A selection of AI systems built in collaboration with artists for exhibits and performances

Get the repo: git clone https://github.com/seraphinatarrant/ai_as_art.git then cd ai_as_art

For each different project, make an environment with: conda create -n MYENVNAME python=3.7

Once it is created, activate it to install everything: conda activate MYENVNAME

Then navigate to the directory for the project (each will have different requirements), and do: pip install -r requirements.txt

To deactivate the environment: conda deactivate

directory: image_generation/. (Under the hood, this is a DC GAN, forked from pytorch examples/.)

1. Use an existing config in the config dir, or make a new one.
2. Activate the appropriate environment.
3. python generate.py -c config/MYCONFIG

directory: text_generation/ TBD adding things here