Seldon
TALK DETAILS:
Speaker: Alexandra Abbas, Data Scientist at Datatonic
Title: Going packaging free with ML on Google Cloud Platform
Talk details:
In their bid to become packaging-free, a global Cosmetics Retailer is developing a mobile app allowing customers to view product information simply by taking a picture. This completely eliminates the need for packaging and labels. However, in order to do this effectively, they needed an accurate Image Classification model available both on Android and iOS. To help them achieve their goal, Alexandra and her colleagues developed a mobile-friendly Image Classification model and an end-to-end, fully automated model training and serving pipeline orchestrated with Google Cloud Composer. This framework enables to re-train the model on Google AI Platform as new data land on Cloud Storage, monitor the new model performance in a BigQuery evaluation table and access the newly created TFLite model on a serving bucket. Alexandra explains how to productionize your Machine Learning pipeline using serverless and managed technologies and what challenges to count on when bringing your model to mobile, through an exciting real-life case study.
Bio: Alexandra is a Data Scientist at Datatonic working on large-scale innovation projects in Data Engineering and Machine Learning, she spends most of her time creating data pipelines using Big Data technologies like Apache Beam and Airflow and building production ready ML models using Tensorflow. Alexandra is a curious mind who enjoys problem-solving and exploring new technologies. She is an initiator of and passionate about the productionisation of Machine Learning models. Twitter @alexandraabbas