Data Science & Machine Learning for Developers – GOTO Academy NL

Data Science & Machine Learning for Developers

 

  

 

GOTO Academy is excited to bring you UK-based Phil Winder of Winder Research, for an intensive 3-day introduction to Data Science, which will leave you with practical tools for utilising Machine Learning principles in your organisation. This course combines beginner, intermediate and advanced courses into one efficient program. At the end of this course you will be considered professionally capable of developing and delivering data science products. You will be able to immediately start applying what you've learnt!
The 3-day course can be reduced to a consecutive 2-day course depending on your skill level and your learning goals. If you're unsure which package is right for you, please contact us.

  

Who is the Masterclass for?

This masterclass is aimed towards developers, in which you will delve into the mathematics behind the code as well as developing real life algorithms in Python. One-to-one help will be provided for developers new to Python and all algorithms, frameworks and libraries used will be demonstrated by the instructor.

 

Are there any prerequisites?

This is a full beginner to advanced level course, which is suitable for most users with some development experience. Some experience of Python is helpful. No data science experience is expected. If your development team already have some experience with data science you might like to inquire about the 2-day intermediate/advanced course.

  

This Masterclass will cover:

    Day 1

  •  How data science fits within a business context
  • Data science processes and language
  • Information and uncertainty
  • Types of learning
  • Modelling
  • Simple data cleaning
  • Regression Classification and clustering

 

   Day 2

  • Application of probability in data science and industry
  • More data preprocessing and cleaning and further experience with messy data
  • A range of simple models (logistic, linear, nonlinear SVMs, trees, etc.)
  • How to handle too many features
  • An in-depth practical example demonstrating the day’s concepts

 

   Day 3

  • When and why advanced methods are appropriate
  • Text mining, feature engineering, representation and learning
  • Neural network architectures and training experience
  • When ensembles are a good idea

 

Agenda: Day 1

  • Introduction: Applications, Disciplines, Lifecycle
  • Technical Overview: Techniques, Technologies, Decisions
  • Phase 1: Introduction to Working With Data

○ Visualising data

○ Scaling data

○ Dealing with corrupted data

 

  • Phase 2: Introduction to Modelling

○ Classification

○ Regression

○ Clustering 

 

  • Phase 3: Introduction to Evaluation

○ Numerical evaluation

○ Visual evaluation

 

Agenda: Day 2

  •  Probability theory and summary statistics
  • Generalisation and Overfitting
  • In-depth Data Cleaning
  • Feature engineering (derived data)
  • Feature selection
  • Time series data
  • Dimensionality reduction
  • Manifold learning
  • More classification
  • More regression
  • More clustering
  • Grand challenge

 

Agenda: Day 3

  • Natural language processing (NLP)
  • Ensembles
  • Deep Learning From the ground up
    • Activations
    • Multi-layer perceptrons
    • Autoencoders
    • Convolutional neural networks
    • Recurrent neural networks

 

Would you like to join as a group? - Contact us here

Upcoming training dates