That session explains how to put AI algorithms on Edge devices, with feedbacks from a computer vision project
When buzzwords like DevOps and Machine Learning collide, you need a demo and talk that shows how to actually increase the delivery speed of Advanced Analytics solutions.
The most challenging area of machine learning are Data acquisition, Feature extraction, Feature Selection. Almost in all data science project, 80% of time people spend in Data acquisition and Feature engineering.
How do you deploy multiple machine learning models in production to solve your challenge? How do enable canary releases and A/B testing? How do you make sure a user is always served by the same version of the model?
Business Intelligence projects, and nowadays Machine Learning projects too, still keep failing at an alarmingly high rate. Making them succeed is easy enough tho, I'll work through a case with you.
Learn how Tailwind Traders data science team uses Azure Machine Learning features and services to create bespoke open source NLP models and optimise them. Includes Automated ML, Azure ML SDK and Hyperparameter tuning