Introduction to Machine Learning

This course takes 3 days with 8 hours each (on site) or 5 hours each (online).
The only difference is that there will be less practical exercises in an online course. However, we will hand them over to you and you can still do them on your own and ask our consultants for feedback or help if needed.

Below you can find the topics that can be covered in this course. The actual choice of topics depends on the needs and interests of the course participants.

Machine Learning: Introduction and explanation of main concepts.

About Python/Jupyter

Overview of main Python libraries to be used

Exploratory Data Analysis: in theory

Exploratory Data Analysis: in practice

Supervised Learning: Introduction and explanation of main concepts.

• Explanation of Training Process: Training/Validation/Testing, Cross-validation
• Recap: Regression vs Classification
• Cost/Loss functions
• What is done in training? Minimization of Loss Function. Example Algorithms
• Performance Evaluation
• Steps for successful Model building

Regression

Classification

• Classification Algorithms
• Ensemble Algorithms
• Performance Evaluation

Unsupervised Learning

• Unsupervised Algorithms
• Ensemble Algorithms
• Performance Evaluation

From lab to production: challenges and common problems

The importance of distributed computing in Machine Learning

Introduction to neural networks

• Definition
• Main Concepts
• Tensorflow Playground demo
• Activation Functions
• NN training process
• Multi-class Problems and Softmax
• Convolutional NN
• Keras Demo and Explanation of Keras Library
• Transfer Learning
• Hyperparameter Tuning

Deep Learning: What is it and where can it be applied?

Machine Learning in my organization: How can I implement ML considering the current problems we face?