You ask about uploading three different things, so it's worth answering with regard to the three parts.
Your own code: Definitely acceptable to upload it. You wrote
the code so
the code is your original work and you would own the copyright in
the code. The reasons for citing the original work in the context of
your own code would be two-fold: (a) to put your own code in context,
and (b) to acknowledge an academic debt to the original author. In
terms of copyright, however, I see no legal requirement to cite
the original paper.
Your table of results: Definitely acceptable to upload it. This is
your original work, however closely your results approximate the
original. The reasons for citation of the original work, in the
context of your table of results, are exactly as in point 1.
Original table, for comparison purposes, with citation: From the point
of view of academic ethics, this would look quite acceptable. However,
the copyright issue is less clear and will depend upon the jurisdiction
of the place where you are publishing, and possibly also, on the place
where the original work was published. The notion of "fair use" in
copyright works varies very substantially from jurisdiction to
jurisdiction. If the original table (i.e., the other author's original
work) is more than a few lines long, I strongly recommend that you
seek permission to publish it. Just have a look at "permission
requests" on the website of the original journal.
Other answers here seem to focus on the issue of copyright.
I will add that it is also a good idea to publish your code and results in the interest of documenting replication of the original study. This is an important aspect of science, being able to replicate/reproduce the results of others, as it helps validate scientific findings.
In this context, I think it is also important to cite the original work to help document your replication of their work. Actually, you could even consider publishing your replication in a journal that focuses on this important aspect which is usually ignored by traditional journals in favour of new results. One such journal is Rescience C.