Transformer Training

Infoshare is the largest tech community in CEE and the organizer of the leading tech conference in Gdańsk. It connects startups, investors, corporations, and innovation enthusiasts. It promotes entrepreneurship, knowledge sharing, and networking. Through events, competitions, and programs, it supports the development of the tech ecosystem in Poland and the region.

The training on transformer models is an intensive two-day course that focuses on the practical application of the latest NLP models, such as BERT and GPT, using the Hugging Face library and TensorFlow Hub. The training program is designed so that 80% of the time is dedicated to practical workshops and 20% to theory. Participants will gain the skills necessary to implement and train advanced transformer models in various NLP applications.

• Programmers and data engineers who want to expand their skills with the latest NLP techniques
• IT specialists who want to use transformer models for automating language processing
• Data scientists and data analysts wishing to process and analyze text using advanced models
• Individuals with basic programming knowledge in Python and basic knowledge of machine learning
• Experience with cloud services will be an additional asset

• How to install and configure the Hugging Face library and TensorFlow Hub to work with transformer models
• How to apply transformer models for text classification, named entity recognition, and text generation
• How to use and fine-tune pre-trained transformer models, such as BERT and GPT
• How to optimize model performance and deploy them in a production environment

• Day 1: Introduction to transformer models and basics of Hugging Face

  • Basics of transformer models

  • Introduction to transformer architecture – history and evolution

  • Key components and principles of operation

  • Introduction to Hugging Face and TensorFlow Hub

  • Installation and configuration of libraries

  • Overview of available models and their applications

  • Basic operations with transformer models

    • Tokenization and text processing

    • Loading and using pre-trained models

    • Implementation in text processing tasks

  • Practical exercises with loading and testing models

  • Results analysis and basic optimization

• Day 2: Advanced techniques and practical applications

  • Training and fine-tuning transformer models

    • Training and fine-tuning techniques for pre-trained models

    • Using own data for training models

  • Applications of transformer models

    • Text classification

    • Named entity recognition (NER)

    • Text generation and machine translation

    • Sentiment analysis

  • Fine-tuning and applications of models

    • Implementing fine-tuning on real datasets

    • Creating projects using BERT and GPT models

  • Optimization and deployment of models

    • Techniques for optimizing model performance

    • Deploying models in a production environment

16 h/2 days