# Changelog

## Version 0.3.2 (2022-09-13)

* Updated the example experiment file
* Updated the system test case
* First pypi version

## Version 0.3.1 (2022-08-25)

* Changed syntax of magic command from `__req` to `$req`
  * Individual values (for agents and environments) can be selected with `"$req(phase=0, index=0)"`

## Version 0.3.0 (2022-08-24)

* Changed the formatting/syntax of the experiment block
  * Each environment defintion now consists of three parts:
    - environment
    - reward (optional)
    - state_transformer (optional)

## Version 0.2.3 (2022-06-09)

* Added state transformer as environment configuration parameter

## Version 0.2.2 (2022-01-18)

* Bugfix: More configs were not inherited correctly.
* Fixed the example experiment file. 

## Version 0.2.1 (2021-12-17)

* Bugfix: Configs in later phases were not inherited as expected.

## Version 0.2.0 (2021-11-16)

* Some API changes caused by the reward experiment.
* Added gitlab-ci

## Version 0.1.1 (2021-06-04)

* Added documentation for more of the functions.
* Refactored old generator test as test_designer
* Updated README

## Version 0.1.0 (2021-06-03)

* Updated arsenAI to be compatible with palaestrAI version 3.2.0.
* Included all the results from the experiment workshops done
  by the ARL developers.
  
## Version 0.0.4 (2020-11-26)

* Optimized sampling design 
* Renamed to arsenAI

## Version 0.0.3 (2020-09-24)

* Created generator classes
* Generator takes a cpsao and builds a couple of experiments
* Each experiment is stored in a separate yaml file.

## Version 0.0.2 (2020-09-23)

* Added description classes for abstract ontology and agent
* Added an example cpsao
* Extended CLI to read a cpsao file

## Version 0.0.1 (2020-09-23)

* Initialized repository
* Added projects skeletton