https://github.com/cair/polyacoplus
https://github.com/cair/polyacoplus
Last synced: 9 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/cair/polyacoplus
- Owner: cair
- Created: 2017-11-01T08:01:12.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2017-11-01T08:03:05.000Z (over 8 years ago)
- Last Synced: 2025-02-04T21:45:10.884Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 17.1 MB
- Stars: 0
- Watchers: 4
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# PolyACO+
An algorithm using Ant Colony Optimization with polygons and ray casting for classification.
## Installation
The easiest way to install the project is by using [Conda](http://conda.pydata.org/). Conda is a Python package manager and environment manager that makes it easy to set up and install Python environments. Follow the instructions on [Condas website](http://conda.pydata.org/docs/install/quick.html) to get started.
After installing Conda, `cd` into the project directory and install the project environment with:
```bash
$ conda env create -f environment.yml python=3.5
```
This will install a full conda environment named *acoc*. Activate the enviroment by running:
```
$ source activate acoc
```
## Configuration
The project looks for a configuration file called *config.py* in the root folder. This file is ignored by the VCS so that you can make changes without affecting the repository. To get started with a config file simply copy *config_template.py* and rename it to *config.py*. Then you can freely change the contents of the config file without affecting the VCS.
### CUDA
The algorithm is optimized for Nvidia GPUs and depends on CUDA. To run the project without CUDA, set `'gpu': false` in *config.py*.
## Usage
A complete classification example can be found in *demo.py* in the root folder of the repository. The demo use the classifier configuration specified in *config.py*.