daviddoji/Kalman-and-Bayesian-Filters-in-Python
1
0 Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,082 Commits 1 Branch 0 Tags
5ab83047d9a3ed82510d48ac1b82161576987148
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Roger Labbe 5ab83047d9 Updated to refer to the kf_book subdirectory. 2017-11-14 09:55:05 -08:00
Roger Labbe 88f03985bf Update README.md 
Added text mentioning the new Azure Notebook badge
 2017-10-20 13:25:54 -07:00
Roger Labbe a67bf5e8a6 Merge pull request #177 from crwilcox/patch-1 
Add launch in Azure Notebooks button
 2017-10-20 13:24:55 -07:00
Roger Labbe 868cc84155 Merge pull request #169 from ycchangtw/master 
Update 02-Discrete-Bayes.ipynb
 2017-09-08 10:05:15 -04:00
Roger Labbe d42a950349 Merge pull request #171 from sleepingAgain/master 
fixed typo
 2017-09-08 09:56:44 -04:00
Roger Labbe 80947df4f3 Merge pull request #168 from fubel/patch-1 
Typo correction
 2017-08-27 14:07:38 -04:00
Roger Labbe f77e2bf19b prcoess var should be used to compute prior, and sensor var 
for the posterior.
 2017-07-27 07:50:14 -07:00
Roger Labbe 40e2c8d830 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2017-07-27 07:48:57 -07:00
Roger Labbe 5a7257dbba Merge pull request #164 from PLeVasseur/04-intro-to-designing-filter 
In Introduction to Designing a Filter updated to use temp_change
 2017-07-27 07:47:30 -07:00
Roger Labbe e70190300a Merge pull request #163 from PLeVasseur/04-prediction-with-gaussians 
Fixed text describing Gaussian prediction calculation
 2017-07-27 07:37:03 -07:00
Roger Labbe 5c25643866 Deleted empty line in code. 2017-07-27 07:35:30 -07:00
Roger Labbe 6b3f902a4f Merge pull request #162 from peteryschneider/master 
Fixed estimates that were not tracking under 'Control Inputs' in 08-Designing-Kalman-Filters.ipynb
 2017-07-27 07:28:56 -07:00
Roger Labbe fea343cc11 Reran chapters. 2017-07-27 07:23:08 -07:00
Roger Labbe 1c11186fa4 Bug in train_filter 
Called normalize(belief), when it should have been normalize(prior).
belief is a variable from another cell, and not used in this function.

Reported by by Shakti Gehlaut.
 2017-07-15 12:08:47 -07:00
Roger Labbe ac1eb9d7fc Updated second binder link 
I altered the one at the top in the previous commit but not the one embedded deeper in the file.
 2017-07-03 13:57:34 -07:00
Roger Labbe f137b27646 Updated binder badge again 
This one follows instructions that I got in gitter from Yuvi Panda
 2017-07-03 13:53:21 -07:00
Roger Labbe b2a2b42a05 Altered binder badge 
Binder has changed it's link and host
 2017-07-03 13:45:24 -07:00
Roger Labbe 7c5aa45019 Weights should be reset to 1/N after resampling. 2017-07-03 13:07:28 -07:00
Roger Labbe ba61f3a96c Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2017-07-03 12:59:43 -07:00
Roger Labbe edb9b4f306 Text used sigma_ps when talking about the wheel. 2017-07-03 12:58:12 -07:00
Roger Labbe 1002744510 Merge pull request #158 from sleepingAgain/patch-1 
Fixed typo in Chapter 7
 2017-06-21 08:01:42 -07:00
Roger Labbe d15845bde8 I reversed the explanation of product/sum for Gaussians, saying the 
sum is proportional and the product is Normal. Of course the reverse
is true.
 2017-06-12 09:35:55 -07:00
Roger Labbe 529585a4c8 Issue #144 - update variable est/prediction names 
I erroneously named the prediction of x x_est, which is confusing at best.

There were also issues with plotsize. The %matplotlib notebook magic
is resetting the size of the plots. Not sure if this has something to do
with some of my plotting code or not. Anyway, I wrote a context manager
to reset the plot size after exiting interactive mode.
 2017-06-12 09:30:27 -07:00
Roger Labbe 000122a3ab Merge pull request #155 from jeffrimko/master 
Minor typo fixes.
 2017-06-12 08:17:54 -07:00
Roger Labbe 0a66218980 Merge pull request #153 from senden9/patch-2 
Fix formation in heading in README.md
 2017-06-12 08:16:06 -07:00
Roger Labbe 7fda1ccd01 Merge pull request #150 from dschuler/master 
Add 'Estimates' to legend

I added a comment header to the plotting routine, and renamed `retval` to `lines` (I prefer meaningful names whenever possible.
 2017-06-12 08:15:43 -07:00
Roger Labbe e16af3c383 Create book_plots.py 2017-06-12 08:14:39 -07:00
Roger Labbe e6f91acf79 Merge pull request #149 from alexwal/patch-1 
Fixed continuous Bayes' Rule formula
 2017-06-12 08:05:54 -07:00
Roger Labbe 04f8235c43 I was multiply the noise generated by randn() by dt, which makes 
no sense.
 2017-05-11 13:27:05 -07:00
Roger Labbe d2f247da20 I dropped two square terms in the derivation for the sum of 
Gaussians; the remaining math was correct.
 2017-05-11 13:24:35 -07:00
Roger Labbe 99193f7e26 this was just in text, but I used self.W[i], and it should have been 
self.Wc[i].
 2017-05-11 13:20:37 -07:00
Roger Labbe e82f2184d8 Merge pull request #136 from asmwarrior/patch-1 
Update 03-Gaussians.ipynb
 2017-05-11 13:18:26 -07:00
Roger Labbe 5a0e84f5fc Merge pull request #142 from horaceheaven/horaceheaven-patch 
Fix readme typo
 2017-05-11 13:18:00 -07:00
Roger Labbe 171b8649c2 Merge pull request #143 from horaceheaven/patch-1 
Fix: 'to' missing from sentence
 2017-05-11 13:16:32 -07:00
Roger Labbe 3d288d60d6 The base python installation has a file code.py; this causes 
problems when trying to import files from the code subdirectory.

I also tried to rearange some code so that book_format does not need
to be imported if you are not running in a Jupyter Notebook.
 2017-05-11 13:13:17 -07:00
Roger Labbe 3639394a12 Slight rewording for clarity. 2017-01-18 12:24:57 -08:00
Roger Labbe c97186c8b4 Fixed PDF generation re %matplotlib magic 
Fixed the scripts that comment out the matplotlib magic.
It was commenting out all of the magics - the first one, and
ones in markdown cells. It should only comment out the ones
in code cells, and leave the first one.
 2017-01-18 12:23:06 -08:00
Roger Labbe 91e9e08377 Removed #reset from magic 
I was using comment to detect which %matplotlib inline to comment
out to generate the PDF but it generates a UsageError. I need to make
my script smarter instead. The problem is that I don't want to remove
the first inline definition
 2017-01-18 10:59:43 -08:00
Roger Labbe f0e78c7037 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2017-01-18 09:21:43 -08:00
Roger Labbe 6bafbd490e I switched the weight gain guess back to 1 (the corrct value - of 
course the filter would find the right answer. Leaving at -1 shows
that the filter works well even when the initial estimate is incorrect.
 2017-01-18 09:20:15 -08:00
Roger Labbe ebb8bceeab Add SpaceX reference to readme 2017-01-16 08:33:18 -08:00
Roger Labbe 6dc3dbaef9 Text had ```Python (capitalized), which renders fine in the local 
notebook, nbviewer, and elsewhere, but not on github for some reason.
 2017-01-14 15:11:32 -08:00
Roger Labbe 8463014958 Disable interactive plots for PDF. 
nbconvert cannot output interactive plots to PDF. I wrote code
to remove all of the %matplotlib notebook magic, rerun the notebooks,
and then create the PDF.
 2017-01-14 13:15:24 -08:00
Roger Labbe e2405b23d5 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2017-01-14 09:05:11 -08:00
Roger Labbe fbd2655dc3 Typo fix. 2017-01-14 09:05:00 -08:00
Roger Labbe b161c6d35c Merge pull request #130 from senden9/patch-1 
Fix typo in README.md
 2017-01-14 09:03:52 -08:00
Roger Labbe c3c5616eac Merge pull request #129 from marcae777/master 
Fixed formula in Kalman Filter Math
 2017-01-14 09:03:35 -08:00
Roger Labbe 9c8b113259 Merge pull request #128 from MattHeard/master 
Correct minor typo
 2017-01-14 09:02:21 -08:00
Roger Labbe e4b05feabf Merge pull request #126 from sauravrt/master 
Chapter 7 Fix LTI system theory equation for phi(t)
 2017-01-14 09:01:38 -08:00
Roger Labbe 360eccb371 Minor spelling correction. 2016-12-09 12:24:37 -08:00
Roger Labbe 5bac8fbbbc Fixed formatting to generate book. 
Some tables had two header specifications for 3 columns. Rendered
in the browswer, but not in latex. Not sure how that happened,
I don't think I changed that code and the pdf used to build.
 2016-11-08 06:55:31 -08:00
Roger Labbe bfdc23fa77 Checked in a temporary test file by mistake 2016-11-02 08:39:38 -07:00
Roger Labbe 83f6795f72 Wording changes. 2016-11-02 08:38:18 -07:00
Roger Labbe 55d73b21de Improved installation instructions. 2016-09-25 12:46:23 -07:00
Roger Labbe 4b93469205 correction: lets to let's. 
grammar is our friend.
 2016-09-25 12:36:14 -07:00
Roger Labbe ae45efc403 Updated binder instructions. 
It is not experimental, it is the best way to use online. Also,
removed the mention of sagemath, that is too complicated to use.
 2016-09-25 11:43:49 -07:00
Roger Labbe c2548702d9 Some work on HTML conversion. 
Not happy with the output yet. The advantage of HTML output is
interactive plots are displayed in the output, whereas they are
not in the PDF output.
 2016-09-25 11:31:48 -07:00
Roger Labbe dbdf3541f2 Updated to specify Jupyter, not ipython. 2016-09-25 11:30:01 -07:00
Roger Labbe e289a2841f Minor word smithing. 2016-09-25 10:10:53 -07:00
Roger Labbe 91e603c592 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2016-09-23 12:49:37 -07:00
Roger Labbe 0b358aca26 Merge pull request #123 from ASunday47/patch-2 
Update 07-Kalman-Filter-Math.ipynb
 2016-09-23 12:49:25 -07:00
Roger Labbe 30874bae62 Issue #122. 
Description of ellipse shape was incorrect.

While I was changing this, I also alterted the plots to make them
more readalbe.
 2016-09-23 12:08:39 -07:00
Roger Labbe b88a05d307 Merge pull request #119 from SonienTaegi/patch-1 
mkf_internal.py : Fix duplicated parameter in plt.scatter(...)
 2016-09-20 11:23:12 -07:00
Roger Labbe 81ab95d5a9 Slighly improved discussion of UKF weights. 
I wrongly stated that weights must sum to one. they don't.
I clarified this, but did not explain *why* Merwe weights are so large,
and/or negative. this needs to be improved.
 2016-09-20 07:40:27 -07:00
Roger Labbe 2743b78db5 Removed temporary files and test code. 
I made some changes while testing the plotting change in the last checkin,
and it inadvertantly got checked in.
 2016-09-20 06:56:44 -07:00
Roger Labbe 96749eb378 Fixed interactive plots for PDF. 
the plots generated by interactive plots (%matplotlib notebook) do
not show up in the PDF. This has meant that I haven't been able to
update the PDF since febuary. I switched the default plotting
mechanism to noninteractive.
 2016-09-20 06:49:15 -07:00
Roger Labbe 97f3f64efb Fixed plots to work in PDF 
The interactive plots do not display in the generated PDF. I have
fixed this so we can generate the PDF again.
 2016-09-12 20:04:53 -07:00
Roger Labbe b28158e6dd Last checkin ran with %matplotlib inline. 
That was test code, it should be %matplotlib notebook
 2016-09-12 18:15:08 -07:00
Roger Labbe 68b2ee27f9 Reran cells to generate plots 2016-09-12 18:12:11 -07:00
Roger Labbe 681e19ba47 Issue #117. Invalid shortened URL. 2016-09-12 18:07:14 -07:00
Roger Labbe 32848496e7 Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python 2016-09-12 17:39:30 -07:00
Roger Labbe b330aacc4b Issue #115. Wording error. 
'measuremet' was used; it should have been 'velocity'.
 2016-09-12 17:37:50 -07:00
Roger Labbe c246aa0acb Bug where animation not cleared between frames 2016-09-12 17:36:30 -07:00
Roger Labbe 9d95bd0714 Merge pull request #116 from esvhd/nb05_typo_fix 
Fixed two typos in notebook 5 Multivariate Gaussians.
 2016-09-12 17:34:10 -07:00
Roger Labbe 19deda37ee Issue #112 interactive_gh uses global data 
interactive_gh() uses zs, which is global data, so if you run
another cell that assigns to z, then drag the sliders without
rerunning the cell that uses interactive_gh(), you get the
reassigned zs value.
 2016-08-28 15:19:37 -07:00
Roger Labbe 2128710f41 Normalized with Preface. 
The README contains text esssentially identical to the preface. However,
there were some notable differences, which I have fixed.
 2016-08-28 14:42:02 -07:00
Roger Labbe 7b6151e1e4 Reran cells. 
Many of the plots were not generated, which makes the chapter
unreadable on github.
 2016-08-28 12:54:50 -07:00
Roger Labbe e9145dbc62 Deleted obsolete file particle_filter.py 2016-08-28 12:48:47 -07:00
Roger Labbe afab63ec2f Issue #103. Completed unfinished section. 
The first section of the README was incomplete.
 2016-08-28 12:23:33 -07:00
Roger Labbe 0ce8792340 Reran cells - checked in version had no plots. 2016-08-28 11:45:22 -07:00
Roger Labbe cab54e11d6 Issue #95. Incorrect notation for process model. 
I used the Fundaental matrix F when I should have used the
system dynamics matrix A.
 2016-08-28 11:00:45 -07:00
Roger Labbe d31721aa0f Discrete_bayes_sim() wasn't displaying the graphs because plt.show() 
was missing from the end of the function.
 2016-08-28 08:12:03 -07:00
Roger Labbe 6914c3be77 Merge pull request #113 from neonquill/fix_typos_05 
Fix a couple of typos found in Chapter 5.
 2016-08-28 07:45:42 -07:00
Roger Labbe 044063363c Merge pull request #111 from neonquill/fix_train_filter 
Fix train example in 02-Discrete-Bayes.
 2016-08-27 08:09:41 -07:00
Roger Labbe a6cba51c0b Merge pull request #110 from Qx5/patch-1 
Missing transpose in text
 2016-08-27 08:01:35 -07:00
Roger Labbe 4328c397d4 Merge pull request #109 from esvhd/nb4-fix 
Added missing closing bracket.
 2016-08-27 07:59:15 -07:00
Roger Labbe 489084aa8f Merge pull request #102 from plusk01/lesson01_fixes 
Lesson 01 Content Fixes
 2016-08-27 07:58:35 -07:00
Roger Labbe 45cf59df72 Deleted reference to 6x9 version 2016-07-07 21:14:03 -07:00
Roger Labbe 5017434494 Updated link to PDF 2016-07-07 21:12:32 -07:00
Roger Labbe 6986b66f87 Merge pull request #104 from asfaltboy/patch-1 
Simple grammar correction from context
 2016-07-07 19:33:43 -07:00
Roger Labbe 4669507d7a Fixed computation of epoch 
Section changed the time delta from 2 seconds to 1, but did not
note that in the text.
 2016-06-25 15:40:53 -07:00
Roger Labbe 5fcdf5af47 Minto typos. 
Courtesy of email from James Kim.
 2016-06-22 19:53:33 -07:00
Roger Labbe b370e9a9a4 Typo fix. 
lubkoll offered a commit, but I got a conflict and it was easier to
make the change myself.
 2016-06-16 12:38:08 -07:00
Roger Labbe 0239608ab2 Issue #101. Missing plots in chapter. 2016-06-16 12:35:47 -07:00
Roger Labbe be073785c9 Revert "Merge pull request #99 from lubkoll/master" 
This reverts commit 4b17de67a9, reversing
changes made to 5559d2e03f.
 2016-05-25 10:48:47 -07:00
Roger Labbe 4b17de67a9 Merge pull request #99 from lubkoll/master 
Fix typos in chapter 08
 2016-05-25 08:39:15 -07:00
Roger Labbe 5559d2e03f Merge pull request #98 from lubkoll/master 
Fix typos in chapter 04
 2016-05-25 07:13:28 -07:00
Roger Labbe 510e720559 I used terms better suited to radars (elevation angle instead 
of bearing, for example).
 2016-05-19 18:39:26 -07:00
Roger Labbe a79153205b Issue 87. Example was linear. 
I provided  a linear example for what was supposed to be a
nonlinear example.
 2016-05-19 18:37:46 -07:00
Roger Labbe d89e35bbdf Fixed latex raising ^* 
It works in mathjax, but the pdf convertor cannot handle it. Changing
it to ^{*} works.
 2016-04-23 04:30:03 -07:00