Skip to content

analondhe/ml-aws-specialty-lab

 
 

Repository files navigation

AWS ML Specialty

This repo includes various jupyter notbooks, notes from courses, white papers that helped me pass the AWS ML Specialty certification in 2020. It is not intended to list every resource that exists out there but should help some of my fellow associates to get started with some good documentation.

I personally passed that exam on 11/10/2020 with score of 922 / 1000 or 92.2%. The passing score is 75%. I found the exam quite hard and I am glad I invested time (roughly 3 months) to refresh and/or learn on various topics in modeling, operations, monitoring, security and data engineering/analysis.

ml-specialty-image

ML Specialty certification overview:

"The AWS Certified Machine Learning - Specialty certification 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."

Topics covered during the exam:

  1. Data Engineering (20%): S3 (and VPC Endpoint Gateway), Kinesis (Streams, FireHose, Data Analytics, Video), Glue (Data Catalog and Crawler), Athena, AWS Data Stores (Redshift, RDS/Aurora, DynamoDB, ElasticSearch, ElastiCache), AWS Data Pipelines, AWS Batch, AWS DMS, AWS Step Functions
  2. Exploratory Data Analysis (24%): Data Types and Distribution, Time Series, Amazon Athena, Quicksight, Ground Truth, EMR, Spark, Dat binning, transforming, encoding, scaling and shuffling, Dealing with Missing data, outliers, unbalanced data, outliers,
  3. Modeling (36%): CNN, RNN, Tuning neural networks, Regularization, Gradient, L1 and L2 regularization, Confusion matrix (Precision, Recall, F1, AUC), Ensemble methods (Bagging and Boosting), Amazon Sagemaker, Amazon Algorithms (Linear Learner, XGBoost, Seq2Seq, BlazingText, DeepAR, Object2Vec, ObjectDetection, Image Classification, Semantic Segmentation, RCF, LDA, KNN, K-Means, PCA, Factorization Machine), Amazon AI Services (Comprehend, Translate, Transcribe, Polly, Rekognition, Forecast, Lex, ...)
  4. ML Implementations and Operations (20%): SageMaker Production Variants, Neo, IoT Greengrass, Encryption at Rest and in Transit, VPC, IAM, Logging, Monitoring, Instance Types and Spot Instances, Elasstic Inference, Auto-Scaling, Availability Zones, Inference Pipelines, ...

Additional Info about the exam

Cost; $300 180 mn long (3 hours) and ~65 questions

  • multiple choice
  • multiple response

Notes:

  • No partial credit for questions (if we get 2 or 3 right out of 5, no credit)
  • Can mark questions an g back to them
  • No points for unanswered
  • Scores: between 100 and 1000
  • Minimum passing score: 750
  • Scaled scoring models are used

4domains

1. Course and lab

1.1 AWS SageMaker Notebooks

I have tried many SageMaker notebooks in my personal account to really get a good feel for the various algorithms and modeling techniques. You can see the list of Sagemaker repos in this other repo: aws-sagemaker-notebooks

1.2 CloudGuru Course

The Cloud Guru - AWS Certified Machine Learning - Specialty 2020 Course includes over 17 hours of videos, 79 lessons, 8 course quizzes and 1 practice exam. There are also some great labs to get hands on.

Below are some of my notes/snapshots from the course:

CloudGuru Lab Work

1.3 Udemy Course

The Udemy AWS Certified Machine Learning Specialty 2020 - Hands On! Course includes over 9 hours of videos, 114 lessons and 1 practice exam. The full list of course slides is available here.

Below are some of my notes/snapshots from the course:

1.4 WhizLabs courses

The Whizlabs Course (and tests) is a great course, with a lot of examples/labs via Jupyter notebook to grasp the materials taught. I also found the Tests a lot harder than Udemy and CloudGuru and would recommend passing these tests last.

Below are some of my notes/snapshots from the course:

1.5 AWS Course

1.6 AWS White Papers

1.7 AWS FAQ

1.8 AWS Machine Learning Cheat Sheet

2. Practice Exams

2.1 CloudGuru Practice Exam

passed on 10/18/2020: scored 75% and used 1h35 from the 3 hours to cover the 65 questions

2.2 Udemy Practice Exam

2.3 Whizlabs Practice Exam

2.4 TestPrep Practice Exam

2.5 AWS Practice Exam

2.6 Misc

Resources

Great set of courses from Andrew Ng in Coursera that I highly recommend:

Article extracts and notes (pdf):

Other:

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 99.8%
  • Python 0.2%