Journey towards AWS Certified Machine Learning Specialty (MLS-C01) Certification
“It is good to have an end to journey toward; but it is the journey that matters, in the end.” — Ernest Hemingway
Above quote from Ernest Hemingway is one of the core ideas to keep in mind during your preparation for AWS Certified Machine Learning Specialty (MLS-C01) Certification. I have passed MLS-C01 certification with the score of 865 in my first attempt. My academic background (MTech in Artificial Intelligence and Robotics) immensely helped me in preparation along with working knowledge of state-of-the-art NLP models. Although I must mention that you don’t need to be MS or PhD in AI/ML to achieve this certification, but hands-on experience is very much required along with strong AI/ML fundamentals. Preparation period for MLS-C01 differs from one to another based on the experience with AWS AI/ML offerings and different machine learning framework such as Keras, Scikit-Learn, TensorFlow, MXNet, PyTorch and Containers services, Docker. I will be sharing stepwise approach for preparation of this certification from my own experience. I would like to mention at the very beginning that this certification shouldn’t be taken as first AWS certification exam. As per AWS official exam guide, recommended AWS Knowledge — The successful candidate likely has 1–2 years of hands-on experience developing, architecting, or running ML/deep learning workloads on the AWS Cloud, along with: · The ability to express the intuition behind basic ML algorithms · Experience performing basic hyperparameter optimization · Experience with ML and deep learning frameworks · The ability to follow model-training best practices · The ability to follow deployment and operational best practices.
It would be a good strategy to take AWS Certified Data Analytics Specialty (DAS-C01) before taking MLS-C01 as it almost overlaps 20% of the content related to data engineering. DAS-C01 will provide solid concept on AWS Glue, EMR, Amazon Redshift and distributed data processing on AWS. Again, this is not a mandatory ask from the MLS-C01 preparation. If you are interested for AWS Certified Data Analytics Specialty, please refer my another blog for DAS-C01 preparation.
Step 1: Go through AWS link to know about details about this certification.
Step 2: Go through official study guide to understand the various required topics to be cover.
Step 3: If you are familiar with AI/ML fundamentals, start with Udemy course AWS Certified Machine Learning Specialty 2021 — Hands On! offered by Stephane Maarek and Frank Kane. Please perform hands on in your own AWS account and ready to incur the cost as mentioned in this course. As per my experience, watching this training video will not help you to grab the titbits of implementation which is highly required to achieve this certification. Please make notes during taking this training as you can review later point of time.
If you are not familiar with AI/ML fundamentals, please start with basic course to grasp it. One of my recommendation is Coursera course on Machine Learning offered by Andrew Ng. Below YouTube free videos/Blogs can be helpful to build your fundamental concepts.
· Normalized Nerd
· Machine Learning for Free on AWS
Step 4: Once you complete above mentioned Udemy course by Stephane Maarek and Frank Kane. I would recommend you take sample questions provided by AWS. Please score yourself and don’t look into given explanations and correct answer for each of the questions. You will understand the reason of this strategy in step#7.
Step 5: My suggestion will be to take another Udemy course AWS SageMaker — Certified Machine Learning Specialty Exam offered by Chandra Mohan Lingam. This is more detailed and will give in-depth idea with hands on activity. Please be prepared to spend 10$ — 30$ if you follow all the hands-on activity in your own AWS account. My recommendation is to perform these hands on if you don’t have exposure to work with these technologies on daily basis. Please augment/add notes during taking this training to your notes which you prepared during previous training.
Step 6: Along with the above course, please read AWS Blog and watch Re:Invent videos to clear your concepts. Please find couple of important link for preparation:
https://aws.amazon.com/training/learning-paths/machine-learning/exam-preparation/ (Free AWS Digital training link will be available through this link. )
A Disciplined Approach to Neural Network Hyper-Parameters: Learning Rate, Batch Size, Momentum, and…
Training a machine learning algorithm requires carefully selecting hyper-parameters. But with neural networks, this can…
Amazon SageMaker Technical Deep Dive Series by Emily Weber.
Step 7: After preparation with various sources and 2 training courses, it’s time to validate the knowledge and get accustomed with question pattern. Don’t expect the same question in actual certification exam from the practice test but you will get enough idea about actual questions from these practice test. I would like to recommend 2 practice questions set. Start with Whizlabs which provides good/detailed explanation of the correct answer. This will help you to detect gaps from your preparation and fine tune accordingly. Please refrain from memorizing the question as it will jeopardize your correct evaluation in case of taking same practice test multiple times. After taking this practice exam, try sample questions provided by AWS again (already you took once during step 4) and measure your development. This time please go through the given explanation for each of the answer. This exercise is going to be a good confidence booster for you and will help to measure your preparedness.
Step 8: After taking Whizlab test and AWS sample questions, time to take another Udemy practice test by Jon Bosno. Expect to score more than 85% for both the tests. If. you score less, request you to continue with your preparation after identifying weak area. If you are still not confident, you can take another practice set offered by BrainCert. Expect some common questions in this practice set as it combines questions from other available practice tests along with their own set of questions. Expect to score 80% based on your preparation.
Step 9: As per my suggestions, one of the last steps before appearing in exam to take AWS Certified Machine Learning Specialist — Exam Readiness offered by AWS. Please go to your AWS certification account for taking this exam. Along with crisp revision of required content, it will give you opportunity to test your knowledge with 35 sample questions which are very similar to actual certification exam. By scoring 85% with these sample question, you can go ahead for real exam.
Step 10: Another final round of knowledge check can be done with paid (40$) practice test AWS Certified Machine Learning Practice (MLS-P01) offered by AWS. Please go to your certification account to take this. You need to answer 20 questions in 60 minutes. Questions are very similar to the actual certification exam. You can take this test at free of cost if you hold any prior AWS certification. Any AWS certification provides benefits of 50% discount in next certification fees and 100% discount on practice test.
Activities to be done in last 2 days before exam day
•Go through practice exam’s answer and corresponding explanation. Read the explanation from AWS documentation to hash out any bit of confusion.
•Watch video tutorial selectively as per your area of weakness.
• Review your notes
•Take adequate sleep and relax your mind to release the pressure created due to prolonged preparation.
•Go through below link as recap.
. Link for additional blog/experiences of MLS-C01
AWS Machine Learning Certification Exam Tips
I just passed the AWS Machine Learning exam. It is the most difficult topic of all AWS certifications, however if you…
· FAQ Links from Required AWS Services
Final Notes and Exam Day Tips
· Please familiar with new AI/ML and Data offering from AWS such as SageMaker Studio, SageMaker Data Wrangler, SageMaker Jumpstart, Amazon Glue Elastic View, SageMaker Experiments, SageMaker Feature Store, SageMaker Edge Manager. I have seen couple of questions in exam on this new service and it will be more with time.
· You may have all options with possible correct answer for a particular question but need to choose best answer with the given scenario. As example, on some occasion it may be asked to come up with most cost-effective solution or there may be ask for completely managed service (such as AWS Services, Serverless offerings). Read the question carefully.
o Time Management — Key to success
· Effective exam duration is 170 minutes. 65 questions need to be answers.
· Don’t spend more than 2 minutes on a particular question. Answer it as your best guess and mark review flag and come back after answering all the questions. It’s tough to do but practice it during taking practice exam itself.
· You should attempt all 65 questions in 130 minutes by spending 2 minutes on each. So, 40 minutes will be there with you to think through the difficult questions which you marked as review. I have finished in 150 minutes and reviewed for 20 minutes.
· Many irrelevant information will be there in question, but you need to skim through and choose keywords to find the correct option.
· Don’t forget to put your laptop in charge if you are taking certification examination from home through Pearson Vue or any other third party proctor. I know the instance where people couldn’t complete certification exam as laptop battery drained out completely.
Keep positive attitude and believe in yourself. Failure in certification exam is not end of anything, knowledge gathered throughout the preparation process is an absolute value addition to your professional life. So, there is nothing to regret. You can always take certification after 15 days in case of failure. At the end of the day. It’s the journey which matters more than destination!!!
Best Wishes to Everyone for AWS Certified Machine Learning Specialty Certification!
Enjoy Preparation and conquer It!!! Stay Safe!!!