Like many people, I kicked off this year with a long list of resolutions. My list includes items like learning how to do a backflip, but that’s a story for another time! One of my top three goals, and the one has come up regularly over the last couple of years, was to make an active effort learning cloud computing. Luckily, my current employer has chosen Microsoft Azure as their cloud platform, hence allowing me to focus on it.
Having just completed my machine learning course with Georgia Tech (in May 2020) and being more fluent in
Python, it was pretty obvious to me to choose Azure DP-100 exam.
What is the DP-100 Exam?
The Azure DP-100 Exam: Designing and Implementing a Data Science Solution on Azure is for data geeks who use machine learning techniques to implement a data science solution using Azure Machine Learning.
The DP-100 Exam is designed to evaluate candidates for the following tasks:
- Create, define and manage Azure Machine Learning workspace.
- Create and run experiments that log metrics and train machine learning models.
- Create and manage datastores and datasets, and use them in machine learning experiments.
- Create and manage compute resources, and use them to run machine learning experiments at scale in the cloud.
- Deploy predictive models as real-time or batch inference services, and consume them from client applications.
- Perform hyperparameters tuning, automate machine learning models.
Here is a diagram displaying the learning outcome visually.
For more details on the scope of the exam, click here.
pip install --upgrade azureml-sdk[explain,automl,notebooks]
Here are some key concepts to know prior to sitting the exam:
- Supervised Learning
- Regression methods: Linear Regression, Decision Trees, Random Forest, Deep Learning, etc.
- Classification methods: Logistic Regression, Linear Discriminant Analysis, Support Vector Machine, Multi-classification, etc.
- Unsupervised Learning methods: Principal Component Analysis, \(K-\)means Clustering, etc.
- Data Wrangling: Imputation, Imbalance Classes, etc.
The bulk of the questions focus on Python programming. In other words, as a candidate sitting the exam, you must be comfortable seeing some
Python codes without any trace of sweat on your forehead (to my R-fellows, you know who you are!). There might be some
R codes being thrown at you in the exam.
Either way, be comfortable reading/writing codes in both languages.
There aren’t that many centralized resources available to prepare for this exam. However, here are the ones that I used during my preparation time:
- End-to-end tutorials, and how-tos on the official documentation site for Azure Machine Learning service.
- Python SDK reference
- Azure ML Data Prep SDK overview, Python SDK reference, and tutorials and how-tos.
- Pluralsight MS Azure consists of 25 lectures teaching how Azure services work together to enable various parts of the Machine Learning workflow.
- Examtopics contains about 60 sample questions for free which may be helpful for exam prep.
Projects using Azure Machine Learning
Visit following repos to see projects contributed by Azure Machine Learning users:
- AMLSamples Number of end-to-end examples, including face recognition, predictive maintenance, customer churn and sentiment analysis.
- Learn about Natural Language Processing best practices using Azure Machine Learning service
- Pre-Train BERT models using Azure Machine Learning service
- Fashion MNIST with Azure Machine Learning SDK
- UMass Amherst Student Samples - A number of end-to-end machine learning notebooks, including machine translation, image classification, and customer churn.
I hope you found this blogpost useful. Please drop a comment/remark below :-)