O'Reilly logo
live online training icon Live Online training

Modern AI Programming with Python

Explore the world of AI and how to implement it's different sub-domains using Python

Alberto Boschetti

Artificial Intelligence is considered to be one of the most advanced areas in tech, and it unfolds into various industry verticals. Python is the go-to programming language, providing a myriad of benefits in helping to develop AI projects that have algorithms with less code, pre-built libraries, and platform-agnostic features.

This live training course will highlight the different aspects of Artificial Intelligence and how they can be put to practical use using the highly popular Python programming language. We will cover the different sub-domains of Artificial Intelligence, such as Machine Learning, Deep Learning, Reinforcement Learning, and Genetic Algorithms, along with real-world implementations of them -- including Natural Language Processing, Computer Vision, Heuristic Search, Semantic Web, and more.

We will also discuss different Python libraries such as scikit-learn, Tensorflow, and Keras, and how they can be used for building smart, intelligent systems.

What you'll learn-and how you can apply it

  • An overview of Artificial Intelligence, and its different sub-domains, such as Machine Learning, Deep Learning, and Reinforcement Learning
  • Implementation of the popular AI algorithms and techniques using Python, and its libraries such as Keras, Tensorflow, and scikit-learn
  • Implementation of real-world use cases of AI, such as NLP, and speech recognition
  • How to put machine learning models in production, on your infrastructure or on a cloud-based environment

This training course is for you because...

This training targets beginners to intermediate Python users.

Prerequisites

  • Python 3.5 or later (may be installed via Conda, via binaries, or using an online Notebook, like Google Colab).
  • A minimumal understanding of Python 3.x and general concepts of data science is recommended.

Recommended Preparation

About your instructor

  • Alberto Boschetti is a data scientist, with strong expertise in signal processing and statistics. He holds a Ph.D. in Telecommunication Engineering and currently lives and works in London. In his work projects he daily faces challenges spanning among natural language processing (NLP), machine learning and probabilistic graph models. He is very passionate about his job and he always tries to be updated on the latest development of data science technologies, attending meetups, conferences and other events.

Schedule

The timeframes are only estimates and may vary according to how the class is progressing

Section 1: Deeper questions are around (10 mins)

  • AI in the era of Big Data: from Machine learning to Deep Learning
  • Why deep learning? Wasn’t Machine learning enough?
  • Challenges and the benefits of deep learning
  • How data monetization has changed with deep learning
  • Agenda

Section 2: A step back: an introduction to AI tasks (15 mins)

  • An introduction of the problems we can solve with AI

Section 3: Python preprocessing (20 mins)

  • Light introduction to Python and the motivations of why it’s the choice of many Data Scientists (and comparison wrt R, C, Java)
  • Variables and features
  • Low level libraries: a short introduction to NumPy and SciPy
  • Big data acquisition and preprocessing: pySpark & pandas

Labs: How to read data with pandas and some basics operation that can be done via pySpark.

Section 4: Python machine learning (15 mins)

  • Practical machine learning with Python: scikit-learn
  • Pipelines with scikit-learn
  • How to get the best out of the problem: hyperparameter search and cross-validation
  • How to handle big data? A guide to Luigi and Apache Airflow

Labs: A simple spam classifier, a simple predictor of house prices

Break (10 mins)

Section 5: Python deep learning (15 mins)

  • Tensorflow basics
  • From CPU to GPU (and TPU)
  • Pipelines with Tensorflow
  • A higher lever approach: Keras
  • Alternatives: pyTorch

Labs: Practical examples: Deep Network, CNN, RNN

Section 6: Python reinforcement learning (10 mins)

  • Keras and Tensorflow for reinforcement learning
  • Practical examples: teach an agent to play games
  • Labs: teach an agent to play games

Section 7: Cloud-based resources (15 mins)

  • Google Cloud
  • AWS

Q&A (10 mins)