Masterclass: AI Bootcamp

Take AI seriously and kickstart your journey!

This training provides a broad introduction to practical machine learning (ML). Training focuses on supervised ML and artificial neural networks. Theoretical background and relevant points of each topic are introduced, followed by instructor-led programming exercises.

Participants will learn how to apply ready-made Python libraries for various ML solutions. First two days focus on establishing basic skills on modeling and simple ML algorithms. Next three days will focus on designing and training artificial neural networks.

What to expect?

This training covers various topics with hands-on programming exercises. Here’s what will be covered during five intensive days:

• Overview of ML-AI
• Artificial neural networks
• Deep learning
• Basics of modeling
• K-nearest neighbors method
• Decision tree algorithms
• Data pre-processing
• Multilayer perceptrons
• Convolution neural networks
• Autoencoders
• Sequence models
• Best practices in machine learning
• Scikit learn
• Tensorflow
• Keras
• Programming exercises 50 %

Who gain the most from this AI Bootcamp and how it is done?

This course is aimed for persons who are familiar with Python programming language and would like to get good understanding on how to develop ML/AI applications.

Learning goals:

• Good general knowledge on machine learning and neural networks for supervised machine learning.
• Ability to design and train simple machine learning models to analyse real-world data.
• Ability to learn more independently.

Learning by doing:

• Lectures (50 %) and hands-on exercises (50 %).
• Theoretical background and relevant points of each topic are introduced, following by instructor-led programming exercises.
• Course includes small programming homework.
• Exercises are fully based on Python libraries and interfaces:
• Numpy
• Keras-Tensorflow
• Scikit-learn
• Pandas

Prerequisites: Fluent/experienced in Python programming language.
Training language: Finnish
Language of training materials: English.

Suggested pre-material before the course:

• How to calculate Standard Deviation and Variance
• Four Types Of Cross Validation| K-Fold | Leave One Out |Bootstrap | Hold Out
• MIT open courseware, Introduction to Machine Learning
• Precision, recall, F-score
• Deep Learning – Computerphile

Contents of AI Bootcamp day-by-day

Day 1:

Review of machine learning history
Basic statistics by Python
Review of machine learning methods
The bias-variance dilemma
K-nearest neighbors algorithm

Day 2:

Cross-validation
Performance KPIs
Decision trees

Day 3:

History of ANNs
About data pre-processing
Multilayer perceptron
Homework

Day 4:

Deep learning
Learning strategies
Autoencoders

Day 5:

Convolution neural networks
Sequence models, LSTM
About best practices

Before coming to the course – technical requirements

• All exercises are run on participants´ computers.
• Install Python Anaconda development environment with Python 3.7. (https://www.anaconda.com/distribution/)
• Participants must have rights to download and install Python libraries to their Anaconda development environment.
• Preferably 16 Gb memory or more.
• Preferably Intel Core i5 or better.
• In principle, any Python environment will do. However, Anaconda is strongly recommended because it has most of the required libraries in-built.

• Keras is implemented into Tensorflow and doesn’t need to be installed separately.
• Easiest way to install libraries is to use PyPi via Anaconda prompt:
  • pip install tensorflow
• To test if everything works, try to import keras from tensorflow in Anaconda’s Spyder code editor. Also pandas and sklearn:
  • from tensorflow import keras
  • import pandas
  • import sklearn
    
• Course material will be shared via OneDrive as the course progress.
• To use OneDrive, participants must have access to email and internet.

Your trainer

Juhani Teeriniemi

Roles:
Automation developer
Trainer

Services:
RPA development
Software development (Python)
Machine learning
Advanced data analytics
Test automation (functional, performance, load)
Training (RPA, machine learning, data analytics, Python)

Studies:
D.Sc. (Tech.) 2016, Aalto University of Technology

Industry expertise:
Machine learning
Analytical and numerical modeling
Training
Software automation
Test automation (functional, performance, load)
Research (machine learning, quantum mechanics, materials science, water treatment)

Courses, cerficates & achievements:
DASA DevOps Enable & Scale
Professional Scrum Master I
UiPath Certified Professional RPA Developer Beginner User
ISTQB Software Testing Fundamentals
Several publications in peer-reviewed international science journals.

Avainsanat: AI, Machine Learning, Python