We at genular are happy to announce the release of our one of a kind
Machine Learning Software – SIMON.
In line with our mission of bringing reusable, user-friendly, executable and reproducible machine learning to the community, we have designed and developed SIMON as an open source automated machine learning software. All thanks to our vibrant genular community.
Before we dig deep into SIMON, we will like to briefly introduce SIMON to you. SIMON is a powerful, flexible, open-source and very easy to use software that implements machine learning (autoML) and statistical data discovery features for the discovery of dynamic relationships inherent within a given data.
With few clicks within a soothing graphical user interface, scientists, makers and a whole of users can make use of SIMON to extract structural and meaningful information from their data and to build machine learning models.
And now to the cool stuffs, with SIMON providing:
- Nicely designed drag&drop user interface for applying data modeling techniques,
- Over 200+ machine learning algorithms to choose from,
- Easy importation of public dataset repository,
- In-built data preprocessing (correlation filtering, normalization, standardization etc.),
- Support for high sparsity data via mulset or imputation,
- Cloud and local backend data storage,
- Clustering and correlation graphs via visual data analysis,
- 280 visual styles via visual feature analysis with dot-plots,
- Multiple language support
- Integrated SAM (Significance Analysis of Micro-arrays) technique, and
- Exportation of developed model and associated data
Take a look at SIMON, Figure 1 depict a general overview of our SIMON knowledge discovery application software. Here you can select desired processing algorithms as you can see under: Available Packages, Selected Packages as well as the Dashboard and Workspace Tabs.
A look at SIMON’s dashboard presented in Figure 2 revealed the nicely designed GUI with different Tabs having distinct symbols and provided summative information. As seen, six tasks were submitted and a complete breakdown of each tasks were provided. In the breakdown, we can see task details like: ID, Name, Created, Processing time, Sparsity, Successful models and many other information relevant to the specific data-set processing.
SIMON also have a workspace, as depicted in Figure 3, where data importation and wrangling is being carried out. Here you can also import publicly available data-sets for them description and other information is made available as well as preview of the data.
Concisely, with SIMON you as a user will easily and automatically discover features, perform exploratory data analysis with instant insights and visuals and make use of automated machine learning in your project. More so, your results and models you build are available for sharing and also downloadable for external usage.
At genular, privacy and security is part of priority. That’s why we ensured that you can host and run SIMON on your own dedicated servers or laptop. IN that way you will not need to worry about someone else stealing or looking at your data and or models.
The easiest way to install SIMON is to follow directions provided here in documentation. In case of any issues/questions or just a small talk please join us on community forms or IRC #genular on freenode.net.
Looking forward to your migration and becoming member of our vibrant community as you make use of SIMON for your machine learning project and contribute toward the betterment of this community. If you like SIMON please be sure to leave us a star at our GitHub repository.