It was rather too large, especially when inline with text. I might
have gone just a *tad* too far in the other direction, but it mostly
looks very good now.
Major revamping - a lot of bayesian math has been added, and the
examples have been tweaked. I corrected a major confusion in the
update function - I was scaling, but said I was multipling by a
probability.
Also modifed the css for mathjax so that the font renders in a more
reasonable form.
The CSS was causing the text bounding box to slightly overlap the
text. A kind person explained the problem by posting to issue 16.
However, his change didn't work. I removed the width field and for
some reason all the headings centered themselves. I tried changing
their alignment and that didn't work. I comprimised by changing 750
to 540 and that worked.
I also slightly changed the font size. It was almost comically large.
Most of the graphs were really too tall. Well, we can quibble about that
but I made them smaller to reduce the PDF size and thus print size
of the full book.
I had checked it in to share it with people while debugging issues
in IPython notebook. But now it is confusing people who think it is
a separate notebook. This is jsut the notebook that I create to generate
the PDF, and should not be checked in.
this was the older license, I have allowed commercial and other usage with CC-BY. The readme was updated awhile ago, but I neglected to fix the license.html file.
Too many to mention. Read through several chapters and made changes
as I went. I've had people (correctly) question me on several points,
and pointing out typos.
I think I prefer just letting nbconvert generating static files,
but if I want to switch to using Bokeh or other javascript based
rendering I'll need the HTML conversion facility.
I had information on these rather buried, and did not have a single example
of how to compute them. Now the first part of the chapter covers this much
more thoroughly.
Also added a section on plotting exponentials. Not sure I want to retain
it; it is a bit 'light'.
If you add a semi-colon to the end of the interact call
it hides the printing of the return function. This happens
because interact is both a function and a generator so it
can be a decorator.
I finally learned that the raw cells let you put latex directly
into the output. So now I have the appendix being created without
the [1] %%latex showing up in the output. Also wrote code to parse
the .tex file to make the Preface a chapter without numbering, so it
is now a proper preface, and not "chapter 0".
I have not touched the Windows generation tools, so PDF generation
in Windows is now broken. I don't think I care.
Some sections jumped from level 2 to level 4, for example.
Also played around with generating book to HTML. I like it, but
it needs some work. The Calico table of content generator cannot
deal with different headers with the same name, for example.
I want to go to HTML so I can use Bokeh to have interactive plots.
But that means no more PDF, because Bokeh will not export to
PDF.
Use of markdown was causing chapters to be merged together.
I made the change to markdown in some place in preparation for
IPython 3.0, but that turned out to be a bad idea (plus, 3.0
will automatically convert the headers correctly).
I was reusing the names for the state and measurement functions,
which made it impossible to run cells out of order, and was no
doubt confusing to the reader.
I think I finally arrived at a good ordering of material.
Started with implementing a linear problem just so we can
see how it differs from the linear KF, then added problems
step by step. Got rid of most of the poor performing filters.
I moved it more to the front as the pos vs velocity explanation
is difficult to understand if you do not already understand how
orthogonal or correlated measurements greatly increase precision.
Also various minor wording changes and fixes in other parts.
Position vs velocity confuses people, so I added a 2 position
reading example.
Fixed the Newtonian equation for jerk.
Changed indefinite integral in Newtonian equations to definite.
I wrote a chapter introducing the problem of nonlinearity in KF because this
information is common to the EKF, UKF, etc. Then I reordered chapters. I want
people ot learn the UKF before the EKF. While I was doing this, I moved the
H Infinity and Ensemble filters to later chapters because they are mostly
curiosities in this book.
The contents section is a pain to maintain, and it doesn't really
add anything of important. Maybe i'll add it back in once the book
is settled down, but probabl not.
Roger Labbe