Practical Guide to Data Engineering

Explain Live Session –

Aimed at independent users, this course in Reinforcement Learning (RL) will take you through some of the practical considerations to make when looking at how Reinforcement Learning can be used as a decision-making tool in your business. This course will take you through practical applications of RL and highlight areas where it can be used in your industry.  

Participants will learn about various types of RL algorithms such as Q-learning, REINFORCE Deep Q-learning, and Actor Critic networks while exploring practical application areas from use cases in industry.  

You will also complete practical exercises where you will gain some hands-on experience by train a Reinforcement Learning agent in a toy problem. In addition, you will gain an understanding about the importance of hyperparameter optimisation in Reinforcement learning.

In this level course, you will:

The Many Varieties of Database and Choosing a Suitable Persistence Solution 

To introduce and explain in plain, accessible language the main types of databases using open-source solutions such as Postgres, Cassandra, MongoDB and Neo4j. We will explain a range of fundamental differences and the factors involved when choosing a particular solution for different use-cases. We will explain important concepts such as data formats, data-files, partitioning, primary and secondary indexes, data normalisation vs de-normalisation, views & materialised-views, and CAP theorem (Consistency, Availability, Partitioning). We will cover classic relational database solutions (RDBMs) and the diverse types of NoSQL DB including Document Stores, Wide-Column Stores, Time-Series, and Graph DBs.  

Caching and In-Memory Data 

When latency is of critical concern, in-memory data caches can significantly increase the performance of your application. We will introduce the main caching strategies using open-source solutions including Redis and in-memory databases.  

Moving Data: Streams, Queues, Topics and Brokers 

To introduce the main concepts and patterns behind data streaming, messaging and asynchronous message channels in plain, accessible language using open-source solutions such as Kafka, Pulsar and RabbitMQ.  We will explain a range of different patterns such as Topics vs Queues, Pub-Sub, Competing Consumer and Message Routing, and explain what specific use-cases they can address.

There are no pre-requisites for this course

Create a free account to our Training Portal to register for a course and browse all available training courses.