Serverless + Machine Learning – Bringing the best of two worlds together

While building and training Machine Learning models, many of us face this question of "Which of the algorithm or library fits well with my model?". To answer this you will create multiple estimators and then come to a conclusion. How about finding the right FIT for your data in one go? and also save the compute as well with Serverless Cloud Functions.

This interactive session walks you through the process of building a predictive machine learning model, deploying it as an API to be used in applications, testing the model. All of this happening in an integrated and unified self-service experience on IBM Cloud. Also, You will create multiple python serverless actions using different ML libraries and algorithms which will be invoked in one go, saving the results for you to chose the BEST FIT.

In short, Machine Learning models exposed as APIs with serverless python action acting as a middleware to invoke the action.Thus bringing the best of Containers through Serverless and Machine learning together.

Transcript

1. Bringing the best of Containers and Machine Learning together Vidyasagar Machupalli, Technical Product Manager, IBM http://vmac.xyz | @VidyasagarMSC
2. 2 http://bit.ly/2Esjpqy Demo
4. Cloud Functions 4
5. Functions as a Service (FaaS) platform to execute code in response to events 5
6. Write your code in the language you like the most 6
7. Expose your actions as API endpoints Cloud Functions Cloud Functions 7
8. Source: https://medium.com/openwhisk/squeezing-the-milliseconds-how-to- make-serverless-platforms-blazing-fast-aea0e9951bd0 Best usage of Containers
9. 9 Future???
13. Let’s see in action… 13
14. The dataset 14
15. What you just saw 15 Input Trigger Sequence https://bit.ly/2qGp57o WATSON STUDIO CLOUDANT NOSQL DB CLOUDANT NOSQL DB CLOUD FUNCTIONS CLOUD FUNCTIONS
16. https://developer.ibm.com/code/ Signup for IBM Cloud https://bluemix.net http://ibm.biz/solution-tutorials Solution Tutorials
17. Thank you