Hi, I open up this topic to discuss and report some things I noticed or found unclear in the course documents for the “Introduction: Machine Learning concepts” section.
Feel free to participate or to contradict me!
Hi, I open up this topic to discuss and report some things I noticed or found unclear in the course documents for the “Introduction: Machine Learning concepts” section.
Feel free to participate or to contradict me!
03:05. The correspondences between colors (on the plot) and iris type are not specified (the is no legend at all in fact – using a screenshot from https://github.com/INRIA/scikit-learn-mooc/blob/master/slides/ml_concepts.md#engineering-rules-data-versus-experts would be clearer in my opinion).
00:51 & 00:56 (minor). Text is quickly displayed and disappears immediately
Question 2. It is not made clear during the video that the target variable cannot be called “a feature” (07:20, “Iris type” appears as a column, while it is said columns are features). The question is about “features” and the anwser “predictive features”; I didn’t understand correctly that “features” were always “predictive features”.
And number two:
This silly forum doesn’t allow me to put two images into one message! And my first attempt at a second message was greeted with “You’re replying too quickly. Please wait a few seconds before trying again.” Treating forum users as potential spammers is not a very nice attitude!
Thanks a lot for your comments!
I fixed the typo in our slides, but fixing the video is another matter and I have to admit some other Pedagogical team members (or you given your experience in the Reproducible Research MOOC) will know more about how to proceed than me .
Treating forum users as potential spammers is not a very nice attitude!
100% agreed, not a Discourse expert, but I am sure there are Discourse settings that can be tweaked. Let’s say that the beta is a good opportunity to find good default settings .
Hello,
The typo in slides will be corrected in the video also.
For the forum, Benoit modified the parameters of discourse (now, same as RR and IOT Mooc).
Fixed ! You can now add up to 20 images and post as many messages as you want