Course Wiki

TensorFlow

Google's deep learning framework and its Keras interface — how ML Zoomcamp trains and serves image models.

Machine Learning Zoomcamp teaches deep learning in both major frameworks, with TensorFlow and Keras as the first path: Keras’s Sequential API builds the clothing-classification model layer by layer, and the same network is reimplemented in PyTorch to show what each framework abstracts.

TensorFlow also owns the serving side of the course: TensorFlow Serving is the model server deployed on Kubernetes in module 10, and the serverless module packages a TensorFlow model into an AWS Lambda container, including the size workarounds its dependencies force.

Taught in

• Machine Learning Zoomcamp

  • Module 1: Introduction to Machine Learning
    • Setting up the Environment
  • Module 8: Neural Networks and Deep Learning
    • Fashion classification
    • TensorFlow and Keras
    • Transfer learning
    • Checkpointing
    • Explore more
    • Installation of tensorflow
  • Module 9: Serverless Deep Learning
    • Introduction to Serverless
    • TensorFlow Lite
    • Explore more
    • Python 3.12 vs TF Lite 2.17
  • Module 10: Kubernetes and TensorFlow Serving
    • Overview
    • TensorFlow Serving
    • Creating a pre-processing service
    • Deploying TensorFlow models to Kubernetes ## Related concepts

• PyTorch

• Neural Networks

• Transfer Learning

• Kubernetes

• Serverless Deployment