daviddoji/Kalman-and-Bayesian-Filters-in-Python
1
0 Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,280 Commits 1 Branch 0 Tags
6cdc0afa150b295b8f3443c49a9bc5a46972ebd0
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Roger Labbe 6cdc0afa15 Merge pull request #272 from WillieMaddox/patch-1 
Update 03-Gaussians.ipynb
 2020-04-26 18:05:22 -07:00
Roger Labbe e84f801836 Merge pull request #269 from rummanwaqar/master 
Bug fix: Issues in Chap 3 and 4
 2018-12-20 08:01:35 -08:00
Roger Labbe 9e4ac42f79 Merge pull request #267 from rummanwaqar/master 
Bug fix: moved __future__ import to the top
 2018-12-13 12:02:12 -08:00
Roger Labbe 17f38c69f2 Fixed for new code organization and book format 
These haven't been run in awhile; I updated them so they
have the same look and feel as the book, and fixed a few
imports and pep 8 compliance as well.
 2018-11-02 16:11:20 -07:00
Roger Labbe 81eff753b5 point to localhost 
The links in the toc were pointing to nbviewer. Github #239
 2018-11-02 16:04:17 -07:00
Roger Labbe d78aaa25e4 Fixed particle animation 
I don't know why it wasn't displaying, but I reran the notebook
and it seems to render in the book now.
 2018-11-02 15:58:38 -07:00
Roger Labbe 9f89adc11f Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-11-02 15:50:29 -07:00
Roger Labbe eb5f7e1b17 Remove weight reset in update() 
As noted in Github #245, this should only happen after resampling, and
the resample functions already perform this computation.
 2018-11-02 15:50:10 -07:00
Roger Labbe 2225b249cd Remove weight reset in update() 
As noted in #245, this should only happen after resampling, and
the resample functions already perform this computation.
 2018-11-02 15:48:49 -07:00
Roger Labbe b59c2c3698 Remove default argument from discrete_bayes_sim 
This was a source of bugs. Github #259.
 2018-11-02 14:59:22 -07:00
Roger Labbe 00126c37f3 Cleaned up gaussian named tuple 
For some reason the code defined it in two places. I also added an
example for accessing individual fields.
 2018-11-02 14:55:40 -07:00
Roger Labbe 7e4925badd Error in code in text. 
Github #254. process_model  assignment erroneously multiplies
velocity by dt. This is only in the text so there is no error
in the code.
 2018-11-02 14:50:16 -07:00
Roger Labbe 20b2e38a22 Added line for actual weight trend line 
This is in response to pull request #257. That hard coded the
line, which meant it had no label in the legend. This adds the
trend line with a label.
 2018-11-02 14:42:38 -07:00
Roger Labbe 034f2b809a Merge pull request #256 from belm0/g_h_filter_example 
improve naming in ch 1 "predict_using_gain_guess" example

I (Roger Labbe) need to make further modifications to the function as explained above.
 2018-11-02 14:17:09 -07:00
Roger Labbe 5cb33dce02 Merge pull request #255 from LinkHS/fix_bug 
fix a bug of forget passing in 'hallway' when invoke 'discrete_bayes_…
 2018-11-02 14:03:53 -07:00
Roger Labbe c5804e3a28 Merge pull request #252 from PedroHRPBS/master 
fix: Fix typo
 2018-11-02 14:02:12 -07:00
Roger Labbe 216efe2ba4 Merge pull request #251 from off99555/master 
Change 10.90 to 11.35
 2018-11-02 14:01:17 -07:00
Roger Labbe 58721a7373 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-11-02 10:02:15 -07:00
Roger Labbe 80b8049e51 Normalize binary line ending 
Github #253. Binary files were corrupted by eol changes because
they were not denoted as binary in .gitattributes
 2018-11-02 10:01:11 -07:00
Roger Labbe 0d1badfee8 Normalize binary line ending 2018-11-02 09:59:58 -07:00
Roger Labbe 06246e9502 Normalize binary line ending 2018-11-02 09:59:21 -07:00
Roger Labbe c68b74b2c5 added binary attirbutes for binary files 2018-11-02 09:53:02 -07:00
Roger Labbe e2339a6ccb Updated files to ignore 2018-10-07 14:47:15 -07:00
Roger Labbe a2b0291d07 GitHub #250 Wrong EV for dice 
3.5, not 3.
 2018-09-19 07:27:36 -07:00
Roger Labbe bf7ad5d5f7 Github #235 2 used instead of z in a subscript 2018-09-16 10:15:42 -07:00
Roger Labbe 4e11dfcc81 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-09-16 10:10:03 -07:00
Roger Labbe fb110beeb0 Github #246 Pi computed incorrectly 
I incorrectly used area * pts/N to comput pi,which happens to
work if radius = 1, but fails otherwise.
 2018-09-16 10:08:46 -07:00
Roger Labbe fd03b1ddd7 Merge pull request #248 from off99555/master 
Several errors:

1. Update called with a move control input, which makes no sense.
2. velocity needed to be multiplied by dt, I was just lucky that dt = 1
3. Typo in code calling nonexistent function likelihood()
4. Fixed length of error bars to match subsequent text and code.
 2018-09-16 09:49:26 -07:00
Roger Labbe 5410867608 Merge pull request #247 from potpath/fix-07-Kalman-Filter-Math 
Fix a few typos
 2018-09-16 09:31:44 -07:00
Roger Labbe 70858d0e34 More improvements to Bayes' theorem coverage. 
Mostly rearranging descriptions from chapter 2 to 3, after I cover
probability distributions. I think some more work needs to be done,
mainly I think I go on for too long on pdfs, using first cars, then
temperatures, to explain the same thing.
 2018-08-29 19:39:14 -07:00
Roger Labbe 3270a5b686 Updated for filterpy version 1.4.4 
I changed the IMM filter to use the predict/update cycle, requiring
a code change in the Adaptive Filtering chapter.
 2018-08-24 08:42:46 -07:00
Roger Labbe 5512aceaef Altering book format 
Changed rest of chapters to use new formatting style. Also,
there is a bug in the current release of Jupyter that overwrites
the rcParams settings when you do the first import, so I had to
split the book format cell into two. This is actually better because
now the PDF, which strips out the formatting cell, will still have
the %matplotlib inline and import from future code.

Also, in the process of refactoring the information about Gaussian
and Bayes. Stuff is still strewn around too much. I need to make
this more streamlined. Chapters 2-4 talk about distributions, Bayes,
and Gaussians, and material is repeated a lot.
 2018-08-05 09:20:48 -07:00
Roger Labbe 909f107a93 Formatting changes, and removal of some text 
I am switching to the default look of Jupyter as much as possible.
Altering the look is not really supported by Jupyter, so rending
on GitHub (for example) is very different than when running in
your own browser.

I also got rid of the animations that really slow down execution
of the notebook in favor of using a slider to display a step in
a computation - the reader can see the plots 'animated' by moving
the slider. It is much faster to compute, and actually gives them
fine grained control to go forwards/backwards in time.

I am also revising my coverage of statistics, and moving the
details about Bayes and total conditional probability to the
third chapter. I shouldn't be talking about probability distributions
before covering that material.
 2018-08-04 10:27:26 -07:00
Roger Labbe edbc98ccb4 Merge pull request #243 from liorgolgher/patch-1 
Update 12-Particle-Filters.ipynb
 2018-08-03 08:32:51 -07:00
Roger Labbe 183b5b69ab Merge pull request #244 from jeremyroy/master 
Fixed typo
 2018-08-03 08:31:55 -07:00
Roger Labbe 0e43c1b9ae All line endings to LF from CRLF 2018-07-14 11:45:39 -07:00
Roger Labbe 59b7120c98 Reran to ensure uptodate with current python environment 
Python 3.6.5
 2018-07-14 10:58:49 -07:00
Roger Labbe 58f0f1cfab Fixed error in .gitattributes 2018-07-14 09:53:51 -07:00
Roger Labbe 2eb5b67ddf Fixed Pandoc warning 
I was using --chapter, which is deprecated by PanDoc.
 2018-07-14 08:47:32 -07:00
Roger Labbe 8507d8baae Removed Matplotlib deprecation warning 
needed to use new kwarg density.
 2018-07-14 08:46:40 -07:00
Roger Labbe c15a693aa6 GitHub #240 
Added missing word 'we'
 2018-07-14 08:41:07 -07:00
Roger Labbe d1425bd565 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-07-14 08:02:48 -07:00
Roger Labbe 68714a8eb8 Added gitattributes file 
This normalizes all line endings to LF, not CRLF.
 2018-07-14 08:01:19 -07:00
Roger Labbe ad231bef86 GitHub #241 - remove deprecation warning 
Newest version of Matplotlib has deprecated using strings
as booleans.
 2018-07-14 07:54:29 -07:00
Roger Labbe a28a75e1eb Merge pull request #242 from gwshaw/master 
Minor fixes
 2018-07-14 07:31:10 -07:00
Roger Labbe 47cbc4bf43 Fixed link to table of contents. #238 
Table of contents pointed to nbviewer, not the local copy of the
table of contents. Made link relative so it works regardless of
where you are working from.
 2018-07-04 17:17:53 -07:00
Roger Labbe 2480f297e5 Test fix for link to nbviewer. #238 
Trying to make a relative link to the table of contents so browsing
works both locally and remotely.
 2018-07-04 17:05:31 -07:00
Roger Labbe b87aa1ef45 Explained IMM requires all x to have same shape. 
IMM will not work if the x attribute in the filters have
different shapes.
 2018-06-24 10:10:30 -07:00
Roger Labbe be4c505251 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-06-12 16:26:09 -07:00
Roger Labbe c74ec6596b moved test functions to correct file. 
I had two test functions of the robot particle filter, but
oddly placed in the wrong file.
 2018-06-12 16:24:46 -07:00
Roger Labbe 0e108ab369 Merge pull request #230 from sleepingAgain/patch-1 
Fixed typo
 2018-06-11 08:38:27 -07:00
Roger Labbe b00adb2786 Spelling error 
Fixed misspelling of NEES, and removed \mathbf for epison, as it
doesn't render properly in the PDF.
 2018-05-11 10:33:25 -07:00
Roger Labbe 6eefb93968 Fixed chapter numbering 
Edit to book.tplx caused all chapters to be 0.x
 2018-05-08 18:21:31 -07:00
Roger Labbe 0c76f189b1 Fixed latex layout error 2018-05-08 17:47:04 -07:00
Roger Labbe a63e03e301 Added in/out back to cell output 
It was too hard to tell what was embedded sample code and what
was true cell input/output.
 2018-05-08 17:36:56 -07:00
Roger Labbe c3022b7654 Added table of contents 
Added table of contents to book. Now annoyingly lists a "contents"
section of two blank pages, but baby steps.
 2018-05-08 17:14:37 -07:00
Roger Labbe b5146e3a18 Modifed 1st example to use equations 
Previous version used dot(F, x) to compute the prior, which
doesn't look much different from the linear KF.
`
 2018-05-08 15:18:40 -07:00
Roger Labbe be136c1425 Added derivation of Joseph equation and optimal K 2018-05-06 08:36:39 -07:00
Roger Labbe e9fe8621a0 Updated for FilterPy 1.4.0 2018-05-05 19:09:21 -07:00
Roger Labbe 4787cf68e9 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-05-05 09:17:31 -07:00
Roger Labbe 335e6fe7fc Merge pull request #228 from tommz9/master 
Fix another typo in Chapter 8
 2018-05-05 09:17:12 -07:00
Roger Labbe 7feaab9ce2 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-05-05 09:16:04 -07:00
Roger Labbe d605763855 Added my own FloatSlider, IntSlider 
They create a slider with continuous_update=False. Just keeping
code more readable by avoiding long lines.
 2018-05-05 09:14:22 -07:00
Roger Labbe 20cc070525 Merge pull request #227 from tommz9/master 
Fix a typo in Chapter 8
 2018-05-04 09:03:55 -07:00
Roger Labbe 9ae79fa625 Added instructions for cloning the book 2018-04-29 08:08:40 -07:00
Roger Labbe 7963e3f15f minor wording changes. 2018-04-28 14:22:29 -07:00
Roger Labbe 49673c6bc2 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-04-28 11:07:51 -07:00
Roger Labbe 3ede7ff965 Added use of the scipy describe function 
Showed how to use describe() get the summary statistics for
a sampled distribution.
 2018-04-28 11:07:04 -07:00
Roger Labbe de5f6eda77 Merge pull request #222 from tthomto/master 
correcting the math in covariance example - missing fraction.
 2018-04-26 07:57:43 -07:00
Roger Labbe 47062b018d Fixed bad description of fat tails. GitHub #217 
Updated my description of the fat tails to be more general, especially
since my tails were truncated not fat!

Also added a list of Wikipedia links at the end for reference of
those reading the PDF.
 2018-04-25 17:05:53 -07:00
Roger Labbe 237ad1f9cb Updated train example, GitHub #214 
Updated the train example to perform the prediction
before the update, and changed some language.
 2018-04-25 17:04:33 -07:00
Roger Labbe 5a0cd1b098 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2018-04-25 15:29:43 -07:00
Roger Labbe ec0cc50ee9 Invalid matrix indexes, GitHub #211 
I used the wrong order for the matrix indexes, fortunately in
the text, not the code!
 2018-04-25 14:50:54 -07:00
Roger Labbe 4758f013e1 Merge pull request #216 from kmalhan/fix_process_model 
Fixed Unit of process_model for First Kalman Filter in Chapter 4
 2018-04-25 14:30:43 -07:00
Roger Labbe 5e6a503619 Updated for FilterPy 1.2.4, moved code around 
Moved code specific to one notebook to the *_internal.py file
for that book.

Updated to require FilterPy 1.2.4 due to a fix of a bad bug in the
UKF RTS Smoother, and the change of likelihood from a function back
to a property.
 2018-04-18 16:00:10 -07:00
Roger Labbe 7cbac1f983 My text improperly stated that the unscented transform takes a 
Gaussian as an input when of course it takes samples from any
arbitrary distribution.
 2018-03-29 16:54:35 -07:00
Roger Labbe 9b80d231e2 Wording improvements, and plot changes 2018-03-29 16:50:10 -07:00
Roger Labbe 4289a5d2a0 Wording changes. 2018-03-29 15:34:36 -07:00
Roger Labbe 7d295ffce5 Added Markov property example. 
I was getting a bit abstract describing Markov chains, so I gave a
concrete example of why it is a reasonable requirement.
 2018-02-10 08:47:21 -08:00
Roger Labbe 782119171d Fixed copy when directory not empty 
Was copying the kf_book subdirectory into the existing one
if it was already there, causing PDF build failure if you
didn't clean prior to running it.
 2018-02-02 10:52:05 -08:00
Roger Labbe 1ca3ff5e37 Fixed numpy set_printoptions 
for reasons I don't understand you have to call with suprres=True
and no othe rparameters for it to 'work', even though the
get_printoptions shows it as correctly set.
 2018-02-02 10:51:00 -08:00
Roger Labbe a71401a7b8 Fixed bulleted list 
I was using HTML, which doesn't render in the PDF. Switched to
markdown and it is better.
 2018-02-02 10:50:15 -08:00
Roger Labbe d86fb065cb Updated for FilterPy 1.2.0, added section 
Added a section on Mahalanobis distance and on gating.
Updated to require filterpy 1.2.0.
 2018-02-01 08:42:39 -08:00
Roger Labbe 0204c02123 Added section on state variable order 
I've alluded to the fact that you can order x however you want,
but didn't have an example.
 2018-01-31 11:16:44 -08:00
Roger Labbe acc59c240c Spelling error 2018-01-31 09:16:50 -08:00
Roger Labbe c473068238 Saved widget state 
I hadn't noticed that the output of the interactive widget calls
are not saved in the notebooks. This changed awhile ago. I've
explicitly saved them.
 2018-01-31 09:14:13 -08:00
Roger Labbe 6b28661f78 Grammar fix. 2018-01-31 07:25:23 -08:00
Roger Labbe 062438bc23 Added caveat about GitHub 
GitHub does not render the math well, so I warn people off from
using it to read the book.
 2018-01-31 05:27:04 -08:00
Roger Labbe 18b06e37dc Added some text 2018-01-30 06:30:20 -08:00
Roger Labbe ff9735d0a0 Merge pull request #207 from Pask2/patch-1 
Incorrect derivative in equation, and labelled a 1st order filter as 2nd order.
 2018-01-29 09:45:42 -08:00
Roger Labbe 15e8574e2e Text used the wrong parameter name for hist() call, and I misattributed 
a SciPy function to NumPy.
 2018-01-29 09:37:23 -08:00
Roger Labbe 423f3e80fe I wrote about the overlap of two measurements, but I really meant 
the gap between two measurements. The subsequent paragraph discussed
the overlap. I refined the language to make this clearer.
 2018-01-29 09:33:11 -08:00
Roger Labbe e8b88adf5f GitHub #154 Description of VAR inconsistent 
I stated that the heights are equally probable, which they are not.
 2018-01-29 09:27:24 -08:00
Roger Labbe 68ca5960cc Simplified namedtuple explanation. 2018-01-28 16:02:51 -08:00
Roger Labbe da4df463b6 Reran without debugging print out 
I had a print statement in kinematic.py, causing a spurious output
when importing from the common submodule.
 2018-01-28 12:05:18 -08:00
Roger Labbe 2775e76cbf Added gaussian named tuple. 
Use a named tuple instead of a tuple for gaussian so the code
is more readable.
 2018-01-28 11:42:35 -08:00
Roger Labbe 8a64d31f6c fixed PDF generation error 
for some reason the \pm latex was not working without a space before
and after the $ signs.
 2018-01-28 11:41:44 -08:00
Roger Labbe 86a7e2288c Merge pull request #205 from ipa-lth/master 
Misspelling of 'measurement'
 2018-01-26 07:40:28 -08:00
Roger Labbe ccaaa4b04b Merge pull request #204 from ernstklrb/typos-09ff 
Various typos
 2018-01-26 07:39:34 -08:00
Roger Labbe 992a70c0dd Removed extraneous matrix print out 
I don't know why, but the *_ notation ended up printing
the last matrix to the output cell, so I went back to _, _, _, _

I think I was just experiementing with that cell as the *_ notation
is not compatibile with Python 2.7
 2018-01-24 07:39:21 -08:00