embeddings through modern LLM topics.
This is the Yandex School of Data Analysis (YSDA) Natural Language Processing course: a full set of lecture slides, seminar notebooks, and homework assignments organized by week, covering NLP from word embeddings through modern LLM topics.Is this a coding library, or a set of course materials?It is course material, not a library: lecture and seminar content for each topic lives in its own 'week' folder, with a README pointing to the slides, notebooks, and homework for that week.
It is course material, not a library: lecture and seminar content for each topic lives in its own 'week' folder, with a README pointing to the slides, notebooks, and homework for that week.How do I start working through the course?Go into the week folder you want and follow its own README.md, which links the lecture, seminar notebook, and homework for that topic; there's a shared thread for installing the libraries the notebooks need.
Go into the week folder you want and follow its own README.md, which links the lecture, seminar notebook, and homework for that topic; there's a shared thread for installing the libraries the notebooks need.What background do I need before starting?Not explicitly stated, but the syllabus moves quickly from basic word embeddings into Transformers, PEFT, RLHF and RAG, so comfort with Python and basic deep learning is assumed rather than taught from scratch.
Not explicitly stated, but the syllabus moves quickly from basic word embeddings into Transformers, PEFT, RLHF and RAG, so comfort with Python and basic deep learning is assumed rather than taught from scratch.Is the whole course finished and up to date?No — this is the live 2025 edition, and later weeks are still being filled in as they are prepared rather than all being ready upfront.
No — this is the live 2025 edition, and later weeks are still being filled in as they are prepared rather than all being ready upfront.Who is this course meant for?It's built for YSDA's on-campus NLP students (with TAs grading their homework) but the notebooks and slides are public, so anyone studying NLP on their own can follow along.
It's built for YSDA's on-campus NLP students (with TAs grading their homework) but the notebooks and slides are public, so anyone studying NLP on their own can follow along.What topics does the 14-week syllabus cover?It starts with word embeddings, language modeling, seq2seq/attention, and Transformers, then moves into transfer learning, scaling LLMs, prompting/in-context learning, PEFT and RLHF fine-tuning, efficiency (quantization, distillation, speculative decoding), RAG, AI agents, interpretability, multimodal LLMs, and building LLM systems in production.
It starts with word embeddings, language modeling, seq2seq/attention, and Transformers, then moves into transfer learning, scaling LLMs, prompting/in-context learning, PEFT and RLHF fine-tuning, efficiency (quantization, distillation, speculative decoding), RAG, AI agents, interpretability, multimodal LLMs, and building LLM systems in production.Does each week come with homework?Most weeks pair a lecture and seminar with a homework assignment, such as building an n-gram language model from scratch, fine-tuning a pre-trained BERT model, or doing manual prompt engineering with chain-of-thought reasoning.
Most weeks pair a lecture and seminar with a homework assignment, such as building an n-gram language model from scratch, fine-tuning a pre-trained BERT model, or doing manual prompt engineering with chain-of-thought reasoning.What are similar projects?llm_interview_note, autumn-recruitment, LLM101n, dive-into-llms, llm-course, ai-engineering-hub, jeeves, train-llm-from-scratch
llm_interview_note, autumn-recruitment, LLM101n, dive-into-llms, llm-course, ai-engineering-hub, jeeves, train-llm-from-scratchWhich topics is it in?interpretability, university, rlhf, attention-mechanism, retrieval-augmented-generation, llms, lora, natural-language-processing, transformers, curriculum, reinforcement-learning, course
interpretability, university, rlhf, attention-mechanism, retrieval-augmented-generation, llms, lora, natural-language-processing, transformers, curriculum, reinforcement-learning, course