Hi all,
I (Shashank Agarwal) am the Founder & CEO of Hopdata.
We have been working hard to create a simplified machine learning as a service platform. We created Hopdata as we realized most of the platforms out there were either complex, expensive, allowed fewer attributes or had lower accuracy.
We wanted to create a super simple ML service.To create a model, you only need to upload a csv file. Each dataset is analyzed, cleaned and trained on over five different algorithms. The algorithms with the highest accuracy is finally selected.
We are still very new and would love to get some feedback.
Please let us know what you think.
-Shashank Agarwal
What are the algorithms are you using?
Are you planning on exposing an API, where clients can stream training data?
This will help them create feedback loops in their apps.
@shehaaz
We are using six algorithms as of now which include Three variants of Naive Bayes , SVM, Random Forests and Logistic Regression. We plan to support more in the future. Also configs for each algorithm are optimized based on the data provided.
@shehaaz
For the second querstion:
We already have a realtime API for categorizing live data.
You can see example code on our homepage and read more about how to get started here ->
http://docs.hopdata.com/docs/get...
