Global Data Science Forum

Real-World Enterprise ML/AI Workflows with TWIML & IBM

By Sam Charrington posted Tue March 03, 2020 08:03 PM

  

Recently we kicked off a TWIML study group so that folks interested in taking the IBM AI Enterprise Workflow courses on Coursera could collaborate and support one another.

Each of the courses in the six-course series is two weeks long (plus another two weeks at the end for a capstone project), and we just completed the second week of the first course. Woohoo! In this post, we’ll recap some of the key concepts we discussed in the first couple of study group sessions and share various resources including the videos and slides from the sessions.

The first week of the course focused on reviewing the course structure and discussing some high-level introductory topics like:

  • Design thinking
  • Data collection
  • Prioritizing opportunities
  • The scientific method
  • Gathering data

We had a lively discussion about these concepts, which you can check out here:

You can also download the slides.

The second week of the course dug a bit deeper into what is undoubtedly one of the first steps of any machine learning project--data ingestion. Other topics discussed include:

  • ETL, etc.
  • Data science vs data engineering
  • Sparse matrices
  • Data testing
  • Automation of data ingestion pipelines

If you missed the live session, the video is below:

Slides for the week 2 discussion are here.

While we’ve completed the first course in the sequence, it’s not too late for you to join in the fun! Visit twimlai.com/aiew for details on joining the study group. We'd love to have you with us!


#AI-skills
#AI-skills-feature
1 comment
1706 views

Permalink

Comments

Tue March 10, 2020 06:02 PM

Last week our study group focused on the first week of Course 2 of the IBM AI Enterprise Workflow specialization on Coursera. This focus of the week was exploratory data analysis (EDA) and data visualization.

Our learning objectives for the week were:

  • Understanding the key steps in exploratory data analysis
  • Refreshing ourselves on key Python tools for EDA (pandas, matplotlib, and Jupyter)
  • Exploring strategies for dealing with missing data
  • Appreciating the role of communication in EDA

The information on strategies for handling missing data was particularly interesting.

If you missed the session, you can catch the recording here:

https://www.youtube.com/watch?v=CB9PeDpfkE0

Next week we’ll be covering Null Hypothesis Testing. Please join us even if you’ve missed previous meetings. 

Also feel free to post your questions here about the material being covered in Course 2 week 2 or any of the previous courses. 

For more about this supportive community program, visit https://twimlai.com/aiew.