NLP with Tranformers
In the training program, you can understand and teach text to a computer using NLP methods and algorithms. In the world of machine learning, solutions that can understand text and generate text in an intelligent way, such as CHATGPT, have become important assistants for all businesses, government agencies, and all people. This training program will teach you fundamental concepts, advanced techniques, and practical applications in the field of NLP. After the training, you can either create your own startup or have all the technical knowledge you need to work as an NLP specialist in a company. This training is an ideal starting point for you to join the magical world of Machine Learning.
Training table
Information about the training
In the world where Artificial Intelligence and Machine Learning are rapidly spreading, NLP algorithms that teach computers to understand and create texts have been widely applied in recent years in areas such as sales, medicine, or the use of self-responding bots in any field, monitoring social media with sentiment analysis, extracting meaningful results from news, articles, contracts and other types of large-scale text data, etc. In the NLP training program with Transformers, you will learn advanced concepts for NLP, Transformer models, Transforming sentences into vectors, question-answer models, semantic (meaning-based) search systems, BERT, ChatGPT and CLIP models and methodologies. You will learn NLP by applying sentiment analysis (mood) in customer reviews, question-answer systems, building chatbots, building a text search system and search systems that find images from words. With Python, you will gain the ability to apply NLP algorithms using transformers, Pytorch, Keras and Tensorflow libraries.
Who is this training for?
Students
Software Developers
Data Scientists
Linguists
Artificial Intelligence Enthusiasts
Researchers and Academics
Certificate
Təlimi ilə bitirən şəxslərə Cerified Data Scientist və digər şəxslərə sertifikatı təqdim edir. Sağ tərəfdə nümunə sertifikatı görə bilərsiniz.
What are Transformers (Machine Learning Model)?
Transformatorlar (Maşın Öyrənmə Modeli) nədir?
IBM Technology
We are talking about a machine learning model, and in this video Martin Keen explains what transformers are, what they're good for, and maybe ... what they're not so good at for.
Syllabus
Session 1
Encoder-Decoder Attention, Self-Attention, and Multi-head Attention
Positional Encoding
Transformer Heads
Dot-Product Attention, Bidirectional Attention, Scaled Dot-Product Attention
Prebuilt Flair Models
Introduction to Sentiment Models with Transformers
Tokenization and Special Tokens for BERT
Making Predictions
Large-Scale Text Classification with Windows
Window Method in PyTorch
Case Study 1
Sentiment Analysis of Customer Reviews Using Transformer-Based Models
Session 2
Open-Domain & Reading Comprehension
Retrievers, Readers & Generators
Building a Q&A Model with SQuAD 2.0
Evaluating Q&A with Exact Match (EM) & ROUGE
Recall, Precision & F1 Score
Longest Common Subsequence (LCS)
Case Study 2
Building a Q&A System for News Using Open-Domain Reading Comprehension
Session 3
Retriever-Reader & Haystack Introduction
Elasticsearch in Haystack
Sparse Retrievers & BM25 Application
FAISS for Vector Search
DPR Architecture
Retriever-Reader Stack
Sentence Vector Extraction & Cosine Similarity
Similarity with Sentence-Transformers
Case Study 3
Building a Document Search and Reading System with Haystack
Session 4
Visual Guide for BERT Pretraining
Introduction to BERT for Pretraining Code
BERT Pretraining - Masked-Language Modeling (MLM) & Next Sentence Prediction (NSP)
MLM Logic & Pretraining
NSP Logic & Pretraining
Setting Up NSP Pretraining Training Loop
Case Study 4
Predicting the Next Sentence in Song Lyrics
Session 5
Introduction to ChatGPT API
ChatGPT API History Tutor - Part 1
ChatGPT API History Tutor - Part 2
Introduction to Whisper API
Whisper API - Audio Processing
Case Study 5
Creating a Chatbot with ChatGPT API
Session 6
Multimodal Generation - An Exciting Adventure
Using the CLIP Model
Building a Generative Transformer Model
Latent Space Parameters for Optimization
Encoding Text and Images with the CLIP Model
Interpolation Between Points in Latent Space
Davinci Sfumato
Case Study 6
Image Search Based on Text Queries Using the CLIP Model