Explore a compelling conference talk that delves into the intricacies of constructing scalable deep learning pipelines for extensive model training on massive time-series datasets. Learn how this innovative approach resulted in an impressive 80% cost reduction. Gain valuable insights from Benedikt Koller, CTO of maiot GmbH, as he shares his expertise on optimizing machine learning processes and achieving significant cost savings in the field of artificial intelligence and data science.
A Tale of A Thousand Pipelines: Building Scalable Deep Learning Pipelines for Time-Series Datasets
Toronto Machine Learning Series (TMLS) via YouTube
Overview
Syllabus
A Tale of A Thousand Pipelines
Taught by
Toronto Machine Learning Series (TMLS)
Related Courses
-
Detecting Data Anomalies using Deep Learning Techniques with TensorFlow 2.4
-
Recurrent Neural Networks (RNNs): Deep Learning for Sequences and Time Series
-
Scalable Data Pipelines for ML - Integrating Argo Workflows and dbt
-
Building Scalable End-to-End Deep Learning Pipelines in the Cloud
-
Writing Machine Learning Pipelines Against Object Storage
-
Building Scalable and Reliable Data Pipelines Using Debezium and Kafka
