CS279 - Reaction-Diffusion Simulation

About

A reaction diffusion simulator for a Stanford CS279 final project (Autumn 2019). Our implemented Simulation class allows user-defined simulations of Gray-Scott reaction-diffusion models as well as custom stochastic reaction-diffusion simulations based on reaction kinetics.

General Project Structure

├── run_simulation.py
├── requirements.txt
├── src
│   ├── VectorizedSimulation.py
│   ├── preset_simulations.py
│   └── config.py
└── simulations
    ├── example_1
    ├── example_2
    └── [...] <other simulation results>

Requirements

Example virtual environment setup (using virtualenv):

pip install virtualenv  # get virtualenv if you don't have it already
virtualenv venv  # create virtualenv named 'venv'
source venv/bin/activate # activate this virtual environment

Install dependency requirements using the following command:

pip install -r requirements.txt  # install the requisite files

Example Simulations

Running simulations using the run_simulation.py wrapper function. Pre-designed simulations are provided in src/preset_simulations.py and can be run using command-line arguments for run_simulation.py.

For example, running the base code with no additional parameters will run a two-particle Gray-Scott simulation with feed=0.0362 and kill=0.062.

python run_simulation.py

You can run another pre-set simulation feed=0.03 and kill=0.062 by running:

python run_simulation.py --type dots

By default, simulation results will be stored in ./simulations/test/.... To name the directory something else, simply provide the --run_name <name> tag.

In general, there are many user-modifiable parameters that can be accessed through command line flags like this one.

python run_simulation.py \
--run_name mySimulation \
--feed 0.035 \
--kill 0.062 \
--updates_per_frame 25 \
--iterations 1000

You can always run the --help flag to learn more.