We cloud-scale your classifier!
EC for Big Learning Competition (GECCO 2014 Workshop)
Join us in fusing our EC classifiers
Saturday, May 10th, 12am to Monday, June 30th, 12am
From the comfort of home
Have you ever wanted to run your EC algorithm in the cloud? Discouraged by the complexity of EC2? We will deploy your EC algorithm on the cloud for you with our FCUBE framework!
FCUBE supports a Bring Your Own Learner (BYOL) model: it deploys your EC algorithm to hundreds of machines and does all the data management for you. No scripts, no launch hassles, no tedious result collection. FCUBE is (EC) deployment as a service.
For this competition, our goal is to unite the developers of interesting EC classifier algorithms. We seek an experienced informed discussion on the various approaches and techniques without being distracted by one problem at hand. Therefore, we have set up the following format:
- Everyone gets the same computational budget in Amazon EC2
- Everyone works on the same datasets
- Organizers select the features
- You contribute a classifier learning algorithm with your own fitness function, operators and search logic
- You contribute a classifier learning algorithm which accepts training data in csv format and references a Java properties file which you provide (details below), and outputs a classifier.
- You contribute a classifier learning algorithm in executable format (Java, python) or as source code (must be compilable in Linux: C, C++ etc)
- You contribute a piece of code which applies your classifier to test data and produces labels.
- FCUBE executes your algorithm with the competition training data
- FCUBE retrieves the solutions from the cloud nodes, computes the testing predictions, and returns them to you.
- FCUBE also filters and fuses the predictions using different methods. Everyone receives their fused results and everyone contributes to a collaborative fused solution among all contributors.
For more detailed information check the workshop website:
or contact us:
Learner Training interface
- a path to a CSV file
- learning time deadline
- Properties File with Java syntax (for your extra parameters)
- a model stored in a single file on disk
Try an example here:
Learner Predict interface
- path to a CSV file
- path to where the model is stored
- path to where the predictions will be stored
- predictions in CSV file
Try an example here: