Merge pull request #7 from AlgTUDelft/compatibility-update

Updated version to 0.0.3

README.md

@@ -8,19 +8,9 @@ _Deep Reinforcement Learning for Active Wake Control_ (to be published at AAMAS'
 
 ### Replicating the Experiments
 
-Because each experiment has many parameters, we use configuration files to  describe each experiment. The configuration
-files for the experiments presented in the paper can be found in `./code/configs`: `action_representations_*.yml` for
-the action representation experiment (Section 4.1), and `noisy_*.yml` for the noisy observation experiment
-(Section 4.2). You can also create new configuration files to run your own experiments.
+The current version of the repository contains only the environment itself, so that users do not have to download the experiments presented in the paper if they do not need them.
 
-Make sure that you are using Python 3.8 or newer. You also need to install required packages by running
-`python3 -m pip install -r requirements.txt` from the project's folder.
-
-To replicate the experiments in the paper, run `python3 main.py --config CONFIG_FILE` from the `code` directory,
-where `CONFIG_FILE` is a path to a configuration file. The output data is written in Tensorboard
-format to `<current-path>/<directory>/<name>/...`, where `<current-path>` is the path the script was executed from and
-`<directory>` and `<name>` correspond to parameters specified in the configuration file. You can use
-`./code/tensorboard_to_csv.py --path <PATH>` to convert the output from `<PATH>` to a `.csv`-file.
+If you are interested in the scripts and configurations for the experiments or the supplementary material for the original paper, please check the release 0.0.1 of this repository for the code and the instructions.
 
 ### Building from the Source Code
 
@@ -34,8 +24,8 @@ Next, build the package by running `python3 -m build` from the project's directo
 This will create `./dist` with 
 a `wind_farm_gym-<VERSION>-py3-none-any.whl` file in it, where `<VERSION>` is the current release version.
 
-Finally, install the package by running `python3 -m pip install wind_farm_gym-<VERSION>-py3-none-any.whl`, or
-`python3 -m pip install wind_farm_gym-<VERSION>-py3-none-any.whl --force-reinstall` if you want to reinstall an existing
+Finally, install the package by running `python3 -m pip install dist/wind_farm_gym-<VERSION>-py3-none-any.whl`, or
+`python3 -m pip install dist/wind_farm_gym-<VERSION>-py3-none-any.whl --force-reinstall` if you want to reinstall an existing
 installation. This should install the package and its dependencies.
 
 To test that the package is available and working, you can run the  following example, also  available at
@@ -58,21 +48,6 @@ env.close()
 
 ![A simulation of an overhead view of a wind farm](build-test.png)
 
-## Contents
-
-- `./code` contains the source code:
-    - `agent` for implementations of RL agents;
-    - `config` for configuration files used in the experiments;
-    - `wind_farm_gym` for the environment.
-- `./data` contains the datasets used in the paper:
-    - `wind_data` contains the measurements from Hollandse Kust Noord (site B) used to estimate the transition model in
-Section 3.4.
-    - the other subdirectories contain the results of the experiments:
-        - `action_representations_tunnel` for the experiment presented in Section 4.1.
-        - `noisy_1` ... `noisy_7` for the experiment presented in Section 4.2; each folder corresponds to a different
-level of noise, 1% ... 7%.
-- `supplement.pdf` is the supplementary material submitted with the paper.
-
 ## Citation
 
 Please, cite the paper if you use it:

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = wind_farm_gym
-version = 0.0.2
+version = 0.0.3
 author = Grigory (Greg) Neustroev
 author_email = [email protected]
 description = An OpenAI Gym environment for active wake control via reinforcement learning
@@ -18,14 +18,12 @@ classifiers =
 
 [options]
 install_requires =
-    pyglet
-    gym
+    pyglet==1.5.21
+    gymnasium==0.29.1
     matplotlib
     numpy
     scipy
-    FLORIS>=2.4
-package_dir =
-    = code
+    FLORIS==2.4
 packages =
     wind_farm_gym
     wind_farm_gym.wind_process