Donovan Rynne
914dc171f5
fix bugs where shuffle and rotate strategy where switched
...
Also fixed rotate surrogate selection, list was being rotate continuously rather than rotating the initial index list, which nullified the dyncorrexcl setting.
2024-05-18 14:24:22 +10:00
Donovan Rynne
9643b1ce6b
update autoanalyser for new theiler window handling
2024-04-28 14:24:12 +10:00
Donovan Rynne
6bf7954d4e
modify generateRotatedSurrogates for theiler window exclusion
2024-04-28 14:23:48 +10:00
Donovan Rynne
484421bf3c
refactor to bring DYN_CORR_EXCL from kernel and kraskov into common
2024-04-28 14:23:10 +10:00
Donovan Rynne
c1225d67c7
add surrogate type to common getProperty
2024-04-20 15:33:11 +10:00
Donovan Rynne
401df8ee45
add surrogate type to AutoAnalyserMI
2024-04-20 11:09:24 +10:00
Donovan Rynne
2c33e36752
add new surrogate types formatting
2024-04-20 11:09:24 +10:00
Donovan Rynne
a3cdcc04b5
add surrogate_type checking into computeSignificance, surrogate property plumbing in interface class
2024-04-20 11:09:23 +10:00
Donovan Rynne
191a81f1bc
add generateRotatedSurrogates method in RandomGenerator
2024-04-20 11:09:22 +10:00
Donovan Rynne
037159cf86
rename numPermutationsToCheck to numSurrogatesToCheck
2024-03-24 19:01:30 +11:00
Joseph Lizier
2130985c34
Adding/implementing interfaces for continuous CMI estimators to allow all combinations of univariate/multivariate arrays to be passed for var1/var2/conditional
2023-10-27 10:04:25 +11:00
Joseph Lizier
4b85a7fd46
AIS and TE estimators (for KSG in particular) utilising the underlying NN Searchers holding the different data sets separately.
2023-10-09 13:16:27 +11:00
Joseph Lizier
ec8e7f848f
CMI common estimators utilising underlying nearest neighbour searchers holding different data sets separately (for dynamic correlation exclusion)
2023-10-09 13:15:56 +11:00
Joseph Lizier
5f87e2018b
MI common estimators utilising underlying nearest neighbour searchers holding different data sets separately (for dynamic correlation exclusion)
2023-10-09 13:14:29 +11:00
Joseph Lizier
938dd7c4fb
Implementing multiple data sets in the nearest neighbour searches, to facilitate dynamic correlation exclusion only for points within the same data set
2023-10-09 13:12:07 +11:00
Joseph Lizier
a554361de9
Adding unit test for no conditional on CMI Gaussian
2023-10-06 12:29:37 +11:00
Joseph Lizier
90baf68ee7
Making AutoAnalyser Python code only start JVM if not already started
2023-08-21 17:55:16 +10:00
Joseph Lizier
817abd1de0
For Autoanalyser Python code adding \ at the end of line for print statement
2023-08-15 16:28:04 +10:00
jlizier
865d82630a
Altering AutoAnalyser generated python code to set flag convertStrings=True in JVM start, as per PR #94
2022-09-08 10:20:20 +10:00
jlizier
2b0a5cb943
Updated mixed-continiuous CMI KSG estimator, for the bias correction based on number of continuous variables. This was identified from testing, and aligns with testing so far.
2022-09-05 17:51:49 +10:00
Joseph Lizier
d750a737be
Removing what appears to be an old test file from the integrated Spiking TE code from David
2022-09-05 17:08:10 +10:00
Joseph Lizier
27f9cccf65
Don't need code for old way of doing spiking TE estimation
2022-09-05 14:45:50 +10:00
Joseph Lizier
47be5df74b
editing comment on how to implement dynamic correlation exclusions in spiking TE estimator
2022-09-05 14:43:18 +10:00
Joseph Lizier
e747e8aa93
Merge pull request #93 from dpshorten/master
...
Spike train TE estimation
2022-09-05 14:38:21 +10:00
jlizier
05248ef105
Extra fixing windows paths generated from AutoAnalyser to be properly escaped in data filename for Python/Matlab
2022-09-01 15:50:27 +10:00
jlizier
c233485c90
Fixing windows paths generated from AutoAnalyser to be properly escaped
2022-09-01 15:28:28 +10:00
jlizier
2714650cfa
Amending Kraskov (KSG) MI estimators to have an experimental method to provide conditional entropy of the first variable given the second. Works by removing the Kozachenko Leonenko entropy of variable 1 from the MI, using the same kNN radii as the MI estimator. Includes Unit tests to provide some initial validation.
2022-06-16 21:29:29 +10:00
jlizier
25159009ca
updating max entropy discretisation to gracefully handle NaNs in the input (which shouldn't be there anyway)
2022-06-09 12:39:59 +10:00
David Shorten
c068c8308d
improving jittered surrogates
2022-02-25 18:49:12 +11:00
David Shorten
285c95947a
integrating artemis changes
2022-02-22 14:39:15 +11:00
David Shorten
b4e32727c9
merge
2022-02-22 11:55:09 +11:00
Joseph Lizier
0e33393cdc
Merge pull request #86 from pmediano/master
...
Thanks Pedro, and sorry to take so long to attend to this
2021-11-15 12:08:57 +11:00
Joseph Lizier
b02760ddd3
Adding overloads on setObservations and addObservations for MI and TE calculators to handle one variable univariate and one variable multivariate
2021-11-15 10:58:59 +11:00
Joseph Lizier
7d7291fb9d
Implementing Analytic null distributions for discrete conditional TE estimator
2021-11-02 23:11:45 +11:00
Joseph Lizier
6f43dfb620
Altering the python-java jpype data conversion in AutoAnalyser to work properly with jpype 0.7 and upwards
2021-10-08 16:03:14 +11:00
Joseph Lizier
21ae0bdeaa
Fixing logic error in printing whether GPU calculation was falling back to CPU
2021-10-08 11:28:37 +11:00
Joseph Lizier
650159661d
Patching conditional MI calculator to properly allow null or empty conditionals if dimension of conditionals is 0
2021-09-06 10:23:31 +10:00
David Shorten
c8d17e9522
functionality allowing specific embedding intervals
2021-08-23 17:50:34 +10:00
Joseph Lizier
aa8148eca9
Updates to KL entropy estimator: removing NORMALISE property, and adding method for cross entropy
2021-08-23 09:09:49 +10:00
David Shorten
dd215fdfbb
spiking surrogates implemented
2021-08-11 09:20:35 +10:00
Joseph Lizier
ef7cfbb525
Fixed missing " in comment
2021-08-06 17:43:56 +10:00
David Shorten
dc85001ff7
refactoring + conditional processes + radius sharing + euclidean norm working
2021-08-02 22:06:07 +10:00
David Shorten
9a17361cca
minor edits
2021-07-26 21:34:22 +10:00
David Shorten
dc226be60a
minor refactorings
2021-07-25 17:47:27 +10:00
David Shorten
8d7c598022
better testing script and a touch of refactoring
2021-07-25 17:30:30 +10:00
David Shorten
cf5f8edf5e
confirmed working on canonical ex
2021-07-25 16:29:31 +10:00
David Shorten
4ece80b5ef
appear to have 4kl estimator working
2021-07-24 23:32:57 +10:00
David Shorten
bb51b09048
converting embedding vectors to arrays for the trees
2021-07-23 17:54:46 +10:00
David Shorten
0e530cdceb
history embeddings done
2021-07-23 17:26:18 +10:00
David Shorten
deb9d97520
advancing previous indices throughout iteration
2021-07-22 19:06:53 +10:00