Main project grading rubrics
<br>
<ol>
<li>
Grading is subjective: Insight over charts/tables
<li>
Rubrics are weighted according to appropriate project context
<br>
<font size=-1>
For example,  project where clustering is inappropriate and data fitting is 
critical, points move from data fitting to clustering
</font>
<li>
+ indicates:  must have (if appropriate, see previous point)
<br>
* indicates:  extra points (desirable, but do not try to artificially introduce)
</ol>
<br>

<ol type="a">
<li>
data:
<ol type="i">
<li>
+ characterize:  incomplete, uncertain, outliers
<li>
+ filter/scrape/clean   (complexity gets * )
<li>
* large data sets
<li>
* streaming data sets
<li>
+ combine different data sets
</ol>
<li>
hypotheses
<ol type="i">
<li>
+ clearly stated 
<li>
+ devise numerical experiment(s) to confirm or deny
<li>
* causality vs correlation
</ol>
<li>
fit data 
<ol type="i">
<li>
+ regression, MLE, learning  (increasing points)
<li>
* prediction
</ol>
<li>
other data science techniques
<ol type="i">
<li>
+ cluster data
<li>
...
</ol>
<li>
presentation.
<ol type="i">
<li>
* engaging visual representation of data 
<li>
+ conclusions
</ol>
</ol>

<!--   <a href="hsb2.csv.xlsx"> data </a>. -->