Objectives

Students will be able to:

  • uderstand that different Machine Learning scenarios require different hardware infrastructure.

Machine Learning

Theme A4

Hardware Considerations

Different Machine Learning scenarios will require different hardware.

For example, a model which predicts global weather patterns will require a completely different set of hardware to a model which predicts what meals visitors might order in a restaurant.

Here is a table of hardware commonly used in a variaty of ML scenarios:

Hardware Description
Standard Laptop/PC Basic CPU setup, suitable for small local experiments and learning
Edge Device Small, low-power device (e.g. car controller, sensor unit) running AI locally in real time.
GPU Excellent at processing large batches of images and parallel data. Often used for deep learning.
FPGA Can be customized for specific machine learning tasks in hardware; flexible and efficient for industrial use.
Google TPU A special type of processor designed by Google specifically for neural network operations.
ASIC A custom-built chip designed for one specific purpose — not general computing. A TPU is an example of an ASIC.
HPC Center A physical facility with many powerful computers connected together.
Cloud-based Platform A virtual service offered over the internet (e.g. Google Cloud, AWS, Azure).

Task

Make a copy of the worksheet. Move it into your Theme A4 folder.

Follow the instructions and complete the tasks.

Journal Task

Make a new entry in the Journal section of the booklet.

Did you discover anything new today? Do you have questions in your mind that you want answers to?

Make these notes in your journal.

Glossary

Machine Learning

Supervised Learning

Unsupervised Learning

Labels

Training

Training data

Test data

Features

Tags

Machine Learninglabels Supervised learning Unsupervised learning features model training data trainingtest data