Databricks – Scalable Deep Learning with TensorFlow and Apache Spark

What is the Scalable Deep Learning with TensorFlow and Apache Spark course about?

This course starts with the basics of the tf.keras API including defining model architectures, optimizers, and saving/loading models. You then implement more advanced concepts such as callbacks, regularization, TensorBoard, and activation functions.

After training your models, you will integrate the MLflow tracking API to reproduce and version your experiments. You will also apply model interpretability libraries such as LIME and SHAP to understand how the network generates predictions. You will also learn about various Convolutional Neural Networks (CNNs) architectures and use them as a basis for transfer learning to reduce model training time.

Substantial class time is spent on scaling your deep learning applications, from distributed inference with pandas UDFs to distributed hyperparameter search with Hyperopt to distributed model training with Horovod. This course is taught fully in Python.

Who is this course for?

• Data scientist
• Machine learning engineer

Prerequisites

• Intermediate experience with Python/pandas
• Familiarity with machine learning concepts
• Experience with PySpark

Skills You’ll Gain

• Build deep learning models using Keras/TensorFlow
• Scale the following:
  • Model inference with pandas UDFs & pandas function API
  • Hyperparameter tuning with HyperOpt
  • Training of distributed TensorFlow models with Horovod
• Track, version, and reproduce experiments using MLflow
• Apply model interpretability libraries to understand & visualize model predictions
• Use CNNs (convolutional neural networks) and perform transfer learning & data augmentation to improve model performance
• Deploy deep learning models

For more information about this course, please check this blog from P2L.