deap_genetic_programming

Requirements

• Python
• Pip

Usage

1. Install assignment packages

$ pip install -e packages/symreg packages/riceclf

2. Run the symbolic regression experiments

$ symreg

• For usage help:

$ symreg --help

3. Run the rice classification experiment.

• We can run the experiment using the following commands:

$ riceclf experiment -o tmp --seed 123 456 789 2974 2479 24755 74593 57993 24749 279

• Now give it some time for each (alot of time, go get a coffee or something... and come back to the program), to allow it to process until its all done.
• Then to plot the results, run the following command:

$ riceclf plot tmp/result.csv tmp/result.png

• Now check the tmp folder for the results in the root directory of the project.

4. For an individual run, create a folder riceclf_output using the following command:

$ mkdir riceclf_output

• Then run the following command to run the experiment and save the results to a file riceclf_output/run123.csv:

$ riceclf run -o riceclf_output/run123.csv --seed 123

• Then to plot the results, run the following command:

$ riceclf plot riceclf_output/run123.csv riceclf_output/run123.png

5. For usage help:

$ riceclf --help
$ riceclf run --help
$ riceclf experiment --help
$ riceclf plot --help

References:

• F.-A. Fortin, F.-M. De Rainville, M.-A. Gardner, M. Parizeau, and C. Gagné, “DEAP: Evolutionary algorithms made easy,” Journal of Machine Learning Research, vol. 13, pp. 2171–2175, jul 2012. https://deap.readthedocs.io/en/master/
• Cinar, I. and Koklu, M. (2019). Classification of Rice Varieties Using Artificial Intelligence Methods. International Journal of Intelligent Systems and Applications in Engineering, vol.7, no.3 (Sep. 2019), pp.188-194. https://doi.org/10.18201/ijisae.2019355381.