Contact Us

Expert course

Practical AI with Python and Reinforcement Learning

Learn how to use Reinforcement Learning techniques to create practical Artificial Intelligence programs!

Rating: 4.61,289 ratings13,232 students26.5 total hours156 lectures
Open course in DoJo Back to courses
Practical AI with Python and Reinforcement Learning Expert

Course facts

  • Last updated 04/2023
  • English English [Auto], French [Auto] , 3 more
  • Instructor: Jose Portilla, Pierian Training
  • technical implementation with AI models and applications

What you'll learn

Practical outcomes

  • Reinforcement Learning with Python
  • Creating Artificial Neural Networks with TensorFlow
  • Using TensorFlow to create Convolution Neural Networks for Images
  • Using OpenAI to work with built-in game environments
  • Using OpenAI to create your own environments for any problem
  • Create Artificially Intelligent Agents
  • Tabular Q-Learning
  • State–action–reward–state–action (SARSA)
  • Deep Q-Learning (DQN)
  • DQN using Convolutional Neural Networks
  • Cross Entropy Method for Reinforcement Learning
  • Double DQN
  • Dueling DQN

Curriculum

14 sections • 156 lectures • 26h 24m total length

Course Overview3 lectures • 17min
  1. Welcome Message00:37
  2. Course Curriculum Overview09:34
  3. Course Success and Overview06:44
  4. Check In Quiz 3 questions
Course Set-Up and Installation Procedures3 lectures • 24min
  1. Anaconda and Jupyter Notebook Install and Setup13:49
  2. Note on Environment Setup00:27
  3. Environment Setup Walkthrough10:09
Numpy Basics Overview6 lectures • 53min
  1. Introduction to Numpy Section02:14
  2. NumPy Arrays22:41
  3. Numpy Operations - Part One11:06
  4. Numpy Operations - Part Two08:14
  5. Numpy Exercise Overview01:18
  6. Numpy Exercise Solutions07:05
Matplotlib and Visualization Overview11 lectures • 1hr 51min
  1. Introduction to Matplotlib04:06
  2. Matplotlib Basics12:35
  3. Matplotlib - Understanding the Figure Object07:32
  4. Matplotlib - Implementing Figures and Axes14:31
  5. Matplotlib - Figure Parameters04:56
  6. Matplotlib - Subplots Functionality19:17
  7. Matplotlib Styling - Legends07:02
  8. Matplotlib Styling - Colors and Styles14:29
  9. Advanced Matplotlib Commands (Optional)03:52
  10. Matplotlib Exercise Questions Overview06:10
  11. Matplotlib Exercise Questions - Solutions16:39
Machine Learning, Deep Learning, and Reinforcement Learning2 lectures • 25min
  1. What is Machine Learning, Deep Learning, and Artificial Intelligence?11:08
  2. Supervised Machine Learning Process13:41
Pandas and Scikit-Learn Crash Course9 lectures • 1hr 43min
  1. Pandas and Scikit-Learn Overview00:09
  2. Pandas - Series Part One09:28
  3. Pandas - Series Part Two10:41
  4. Pandas - DataFrames - Part One19:27
  5. Pandas - DataFrames - Part Two08:18
  6. Pandas - DataFrames - Part Three13:57
  7. Pandas - DataFrames - Part Four14:30
  8. Scikit-Learn - Using Train-Test-Split11:20
  9. Scikit-Learn - Using Metrics14:48
Artificial Neural Network and TensorFlow Basics27 lectures • 5hr 1min
  1. Introduction to Artificial Neural Networks02:15
  2. Perceptron Model10:39
  3. Neural Networks07:19
  4. Activation Functions10:39
  5. Multi-Class Classification Considerations10:34
  6. Cost Functions and Gradient Descent18:13
  7. Backpropagation14:47
  8. TensorFlow vs. Keras Explained02:13
  9. Keras Syntax - Preparing the Data10:49
  10. Keras Syntax - Creating and Training the Model13:59
  11. Keras Syntax - Model Evaluation12:56
  12. Keras Regression - Exploratory Data Analysis18:50
  13. Keras Regression - EDA Continued13:15
  14. Keras Regression - Data Preprocessing and Model Creation08:42
  15. Keras Regression - Model Evaluation and Predictions11:23
  16. Keras Classification - EDA and Preprocessing08:05
  17. Keras Classification - Overfitting and Evaluation16:50
  18. Keras Classification - Overview of Project Options01:40
  19. Keras Project Notebook Exercise Overview07:41
  20. Keras Project Solution - Exploratoy Data Analysis20:35
  21. Keras Project Solutions - Missing Data - Part One14:46
  22. Keras Project Solutions - Dealing with Missing Data - Part Two12:02
  23. Keras Project Solutions - Categorical Data17:23
  24. Keras Project Solutions - Data Preprocessing03:45
  25. Keras Project Solutions- Creating and Training the Model03:57
  26. Keras Project Solutions - Model Evaluation09:42
  27. Tensorboard18:22
Convolutional Neural Networks with TensorFlow17 lectures • 2hr 42min
  1. Convolutional Neural Networks Section Overview01:33
  2. Image Filters and Kernels11:35
  3. Convolutional Layers14:01
  4. Pooling Layers06:47
  5. MNIST Data Set Overview04:41
  6. CNN on MNIST - The Data12:57
  7. CNN on MNIST - Creating and Training the Model16:14
  8. CNN on MNIST - Model Evaluation06:53
  9. CNN on CIFAR-10 - The Data11:23
  10. CNN on CIFAR-10 - Evaluating the Model07:05
  11. Downloading Data Set for Real Image Lectures05:22
  12. CNN on Real Image Files - Reading in the Data14:54
  13. CNN on Real Image Files - Data Generation15:37
  14. CNN on Real Image Files - Creating the Model13:37
  15. CNN on Real Image Files - Model Evaluation08:49
  16. CNN Exercise Project Overview02:10
  17. CNN Exercise Project Solutions08:31
Reinforcement Learning - Core Concepts5 lectures • 32min
  1. Overview of Core Concepts for Reinforcement Learning Section01:37
  2. Agents, Environments, and Policy11:43
  3. Rewards, Discount Factors, and Bellman Equation13:35
  4. Deterministic vs. Stochastic Processes05:17
  5. Tabular Reinforcement Learning00:08
Open AI Gym Overview6 lectures • 1hr 21min
  1. Introduction to OpenAI Gym Section01:04
  2. OpenAI Overview and History11:40
  3. OpenAI Gym - Documentation Tour12:38
  4. OpenAI Gym - Environment Key Ideas07:51
  5. OpenAI Gym - Working with the Environment27:23
  6. OpenAI Gym - Agent Interacting with the Environment20:30
Classical Q Learning19 lectures • 3hr 53min
  1. Introduction to Classical Q-Learning Overview04:12
  2. History of Q-Learning03:53
  3. Q-Learning Theory - Part One - Table Intuition15:38
  4. Q-Learning Theory - Part Two - Q Target Equation12:08
  5. Q-Learning Theory - Part Three - Q-Update Equation08:28
  6. Q-Learning Theory - Part Four - Programmatic Q Updates10:56
  7. Q-Learning Implementation - Part One - Environment Setup14:15
  8. Q-Learning Implementation - Part Two - Table and Hyperparameters10:34
  9. Q-Learning Implementation - Part Three - Update Functions17:10
  10. Q-Learning Implementation - Part Four - Agent Training17:03
  11. Q-Learning Implementation - Part Five - Visualization and Utilization09:50
  12. Continuous Q-Learning Theory - Part One - Environment Setup12:33
  13. Continuous Q-Learning Theory - Part Two- Q-Table Shape16:33
  14. Continuous Q-Learning Theory - Part Three - Discretization Theory04:57
  15. Continuous Q-Learning - Part Four - Discretization Implementation17:38
  16. Continuous Q-Learning - Part Five - Functions and Hyperparameters10:08
  17. Continuous Q-Learning - Part Six - Training and Usage19:35
  18. Q-Learning Exercise Project07:06
  19. Q-Learning Exercise Project - Solutions20:51
Deep Q-Learning17 lectures • 2hr 49min
  1. DQN Section Overview01:57
  2. History of DQN04:45
  3. DQN Theory and Intuition - Part One - Review of Core RL Ideas04:48
  4. DQN Theory and Intuition - Part Two - Neural Networks for RL10:50
  5. DQN Theory and Intuition - Part Three - Feedback and Function Approximation21:10
  6. DQN Theory and Intuition - Part Four - Experience Replay18:46
  7. DQN Theory and Intuition - Part Five - Mapping Key Ideas to Code16:04
  8. DQN Manual Implementation - Part One - Imports and Environment04:44
  9. DQN Manual Implementation - Part Two - Artificial Neural Network07:16
  10. DQN Manual Implementation - Part Three - Hyperparameters and Functions18:43
  11. DQN Manual Implementation - Part Four - Model Training15:51
  12. DQN - Keras-RL2 - Part One - Overview07:20
  13. DQN - Keras-RL2 - Part Two - Imports and Environment03:26
  14. DQN - Keras-RL2 - Part Three - Creating the ANN05:40
  15. DQN - Keras-RL2 - Part Four - DQN Agent13:43
  16. DQN - Exercise Overview03:28
  17. DQN - Exercise Solutions10:07
Deep Q-Learning on Images11 lectures • 2hr 1min
  1. Introduction to Deep Q-Learning on Images04:44
  2. Files for DQN on Images00:02
  3. Key Image Concepts Review07:00
  4. Image History in Replay Buffer - Concept Review06:25
  5. Processing Images Part Three- Coding Replay Buffer and Sequences14:22
  6. Processing Images Part Four - Coding Preprocessing12:08
  7. DQN on Images - Part One - Imports and Processing18:31
  8. DQN on Images - Part Two - Constructing the Network12:00
  9. DQN on Images - Part Three - Setting up the Agent19:25
  10. DQN Exercises Overview06:05
  11. DQN Exercises Solution20:05
Creating Custom OpenAI Gym Environments20 lectures • 2hr 32min
  1. Introduction to Custom OpenAI Gym Environments02:53
  2. Custom Gym Environments - Class Structure Overview08:35
  3. Creating Custom Game - Part One - Snake Game Overview04:08
  4. Creating Custom Game - Part two - Class Structure and Setup07:25
  5. Creating Custom Game - Part Three - Game Reset06:58
  6. Creating Custom Game - Part Four - Direction and Movement07:55
  7. Creating Custom Game - Part Five - Eating and Spawning Food03:57
  8. Creating Custom Game - Part Six - Human Input07:12
  9. Creating Custom Game - Part Seven - Displaying Score04:31
  10. Creating Custom Game - Part Eight - Game Over Conditions03:38
  11. Creating Custom Game - Part Nine - Ending the Game03:18
  12. Creating Custom Game - Part Ten - Play Logic16:35
  13. Important Note on __init__.py Files00:34
  14. OpenAI Gym Environment - Directory Structure Overview05:57
  15. Gym Directory and Setup File07:06
  16. Filling in the __init__.py Files04:27
  17. Conversion of Game to OpenAI Gym Environment Class25:15
  18. Conversion of Game to OpenAI Gym Environment - Part Two14:34
  19. Registration of the Environment09:23
  20. Training an Agent on the Custom Environment07:51

Who it is for

  • Python developers familiar with basics of machine learning, such as Scikit-Learn, but now want to learn how to create Artificially Intelligent Agents through Reinforcement Learning

Course description

Overview

Please note! This course is in an "early bird" release, and we're still updating and adding content to it, please keep in mind before enrolling that the course is not yet complete. “The future is already here – it’s just not very evenly distributed.“ Have you ever wondered how Artificial Intelligence actually works? Do you want to be able to harness the power of neural networks and reinforcement learning to create intelligent agents that can solve tasks with human level complexity? This is the ultimate course online for learning how to use Python to harness the power of Neural Networks to create Artificially Intelligent agents! This course focuses on a practical approach that puts you in the driver's seat to actually build and create intelligent agents, instead of just showing you small toy examples like many other online courses. Here we focus on giving you the power to apply artificial intelligence to your own problems, environments, and situations, not just those included in a niche library! This course covers the following topics: Artificial Neural Networks Convolution Neural Networks Classical Q-Learning Deep Q-Learning SARSA Cross Entropy Methods Double DQN and much more! We've designed this course to get you to be able to create your own deep reinforcement learning agents on your own environments. It focuses on a practical approach with the right balance of theory and intuition with useable code. The course uses clear examples in slides to connect mathematical equations to practical code implementation, before showing how to manually implement the equations that conduct reinforcement learning. We'll first show you how Deep Learning with Keras and TensorFlow works, before diving into Reinforcement Learning concepts, such as Q-Learning. Then we can combine these ideas to walk you through Deep Reinforcement Learning agents, such as Deep Q-Networks! There is still a lot more to come, I hope you'll join us inside the course! Jose

Instructor

Jose Portilla, Pierian Training

Jose Portilla Head of Data Science at Pierian Training Jose Marcial Portilla has a BS and MS in Mechanical Engineering from Santa Clara University and years of experience as a professional instructor and trainer for Data Science, Machine Learning and Python Programming. He has publications and patents in various fields such as microfluidics, materials science, and data science. Over the course of his career he has developed a skill set in analyzing data and he hopes to use his experience in teaching and data science to help other people learn the power of programming, the ability to analyze data, and the skills needed to present the data in clear and beautiful visualizations. Currently he works as the Head of Data Science for Pierian Training and provides in-person data science and python programming training courses to employees working at top companies, including General Electric, Cigna, SalesForce, Starbucks, McKinsey and many more. Feel free to check out the website link to find out more information about training offerings.