Research
TunedIT Research is an open platform for reproducible evaluation of machine learning and data mining algorithms.
Everyone may use TunedIT tools to launch reproducible experiments and share results with others.
Reproducibility is achieved through automation.
Datasets and algorithms, as well as experimental results,
are collected in central databases: Repository and Knowledge Base,
to enable comparison of wide range of algorithms, and to
facilitate dissemination of research findings and cooperation between researchers.
Everyone may access the contents of TunedIT and contribute new resources and results.
- Run automated tests of machine-learning & data-mining algorithms,
according to your test specification
- Log results in Knowledge Base for further analysis
- Run tests for any algorithms, datasets and evaluation procedures
available in Repository, uploaded by you or someone else
- Safely test untrusted code
- Test existing algorithms on your new datasets
- Test your new algorithms on existing datasets
- Reproduce, validate and extend results of tests done by others
- Write new evaluation procedures for new types of algorithms
learn more »
|
- Collect test results from TunedTester in a central database
on TunedIT server
- Merge your results with the ones submitted by others
- Easily browse them using web interface
- View in graphical or tabular form
- Download as CSV files for off-line analysis
- Go directly to Repository and view the tested resources
- Find the best algorithm for a given dataset
- Assess real performance of algorithms on many datasets
- Compare your algorithms with benchmark implementations
learn more »
|
- Upload new algorithms, datasets and evaluation procedures
to TunedIT server and test them with TunedTester
- Upload new versions of resources and don't worry about consistency
of test results in KB - out-dated results are automatically removed
- Share your implementations and datasets with others...
- ...or keep them (and related results) private
- View and download resources shared by others
- Easily interpret results collected in KB thanks to direct link
between them
- and corresponding resources in Repository
learn more »
|
