Scaling Deep Learning with TensorFlow and Apache Spark

Role: ML Engineer, Data Scientist

Duration: Half-day

Labs: Yes


This course covers how to scale your deep learning models with Apache Spark, including distributed model training, hyperparameter tuning, and inference. Taught entirely in Python, throughout this course students will be guided to:

  •  Build deep learning models with TensorFlow
  •  Perform distributed inference with Spark UDFs via MLflow, as well as distributed hyperparameter tuning with HyperOpt
  • Train a distributed model across a cluster using Horovod


  • Experience programming in Python and PySpark
  • Basic understanding of Machine Learning concepts
  • Prior experience with Keras/TensorFlow highly encouraged