This is a great first start. Some resources the author may consider drawing upon, depending on whether and how they choose to expand:<p>[1] <a href="http://quant-econ.net/py/index.html" rel="nofollow">http://quant-econ.net/py/index.html</a><p>[2] <a href="http://people.duke.edu/~ccc14/sta-663/" rel="nofollow">http://people.duke.edu/~ccc14/sta-663/</a>
One doc to learn them all :)<p>The `Optimizing and debugging code` part is where most data scientists falter.<p>So, this is a really nice effort for bringing it all together!
I wish I had this when I was learning the Python data analysis ecosystem. Does a nice job of clearly distinguishing the differences of the major elements.
I hope you don't take this as trolling, but: What's the deal with matrix multiplication in numpy? I wanted to dot-product two vectors yesterday, and I got it right only on the third try:<p><pre><code> x.T * y # nope
x.dot(y) # still no
x.inner(y) # ok
</code></pre>
This is a disaster. I'm sure there are valid historical reasons for this state of affairs, but this makes numpy an environment where random idiosyncrasies get cast in concrete.
This looks like a great document to help one get started on the road to using Python for lots of STEM type tasks. Pandas fans: don't be put off by the lack of mention of it on the title page, as it is covered in there, too.