Python Reinforcement Learning Tutorial for Beginners in 25 Minutes

preview_player
Показать описание
Want to break into Reinforcement Learning with Python?

Just not too sure where or how to start?

Well in this video you’ll learn the basics of creating an OpenAI gym environment in Python and training a reinforcement learning algorithm to solve the Lunar Lander problem. You’ll be able to leverage the stable_baselines algorithms to quickly and effectively train a deep reinforcement learning model in Python, the same pattern can be used over and again to train and solve multiple reinforcement learning problems.

In this video, you'll learn :
1. Installing Stable Baselines for Reinforcement Learning with Python
2. Training a Reinforcement Learning model using the ACER Algorithm
3. Running and Evaluating a Stable Baselines RL Model on LunarLander-v2

Chapters:
0:00 - Start
1:55 - Reinforcement Learning Flow
4:09 - Installing Python Dependencies
6:10 - Importing RL Dependencies including stable_baselines
9:11 - Testing the LunarLander-v2 Environment
11:47 - Training an ACER Reinforcement Learning Model
17:52 - Evaluating the Model
20:30 - Saving and Reloading RL Model Weights

Oh, and don't forget to connect with me!

Happy coding!
Nick

P.s. Let me know how you go and drop a comment if you need a hand!
Рекомендации по теме
Комментарии
Автор

Great tutorial once again! installing dependencies can be a major hassle specially when you're new (to everything). This guy explains everything step by step and makes it easy.

haiderjan
Автор

Great tutorial again! Worth mentioning I had to create a new conda environment with python=3.7 to be able to install tensorflow 1.15.0, and conda install swig to get box2d to install.

anthonylwalker
Автор

Why don't I found you early. You are one of the best RL Tutor. Thank a Lot.

wiwatpitakworarat
Автор

I really liked your python tutorial. I hope that your channel continues to grow. It's amazing to see that we have similar passions for teaching others Python on YouTube

JonathanKarr
Автор

I would be very interested if you would make a video and explain all the different RL algorithms (A2C, DQN, PPO etc etc) and their use cases. I have to admit, I have no idea at all what these RL algorithms involve, how they work and when it is best to use which RL algorithm.

franky
Автор

You have the best python reinforcement learning channel, very through

bmack
Автор

Dear Nicolas, you are the MAN. Thanks for the cool vids!

theartistformallyknownas
Автор

As usual, excellent and very informative tutorial.
Thank you Nicholas, You have the best channel in YouTube when it comes to RL and ML tutorials.
Can not wait for your next video.

OmarAlolayan
Автор

WONDERFUL. Superb tutorial, thank you. Clear explanations to each line of code, allowing me to follow along. I am also able to take what I learned and apply it to different environments. I have developed an even greater interest in the subject matter. BTW, I learned of this tutorial from word-of-mouth. I will spread the word. Please make more and again, thank you!

marlonscott
Автор

I cannot wait to try it out by myself 👍

feifeizhang
Автор

How might we infuse more variability into the foundational problem to assess the model’s robustness under different conditions?

sanchitbhardwaj
Автор

can you remake this video? the libraries are out of date.

rverm
Автор

I use Windows and I am having problems with this example, so I found this remark:

"To download an older version of TensorFlow make sure you are using an older version of python as well. Otherwise, you will run into an issue like no version satisfying requirement found.

Create a virtual environment for this and install python==3..5
Use pip install tensorflow==1.4 or so. "

Is It really necessary?

gplgomes
Автор

thanks for making this video! ill have a go this weekend

tommclean
Автор

I still don’t understand how the algorithm is able to move the spaceship? Where are the bindings with the position controls of the spaceship? Also, is there a reward scheme for landing between the flags?I mean, how does the algo decide that this is the correct way to land?😳

fxsurgeon
Автор

you need to point out in description,
1. that this is only works with python 3.7 because tensorflow
2. also, installing box2d-py not always work

borneoland-hkil
Автор

Hey Nick, do you have any video on an agent that copies another agent? Because I'd love one! One reinforcement learning agent does something, and another one tries to do what the first agent did. Basically, the second agent's got to do its best to copy the first agent. I love your videos!

coded
Автор

I have created a dataset and i want to work on that using the RL model with three states ..can you please make a video to explain working of RL model based on the dataset that was provided

infinite_interestzz
Автор

Hey,
Nicholas Renotte!
Thanks for sharing your knowledge.
I request you to please make a video on creating custom environment for training RL agent using stable baseline

akshaymewada
Автор

Why cant we use tensorflow 2.x.x
BTW love your videos

prasaddash