# AutoArtGPU
AutoArt that uses OpenCL to run on your GPU

This is the GPU version of [AutoArt](https://github.com/pommicket/AutoArt).

For a live version of AutoVideos that creates them as they play, go to https://github.com/pommicket/AutoVideosLive.

**AutoArtGPU requires the following dependencies:**

AutoImages:
- Python
- PIL/Pillow
- Numpy
- PyOpenCL

AutoVideos:
- Python
- OpenCV
- Numpy
- PyOpenCL

AutoAudio:
- Python
- Numpy
- PyOpenCL

On Ubuntu/Debian you can get them using:  
```
sudo apt-get install python python-pil python-numpy python-pyopencl python-opencv
```

On other platforms, you can use [pip](https://pip.pypa.io/en/stable/), or visit the libraries' websites.

To run AutoArt:
```bash
cd /directory/where/AutoArt/is/
```

To run AutoImages:  
`python AutoImagesGPU2.py`  
To run AutoVideos:  
`python AutoVideosGPU2.py`  
To run AutoAudio:  
`python AutoAudioGPU.py`

Don't worry if when you run it, it gives the following message:  
```
/usr/lib/python2.7/dist-packages/pyopencl/__init__.py:61: 
CompilerWarning: Non-empty compiler output encountered. 
Set the environment variable PYOPENCL_COMPILER_OUTPUT=1 to see more. 
"to see more.", CompilerWarning)
```

Also don't worry if AutoArt closes at, say, 10/15, prints `Segmentation Fault`, and re-opens saying 0/5, all 15 images will be made.

You should not create images/videos with a resolution of over 3000x3000, or audio files that are over 10 minutes in length (but this depends on the amount of memory your GPU has). If you create an image/video that is too large or an audio file that is too long, you will see the following error message:

`LogicError: create_buffer failed: invalid buffer size`