Covering Computer Vision Problems and General ML problems in an interactive way
When we start learning Machine Learning, our main focus is building the model! The data usually is clean and ready. The task usually is a simple classifier or regressor. We keep learning several models and the math behind them!
What you’ll learn
- Machine Learning System Design Interviews.
- Machine Learning Pipeline.
- Concepts to start learning more behind them.
- Interactive interviewer-interviewee dialogue.
Course Content
- Getting Started –> 2 lectures • 8min.
- Machine Learning Stages –> 1 lecture • 4min.
- Twitter Timeline System –> 14 lectures • 2hr 24min.
Requirements
When we start learning Machine Learning, our main focus is building the model! The data usually is clean and ready. The task usually is a simple classifier or regressor. We keep learning several models and the math behind them!
In reality, we need to formulate the problem as a machine learning problem! We need data and the corresponding annotations. Most probably we need to do a lot of cleaning, preprocessing and visualizing the data. And then comes the model! A missing stage for many people is deploying the model and integrating it with a product!
In this course, we focus on highlighting all the machine learning pipeline:
- Scoping the problem
- Data: collection and annotation
- Metrics: online and offline
- Modeling
- Evaluation
- Deploying
What to expect in this course:
- To emphasize the machine learning pipeline, not just the modeling!
- To get deep insights about what does it mean to build a ML system!
- A good reference of questions to ask for yourself in your projects
- To prepare for the ML system design interviews!
- This is actually the major concern and what drives the content
- An interactive content: Question and Answer
Audience
- If you don’t know machine learning, this course is not for you
- If you just build toy ML projects, this course may not be for you
- If you build some projects or non-trivial Kaggle competitions, this course is for you
- If you build have market experience, this course is a must for you
Critical notes:
- Don’t take my thoughts for granted. Challenge them. Brainstorm in the QA section.
- I don’t explain machine learning concepts. I highlight them. It is your responsibility.
- You will be exposed to a wide range of terminologies
About the Instructor (relevant experience): I have started worked in machine learning since 2010. I am a Computer Vision Scientist with PhD from Simon Fraser University. My experience covers many areas such as algorithms design, software engineering, machine learning and teaching.
Don’t miss such a unique learning experience!
Acknowledgement: “I’d like to extend my gratitude towards Robert Bogan for his help with proofreading the slides for this course”