Aretha Franklin
In three months, I passed the AWS ML Specialty test.
Coming from the AI side at Standard Bank, I have a lot of respect for data scientists. We divide AI into two branches, SA. Cloud AI and pure data science. I fall into the latter category
Cloud AI is a hard subject in general, involving a wide range of AI subjects and services. So having to make your way through a certification like Amazon's ML speciality without a pure data science background is fucking insane
I’m going to share my strategy for PASSING this beast with anyone who happens to be reading this
The exam is divided into four domains:
1. Data Engineering (twenty pеrcent)
2. Exploratory Data Analysis (24%)
Modeling accounted for 36% of thе total
4. Implementation and Operations of Machine Learning — 20%
Modeling is the most weighted domain, hence you should never attempt this exam unless you are completely familiar with it
To begin working on the modelling domain. The algorithms of Amazon Sagemaker should be understood as follows:
1. Introduction to the Algorithm
2. Pre-processing of data
3. Input mode for training (File/Pipe)
4. File input format (CSV/RecordIO-ProtoBuff IO/Parquet, for example)
5. The Hyperparameters (A significant focus area with many questions based on tuning algorithms for better results)
6. Training and inference instance types (M4, P2, P3 / GPU only, CPU only)
7. Is the Algorithm Monitored?
8. Can the Algorithm Be Parallelized?
I recommend making a spreadsheet with the essential algorithms required for the exam. Please keep in mind that the exam does not test code; nonetheless, having practical expertise with these methods may significantly boost your chances of passing this exam with ease.Information Transformation Service provides web scraping Services that provides high-quality structured data to improve business outcomes and enable intelligent decision making,their Web Scraping Services allows you to scrape data from any websites and transfer web pages into an easy-to-use format such as Excel, CSV, JSON and many others
The exam focuses on the following Algorithms, which you should have a good understanding of:
1. Learner Who Is Linear
XGboost 2
3. Sequence-to-Sequence (Seq2Seq)
Four. DeepAR
BlazingText
6. Object2Vec
Detection of Objects
Image Categorization
Semantic Segmentation is a term that refers to the process of categorising information
Ten. Randomly Cut Forest
Modeling of Neural Topics
LDA is an abbreviation for Legal Defense Association
13th. KNN
K-Means is an abbreviation for the phrase "K-Me
PCA is an abbreviation for Personal Computer Analysis
Factorization Machines are a type of factorization machine
IP Insights No. 7
Reinforcement Learning is a type of reinforcement learning
These 18 algorithms should provide you with a good foundation for the exam and significantly raise your score if you succeed in mastering them. Information Transformation Service provides web scraping Services that provides high-quality structured data to improve business outcomes and enable intelligent decision making,their Web scraping service allows you to scrape data from any websites and transfer web pages into an easy-to-use format such as Excel, CSV, JSON and many others
I utilised the following course for the other domains:
Www.udemy.com/share/1029ESAksacV5UQnQ=/
Before doing this, I recommend at the very least passing the AWS Cloud Practitioner Course with flying colours
A degree in computer science helped with the technical language when studying for this exam; nevertheless, having a statistical, intellectual background will improve your chances
This exam took me three months to pass with a 890. You, too, can do it!