Machine Learning 2: Tooling
Time & Location
About the Event
New in ML. What should I choose?
Every new start comes with choices. And often, these (uninformed) choices impact the result you'll ultimately get. If you've made the right choices in the beginning, the learning should be meaningful and enduring. On the other hand, if your choice was somehow unlucky, you may end up frustrated and give up at some point. Imagine having to choose between learning Perl or Python in 2000, or between programming in NetBeans, Eclipse or IntelijIDEA in 2008, or between developing for Symbian or iOS in 2009, or learning CoffeScript vs TypeScript in 2014 etc..
At some point, all these options look equally valid, but looking back, only one was correct.
In this talk we'll make a survey of what are your options when trying to start learning ML. We will look at programming languages, IDE's and tools, libraries you need to master, and ML frameworks. At the end of this talk you should have a clear grasp of what lies ahead, and what to choose from. I'll also share my view of what to use, if you're just starting out.
Over-fitting and Under-fitting
Maybe you've heard these terms before. They represent two cases in which an ML model usually under-performs.
In this talk we will visually discuss:
- what each term means
- how you can usually spot these problems
- what strategies can we use for fixing them