https://github.com/achoum/exploratron
Exploratron is a open-source turn-by-turn sandbox puzzle game with multiple simple mechanics that interact with each other and create complex and interesting emergent gameplay situations.
https://github.com/achoum/exploratron
genetic-algorithm machine-learning puzzle roguelike sandbox simulation
Last synced: about 2 months ago
JSON representation
Exploratron is a open-source turn-by-turn sandbox puzzle game with multiple simple mechanics that interact with each other and create complex and interesting emergent gameplay situations.
- Host: GitHub
- URL: https://github.com/achoum/exploratron
- Owner: achoum
- Created: 2022-06-19T19:53:02.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2022-12-26T10:01:45.000Z (over 2 years ago)
- Last Synced: 2024-10-07T05:41:05.016Z (7 months ago)
- Topics: genetic-algorithm, machine-learning, puzzle, roguelike, sandbox, simulation
- Language: C++
- Homepage: https://achoum.github.io/exploratron
- Size: 5.46 MB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
Awesome Lists containing this project
README
# Exploratron
**Exploration** is a open-source turn-by-turn sandbox puzzle game with multiple simple mechanics that interact with each other and create complex and interesting emergent gameplay situations.
Exploration is compatible with Linux, Windows and Web browser (using Webassembly).
You can play the Web browser version [here](https://achoum.github.io/exploratron).
## Compilation from source
Requires _Bazel_.
On Web browser:
```shell
./tools/game_wasm.sh
```
On Linux:
```shell
./tools/game_terminal.sh
```
On Windows:
```shell
tools\game_terminal.bat
```
## Map creation
Maps are created using [Tiled](https://www.mapeditor.org/download.html). See examples in the `exploratron\assets\map` directory.
## Agent training
Exploratron is also a framework to train and evaluate agent using Machine Learning.
Read, modify and and run `./tools/train_agent.sh` to train and evaluate agents.
Some examples of commands (see `./tools/train_agent.sh` for details):
```shell
LOGDIR=$(pwd)/training_logs
FLAG="--config=linux --define=terminal=linux_console"
# MODE="-c dbg"
MODE="-c opt"
# Run the evaluation on the "Gather" area using the Random controler (no leaning).
bazel run $MODE --config=linux --define=terminal=linux_console //exploratron/cli:evaluate_main $FLAG -- \
--arena=Gather --controller_key=Random --num_repetitions=100
# Train and evaluate a genetic controller on the "Gather" area.
# Hyper-parameters are available in `exploratron/controller/genetic/genetic.h`
bazel run $MODE --config=linux //exploratron/controller/genetic:train_main $FLAG -- \
--training_log_base=${LOGDIR}/gather/genetic/r1_
# Train and evaluate a hill climbing controller on the "Gather" area.
# Hyper-parameters are available in `exploratron/controller/hill_climb/hill_climb.h`
bazel run $MODE --config=linux //exploratron/controller/hill_climbing:train_main $FLAG -- \
--training_log_base=${LOGDIR}/gather/hill_climb/r1_
# Run the evaluation on the "Gather" area using the keyboard controler (no leaning).
bazel run $MODE --config=linux //exploratron/cli:evaluate_main $FLAG -- \
--arena=Gather --controller_key=Keyboard
```
Results can be compared using the R scipe `./tools/plot_training_logs.R`.