-
Notifications
You must be signed in to change notification settings - Fork 1
polarcoconut/thesis-reactive-learning
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Re-active Learning ============== This repository contains code that can be used to reproduce results found in -Christopher H. Lin, Mausam, and Daniel S. Weld. "Re-active Learning: Active Learning with Relabeling." In Proceedings of the 30th AAAI Conference on Artificial Intelligence (AAAI). Phoenix, Arizona. February 2016. -Christopher H. Lin, Mausam, and Daniel S. Weld. "To Re(label), or Not To Re(label)." In Proceedings of the 2nd AAAI Conference on Human Computation and Crowdsourcing (HCOMP). Pittsburgh, PA, USA. Nov 2014. THIS CODE IS RESEARCH CODE and thus, horrible code. USE AT YOUR OWN RISK. ## Development configuration - Dependencies: see requirements.txt ## Run instructions - If you wish to use a non-synthetic dataset, first copy the desired dataset into the data folder. - Modify settings in `simulate.py` to pick the dataset and the re-active learning algorithms you want to use, as well as other settings like total budget, and worker accuracy. - Create a folder named `outputs` and inside this folder, create a folder whose name is what `folderName` is on line 441 of `simulate.py`. - Run `./runsim.sh` or `python simulate.py` - Results are output to the `outputs` folder. - If you want to assume knowledge of worker accuracy, then you must comment out lines 198 to 201 and uncomment lines 204-209 in samplingMethodClasses.py. ## Data Sets - We only provide the relation extraction data and the code to generate synthetic data. Other datasets can be found at the UCI Machine Learning Repository.
About
No description, website, or topics provided.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published