Boost your career with AWS Machine Learning — Specialty Certification
Machine Learning (ML) is one of the hot skills in the market currently. If you have fair knowledge of ML already then it definitely helps you to get your first machine learning project. But what if you don’t have the knowledge about it but still want to make an entry into it. One of the ways is to do a certification program like AWS. This not only helps you build the knowledge about Machine Learning but also certifies it. It is one of the most difficult certification exams but it is definitely worth doing. More than AWS knowledge, the exam covers the ML knowledge in general.
AWS Machine Learning Certification Exam
As per the AWS Certified Machine Learning — Specialty exam home page, it is intended for individuals who perform a development or data science role. It validates a candidate’s ability to design, implement, deploy, and maintain machine learning (ML) solutions for given business problems.
Like any AWS exam it is a multiple choice, multiple answer exam and you get close to three hours for 65 questions. Questions are not that lengthy like some AWS professional exams. AWS has divided exams into three categories. Associate, Professional and Specialty. While Associate and Professional exams focus primarily on AWS knowledge, Specialty exams test the particular field knowledge along with AWS. so you will find a lot of questions which are not about AWS at all. The point being covering just AWS services knowledge is not enough at all.
It is not a SageMaker or just AWS exam!
When I began reading about the exam, a lot of blog posts mentioned that the exam is about SageMaker, SageMaker and SageMaker! It might have been true at some point but it is definitely not true anymore. You might get 10–20 questions on SageMaker but you will get many more questions on the Machine Learning field.
With enough context set, I will take you now on how I prepared for the exam. Please note that I do not have much experience in the field so I had to start from zero. If you are like me you can follow all the material. If you have knowledge you can skip initial basic material.
Study Resources
The first thing you should start with is learning about the field in general and not focus on AWS services. I will divide this section into two halves where first focuses mostly on Machine learning general knowledge and second covers specific AWS AI/ML services. Good thing is AWS itself provides a lot of free material and yes these are not geared or biased towards their services. These are available freely on aws.training website. This is what I covered.
aws.training courses
The Elements of Data Science (Highly recommended to do it few times)
Types of Machine Learning Solutions
ML Building Blocks: Services and Terminology
Process Model: CRISP-DM on the AWS Stack
Developing Machine Learning Applications
Exam Readiness: AWS Certified Machine Learning — Specialty (Recommend doing it towards the end closer to exam)
Udemy Course
This is the paid course I enrolled into. All topics start with intuition video followed by implementation in Python and R. I did cover Python implementation but most important is to focus on intuition.
Machine Learning A-Z™: Hands-On Python & R In Data Science
YouTube Videos
I subscribed to Krish Naik channel and watched almost all his videos.
I recommend watching his following playlists many times.
Complete Machine Learning PlayList
DataScience and Machine Learning with Python and R
I also covered lot of videos from StatQuest Machine Learning Playlist
Blogs
This is one Blog I covered from AnalyticsVidhya on distribution model.
It is very important to understand the Confusion Matrix.
AWS Machine Learning Services
So far, most of the content I have shared is mostly around Machine Learning in general. Make sure you cover it well. Once you are comfortable with it, you can move to AWS Machine Learning Services. These are the content I went through to cover the AWS part.
ReInvent Videos
PluralSight Courses
I used PluralSight to cover AWS service-wise knowledge.
Getting Started with Amazon Lex
Build, Train, and Deploy Machine Learning Models with Amazon SageMaker
Analyzing Text on AWS with Amazon Comprehend
Turning Speech into Text on AWS with Amazon Transcribe
Translating Languages on AWS with Amazon Translate
Deep Learning Instances and Frameworks on AWS
AWS Documentations
It is a must that you understand following section of SageMaker pretty well. Go through each and every SageMaker Built-in Algorithm, understand the use cases, underlying algorithm implementation, Common Parameters, Hyper Parameters, CPU/GPU/multi-GPU Support etc.
Apart from SageMaker and AI services, there are other services which has ML capabilities, For Example: QuickSight, Glue, EMR, Kinesis Data Analytics etc. make sure you understand those as well. Following link talks about Glue ML capability.
Practice Exam
Finally it is important that you practice well. I used the following practice exam from Udemy.
AWS Certified Machine Learning Specialty: Full Practice Exam
I am sharing my notes which I prepared going through various courses. Here is a link to my notes.
I hope this helps you make an entry into the field and boost your career prospects. All the best!
