update fpca references #636
Conversation
examples/plot_fpca.py
Outdated
| @@ -25,7 +25,22 @@ | |||
| # to explore datasets and obtain conclusions about said dataset using this | |||
| # technique. | |||
| # | |||
| # First we are going to fetch the Berkeley Growth Study data. This dataset | |||
| # First we will introduce what FPCA is and why it is commonly used in many | |||
There was a problem hiding this comment.
I would just introduce it, without this line. I think it is safe to assume that the reader already knows what FDA is, at least by skimming the readme and/or the tutorial (otherwise we would need to add such an introduction to every example).
examples/plot_fpca.py
Outdated
| # studies. Generally, functional data is represented as functions X(t) where | ||
| # t might be time or some other continuous variable. In theory, these | ||
| # functions can have an infinite number of data points (they can be treated as | ||
| # infinite dimension vector spaces) so the goal of FPCA is to reduce the |
There was a problem hiding this comment.
I would expect some definition ("FPCA is ..."), like "FPCA is a dimensionality reduction method for functional data. The goal of FPCA is to reduce [...]".
examples/plot_fpca.py
Outdated
| # variation across the function (the most important directions in which the | ||
| # curves vary). FPCA can be though of as a basis expansion, but what | ||
| # distinguishes FPCA is that among all basis expansions that use K components | ||
| # for a fixed K, the FPC expansion explains most of the variation in X. |
There was a problem hiding this comment.
| # for a fixed K, the FPC expansion explains most of the variation in X. | |
| # for a fixed K, the FPCA expansion explains most of the variation in X. |
examples/plot_fpca.py
Outdated
| # For more information abour FPCA and its objectives, see | ||
| # :footcite:ts:`wang+chiou+muller_2016_fpca`. | ||
| # | ||
| # In second place, we are going to fetch the Berkeley Growth Study data. This dataset |
There was a problem hiding this comment.
This is really the first step of the tutorial, the previous paragraph being a short explanation of the method. I would drop that commit.
examples/plot_fpca.py
Outdated
| # principal components, which are the directions that capture the main modes | ||
| # of variation across the function (the most important directions in which the | ||
| # curves vary). FPCA can be though of as particular case of a basis expansion | ||
| # that uses K components for a fixed K, with the particularity being that the |
There was a problem hiding this comment.
I actually liked very much your sentence "among all basis expansions that use K components for a fixed K, the FPC expansion explains most of the variation in X.". Why did you remove it?
I also noted that your commit messages are not very informative. If you want, you may try to use git rebase in interactive mode to rewrite your git history and squash all your commits together (or in other situations, to squash some of them, change the messages or reorder them). It is not strictly necessary (I could squash them when merging, or leave them) but it is a tool that is sometimes useful and you may want to try.
Note that, in general, rewriting git history should only be done in private branches that you share with nobody, or with a few other developers you communicate with (so, never rebase master or develop). After rebasing you necessarily have to push with the "--force" (or better "--force-with-lease") option, which means that any other developer that is using the branch would have it in an incompatible state and should reset it.
aleexarias
force-pushed
the
fpca_references_fixed
branch
2 times, most recently
from
2026b93
to
35bf3bf
Compare
update fpca references FPCA reference added to docstring + added FPCA explanation in example Fix references in FPCA
aleexarias
force-pushed
the
fpca_references_fixed
branch
from
35bf3bf
to
9031ea2
Compare
aleexarias
force-pushed
the
fpca_references_fixed
branch
from
f2e5cda
to
0ea1472
Compare
| # for a fixed K, the FPCA expansion explains most of the variation in X. | ||
| # | ||
| # For more information abour FPCA and its objectives, see | ||
| # :footcite:ts:`wang+chiou+muller_2016_fpca`. |
There was a problem hiding this comment.
There is no footbibliography, so the references are not shown in this file.
| when fitting a FDataBasis or FDataGrid, except for ``components_basis``. | ||
|
|
||
| For more information about the implementation of the computation of the | ||
| first principal components see :footcite:ts:`silverman_2005_basisfuncexp`. |
There was a problem hiding this comment.
The reference is wrong. Please, check the generated documentation (either locally, or in the PR, in the details link of the Readthedocs check).
Co-authored-by: Carlos Ramos Carreño <carlosramosca@hotmail.com> New commit
aleexarias
force-pushed
the
fpca_references_fixed
branch
from
436c23a
to
68cb73e
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #636 +/- ##
========================================
Coverage 86.83% 86.83%
========================================
Files 157 157
Lines 13522 13522
========================================
Hits 11742 11742
Misses 1780 1780 ☔ View full report in Codecov by Sentry. |
Resolves #634
The reference in _fpca is moved into the class docstring instead of the private function it was on, using footcite/footbibliography