Skip to main content
Welcome to the prerequisites lesson. This short guide outlines what you need to start building, training, and hosting machine learning models using AWS SageMaker. We focus on lowering barriers for beginners, assuming minimal prior ML experience. You will learn with a code-first approach: write Python to prepare data and train models, and use AWS SageMaker to manage training, model artifacts, and hosted endpoints for inference. You don’t need to be an expert Python developer — familiarity with basic constructs is enough. Main objectives
  1. Lower the prerequisites for beginning your machine learning journey.
  2. Target beginners with little or no ML experience.
  3. Use AWS SageMaker as the platform for building, training, and hosting models.
A presentation slide titled "Agenda" listing three points: 01 Lowering the prerequisites for ML learning, 02 Targeting beginners in their ML journey, and 03 Focus on AWS SageMaker as the preferred ML platform. The layout has a dark left column with the title and numbered teal markers along the center.
Callout the approach and expectations:
This course uses Python to demonstrate ML workflows on AWS SageMaker. Basic Python knowledge (functions, lists, loops, conditionals, print) is sufficient. We introduce additional libraries and SageMaker utilities as needed.

​
Python basics we expect

To follow the hands-on exercises, you should be able to:
  • Define and call simple functions (using def).
  • Work with lists and basic variables.
  • Use iteration (for loops) to process collections.
  • Use conditional statements (if/else) to branch logic.
  • Print output to the console (print()).
Example: summing even numbers in a list 
# Define a list of numbers
numbers = [1, 2, 3, 4, 5]

# Define a variable to hold the sum of even numbers
sum_of_evens = 0

# Iterate over the list
for number in numbers:
    # Check if the number is even
    if number % 2 == 0:
        sum_of_evens += number  # Add even numbers to the sum

# Output the result to the console
print("The sum of even numbers is:", sum_of_evens)
Why this example is relevant:
  • number % 2 == 0 identifies even numbers.
  • The loop accumulates even numbers into sum_of_evens.
  • print(...) displays the result — a simple I/O pattern you’ll use when inspecting model outputs or debugging data transforms.
If you can read and understand the code above, you have sufficient Python to complete the course. We will introduce helper functions and third-party packages (NumPy, pandas, scikit-learn, and SageMaker SDK) as they become necessary.

​
AWS basics we expect

You should be comfortable with:
  • Having an AWS account and signing in to the AWS Management Console.
  • Uploading and downloading files (for example, CSVs) to/from an Amazon S3 bucket using the Console. This basic S3 interaction is sufficient to run the hands-on labs.
  • Being aware of other AWS compute and networking services (EC2, ECS, Route 53) is helpful for context but not required.
A slide titled "AWS Basics" with three numbered boxes. They list prerequisites: must have used the AWS Management Console; can upload/download files to an S3 bucket; and awareness of EC2, ECS, Route53 is helpful but not required.
If you haven’t used Amazon S3 before, practice by creating a bucket in the Console, uploading a small CSV, then downloading it back to confirm access. This hands-on step removes a common friction point when training models on SageMaker.

​
Machine learning awareness (high-level)

This course is designed for beginners. You don’t need deep ML theory up front, but it’s useful to understand the typical ML pipeline at a conceptual level:
  • Source data (e.g., tabular house-price dataset).
  • Select a candidate algorithm (we’ll demo linear learning; other choices include XGBoost, LightGBM, etc.).
  • Create and run a training job to let the algorithm learn patterns from the data.
  • Training produces a model artifact (e.g., model.tar.gz) that contains learned parameters.
  • Host the model on compute (a VM, container, or a managed service) to serve predictions (inference). In this course we’ll use SageMaker hosted endpoints for inference requests.
A flowchart titled "ML Awareness" showing a machine learning pipeline from source data through algorithm and training to a trained model. The trained model is then hosted for inference, producing a final prediction.
In short: ML is about teaching a model to generalize relationships between input features (bedrooms, bathrooms, square footage, area) and a target (house price). After training, the model predicts outcomes for unseen inputs.

​
Prerequisite checklist (quick reference)

CategoryWhat you should knowExample task
Python basicsFunctions, lists, loops, conditionals, printingRead and modify a small script that transforms CSV data
AWS & S3AWS account, Console, upload/download to S3Create an S3 bucket and upload a sample CSV
ML workflowHigh-level pipeline: data → training → model → hosting → inferenceRun a training job and call a hosted endpoint for prediction

​
What you’ll learn in this course

  • Choose an appropriate algorithm for a problem and data type.
  • Create and run a SageMaker training job.
  • Register, package, and host your trained model on a SageMaker endpoint.
  • Send inference requests to the hosted model and interpret predictions.
A presentation slide titled "Summary" with two points: (1) basic Python and ML concepts are sufficient to get started, and (2) familiarity with AWS and its services enhances the learning experience. The slide has a dark left panel and teal numbered markers for each point.

​
Summary

If you understand basic Python (lists, loops, conditionals, print) and the high-level ML workflow (data → training → model artifact → hosting → inference), you’re ready to begin. Familiarity with AWS and S3 will make the hands-on portions smoother, while knowledge of EC2, ECS, or Route 53 is optional background. This completes the short prerequisites lesson. In the next lesson we’ll cover key machine learning fundamentals to give you a clear view of what happens during training and inference.

Watch Video