Closed jenshnielsen closed 9 years ago
I don't think adding the additional cognitive load is worth this clarification.
Agreed, but the PR satisfies requirements for becoming an instructor. Thanks, Greg
Thats fair enough thou and I agree that it is a trade off but I would argue that it is not just a clarification. The present version is actually mistreating the data by interpolating new artificial data points in the heatmap (Since when has patient 1.5 made any sense) That is to say not only is it ugly it is also wrong.
Good point - can you turn this into an exercise at the end of the lesson in which you point that out and ask people to fix it? That will let instructors whose classes are running slow skip over it, and others use it to provoke thinking.
@gvwilson That sounds like a good solution. I will do that.
Great idea Greg! On Nov 7, 2014 9:46 AM, "Jens Hedegaard Nielsen" notifications@github.com wrote:
@gvwilson https://github.com/gvwilson That sounds like a good solution. I will do that.
— Reply to this email directly or view it on GitHub https://github.com/swcarpentry/bc/pull/841#issuecomment-62153823.
Closing in favour of #849
I realise that this adds additional complexity to the very early stages of the Python session. However the present heatmap doesnt' represent the data very well since additional points are interpolated between the data points. One could argue that the matplotib default should be 'none' but that would be a major api break.
In a more advanced example I would also suggest changing the colormap to something better than the default (Jet) but at this stage I think it is an unnecessary complication.