Jun 17, 2020

Passing the AWS Machine Learning Speciality Exam

The Amazon Machine Learning Specialty Exam is a 3-hour, 65 question test. It is designed to test your skills in AWS specific Data Engineering and Machine Learning Practices along with Machine Learning in general. I hope to provide a collection of resources for individuals to not only pass the Amazon Machine Learning Speciality Exam but, be able to effectively implement, architect, design, and maintain machine learning solutions in the AWS cloud environment.

The Machine Learning Speciality Exam is comprised of 4 different domains.

  • 20% – Data Engineering
  • 24% – Exploratory Data Analysis
  • 36% – Modeling
  • 20% – Machine Learning Implementation and Operations

There are quite a few resources out there to help you study for this exam. Amazon Training also provides great content to help teach you about their machine learning and AI services. I highly recommend watching their videos as well. The best part of Amazon’s Training videos is that they are free! Aside from Amazon’s training the following is what I used to help me prepare for the Amazon Machine Learning Speciality Exam.

Disclaimer

I practice machine learning and deep learning on personal projects and machine learning competitions almost on a daily basis. There are some things you honestly can’t teach. You have to experience them to fully grasp the nature of some problems. That being said, I believe if one studied hard enough and went through all the material provided, they would be able to pass the test.

1. Udemy

Udemy provides a large variety of courses and online test prep guides. The following three courses are what I used to prepare for the Amazon Machine Learning Speciality Exam. I recommend taking a lot of good notes when going through this course. The practice exams are optional, but I would recommended it. It had information that wasn’t necessarily covered in the Full Course because it relates to experienced based knowledge.

2. Deeplearning.ai

Deeplearning.ai courses are by far my favorite. If you really want to understand the inner workings of Neural Networks and Deep Learning I HIGHLY recommend the following courses. The best part is that they are free. If you want a certification you can pay to get one. I was able to get the both courses done within a month. So I only had to pay $50 dollars at the time for my certification. Andrew Ng is an amazing professor and the deeplearning.ai team produces amazing content. Disclaimer, If you are not confident in your mathematical skills I recommend looking at Fast.ai instead, or take the prerequisites they recommend in the courses.

3. Flash Cards

If you are like me and don’t have a photographic memory, I highly recommend creating some flash cards. There are a lot of fine details in the question that can help you. The details are usually related to information about a particular machine learning algorithm in Amazon SageMaker. To be honest, knowing the finer details of the algorithm will help when addressing problems in real-life. Here is a list of things I would recommend creating flash cards on:

  • Machine Learning Algorithms in Amazon SageMaker
  • Machine Learning Algorithms Purpose/When to use <- Probably MOST IMPORTANT
  • Machine Learning Important Hyperparameters
  • Machine Learning AWS Instance Types
  • Machine Learning Algorithms in Amazon SageMaker Training and Inference Endpoint Input Types <- IMPORTANT
  • Metrics like Recall, Precision, F1 Score, and etc. <- IMPORTANT
  • Machine Learning Algorithms in Amazon SageMaker that ARE/ARE NOT parallelizable
  • How to effective clean/fix your data to give you better results from your machine learning model
  • How to know when you are underfitting or overfitting your model from your learning curve plot. Related to terms Bias & Variance
  • AWS AI/ML Services what they do and how they are best used

There are many more topics to create flash cards on, but this list is a great start.

4. Practice (Numer.AI or Kaggle Competitions)

Finally, after going through all that content I recommend trying to compete in a Kaggle Competition or Numer.AI competition. Doing this will give a great way to understand the inner complexities of machine learning and how it works in practice! Numer.AI is a great way to just get started. You don’t have to perform any cleaning of data or data augmentation. You mostly have to implement different algorithms and/or implement your own neural network architecture. A good understanding of python is needed though.

Kaggle Competitions are a little bit more involved. Most of the time you have to perform some sort of data cleaning, data augmentation, feature analysis, and etc. on your data before you get to the model creation. It gives a better sense of what a data scientist/ machine learning engineer would do though.

That’s all folks

I hope I provided a sufficient amount of information and resources, or atleast help bridge the gap for going about passing the AWS Machine Learning Speciality Exam. Be on the lookout for more content related to Amazon SageMaker. Possibly soon to come an Amazon Braket review! I also hope to create another post “Passing the GCP Professional Data Engineer Exam”. Thanks for reading! If you have any thoughts, suggestions, or questions please leave a comment below!

About the Author

Scott Poulin profile.

Scott Poulin

Sr. Consultant
Leave a Reply

Your email address will not be published.

Related Blog Posts
Natively Compiled Java on Google App Engine
Google App Engine is a platform-as-a-service product that is marketed as a way to get your applications into the cloud without necessarily knowing all of the infrastructure bits and pieces to do so. Google App […]
Building Better Data Visualization Experiences: Part 2 of 2
If you don't have a Ph.D. in data science, the raw data might be difficult to comprehend. This is where data visualization comes in.
Unleashing Feature Flags onto Kafka Consumers
Feature flags are a tool to strategically enable or disable functionality at runtime. They are often used to drive different user experiences but can also be useful in real-time data systems. In this post, we’ll […]
A security model for developers
Software security is more important than ever, but developing secure applications is more confusing than ever. TLS, mTLS, RBAC, SAML, OAUTH, OWASP, GDPR, SASL, RSA, JWT, cookie, attack vector, DDoS, firewall, VPN, security groups, exploit, […]