DISCLAIMER: This project is under development and might not be fully functional.
autodoe helps you generate, process and analyze experimental plans based on Design of Experiments.
It generates experimental plans, maps the plan to your given factor values, creates a well ordered folder structure, processes given input scripts and analyzes the experiments output. That might help you when you want to perform parameter analyses or create statistical significant data for machine learning based on an experiment or a simulation.
Before you begin, ensure you have met the following requirements:
- You have a Unix machine (Mac/Linux). Currently only tested with Ubuntu20.04
- You have python3.8 or newer and pip is installed
- You have a command line interface or script based experiment / simulation
You can clone the git repo and run the program in a virtual environment:
git clone https://github.com/csRon/autodoe.git
cd autodoe
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
Test you installation by running the first example helloDoE
by running
python3 autodoe.py ./examples/helloDoE
You should see some output of simulations running.
It is recommended to take a look at the given examples and its readme's and the documentation for more detailed explanations. Some theoretical knowledge of Design of Experiments would also be advantageous.
Nevertheless the basic setup of autodoe remains the same.
A working folder must be created with the following content (in some exceptional cases input can be different):
config.conf
: file contains your settings for the DoE Analysisfactors.csv
: file contains the factors you want to analyze with min and max valuestemplates
: folder contains all necessary input for your experiment of simulation
For the beginning just copy an existing exmample folder (e.g. helloDoE) to your preferred location and modify the files based on your needs.
Just run the following command directly inside the autodoe folder
python3 autodoe.py /path/to/your/working/folder
All output will also be in your selected working folder.
All output will be in your selected working folder. The results will be 2 files + 2 folders:
factorPlan.csv
: file contains the raw experimental plan with values between min and max of the factor valuerawPlan.csv
: file contains the raw experimental plan with values between 0 and 1simulationFolder
: folder contains raw experiment / simulation results for each run of your experimental planresults
: folder contains the combined results of all experiment / simulation runs as csv-file.
The experimental plan creation is based on pyDOE2
which is based on pyDOE
created by:
- Copyright (c) 2014, Abraham D. Lee (
pyDOE
) - Copyright (c) 2018, Rickard Sjögren & Daniel Svensson (
pyDOE2
)
autodoe is licensed under the MIT license.