All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog and this project adheres to Semantic Versioning.
- Separated content into
basic_tutorial_stepsandadvanced_tutorial_steps - Introduce reference implementation of a learning agent with cogment enterprise in
cogment-enterprise-tutorials
- Modified content of steps 1-4 of previous tutorials to present concepts in a different order.
run.shprocesses changed- install Cogment in virtual environment to reflect more up-to-date method for Cogment application implementation
- changed
buildtoinstall
- Upgrade Cogment js SDK to
2.0.8
- Implementing a learning agent with DQN -- does not use an up-to-date Cogment training loop.
- Add a step 8 showcasing the directory.
- Leveraging
cogment launch - Upgrade to React 18
- Use python built-in
venvinstead ofvirtualenv - Upgrade to cogment python sdk 3.3.0
- Add
./run.shscripts to start the services without relying on Docker.
- Show the ability to provide
controller.start_trialwithtrial_params. - Requires cogment v2.2 and cogment-py-sdk v2.1.1
- Updated to API v2.0
- Update to Cogment Orchestrator v1.0.3, Python SDK v1.2.0 and Javascript SDK v1.25.2.
- Update copyright notice to use the legal name of AI Redefined Inc.
- Naming changes to the DQN player tutorial to follow more closely convention of the algorithm.
- New tutorial showing how to train an actor using reinforcement learning techniques
- Update to the latest versions of the cogment modules
- Update to the latest versions of the cogment modules
- Initial public release.