This is a self-contained laboratory session of a challenge to create a model achieving highest score on OpenAI's CartPole environment.
These Python based notebooks are designed to work inside Google's free research and education tool Colaboratory, which requires only a Google account. Check out their FAQ.
This is designed to be part of a module, directly after https://github.com/KiranArun/Reinforcement_Learning-101-demo as a challenge against other people.
- The Starter notebook already has a working keras-rl model with low performance
- When doing this challenge as one of my modules, you will give me the notebook and I will run it to test performance
1. Download or clone the repository
- if you downloaded the ZIP, extract on your local machine
2. Go to google drive, and upload this folder from your local machine
3. From Drive, open the notebook with Colaboratory (double-click then choose Connected apps Colaboratory)
- If Colaboratory is not shown, you'll have to first add it from Open With, then search Colab, then connect.
- https://colab.research.google.com
4. Select runtime, change runtime type, and set hardware accelerator to GPU
- if it doesn't let you, that's fine (it'll just be a bit slower)
- if are using the GPU, it may run out of memory so you'll have to use CPU
Contributions are welcome, I particularly appreciate corrections from PR's or raised through Issues. Please make an individual PR for each suggestion.
Stack Overflow would be the best place for help with using the frameworks.
Licence: Apache 2.0. © 2018 Kiran Arun