Text-only version of this email
Hi, you must have a systematic process when working through a predictive modeling problem.
Do you make things up every time you start a new software project? The same applies to machine learning.
You can guarantee at least above average results on each predictive modeling problem you encounter by using a systematic process
to work through it.
* A good process guides you end-to-end, from specification to deployment.
* A process is step-by-step so that you always know what to do next.
* A process ensures a "good" result, and results improve on all projects as the process improves.
* A process is not tied to any specific language, library or tool, they are a means to an end.
You don't see much mention about applied machine learning processes because few people are actually working through problems and
writing about it.
A good process template I like to use involves the following steps:
* Step 1: Define the problem
* Step 2: Prepare the data
* Step 3: Spot check algorithms
* Step 4: Improve results
* Step 5: Present results
We will go into these steps in more detail in later emails. If you would like more, see the post:
>> Process for working through machine learning problems
I'll speak to you soon,
Jason.
To make sure you keep getting these emails, please add [email protected] to your address book or whitelist us. Want
out of the loop? .
Our postal address: 151 Calle de San Francisco, Suite 200 - PMB 5072, San Juan, PR 00901
Show all