Use fftconvolve instead of np.correlate in cross_correlation_histogram #273
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Need to look at |
|
#111 |
|
The implementation of |
pbouss
pushed a commit
to INM-6/elephant
that referenced
this pull request
Nov 25, 2019
commit 59e6412 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Wed Nov 20 09:46:54 2019 +0100 Feature/spade duration (NeuralEnsemble#263) * Corrected error message * Doc. for Holm-Bonferroni * Changed calculation of duration for output of patterns * fixed calculation of duration for the signature * Pass spectrum to concepts_to_output function * Bug fixed in vairable assignment: concept commit de077d6 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Nov 19 16:17:52 2019 +0100 Use fftconvolve instead of np.correlate in cross_correlation_histogram (NeuralEnsemble#273) * cch np.correlate -> fftconvolve * moved internal functions in cross_correlation_histogram to a separate class * added BinnedSpikeTrain.get_num_of_spikes() function * added BinnedSpikeTrain.binarize() function commit 059ccf5 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Nov 15 16:24:58 2019 +0100 Homogeneous Poisson Process with refr. period (NeuralEnsemble#261) commit f378f07 Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Tue Nov 12 10:23:29 2019 +0100 Feature timescale (NeuralEnsemble#271) * Implementation of timescale calculation * Use cross_corr_coef of CCH function * Check that tau_max is an integer multiple of binsize commit 82f307f Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 11:09:44 2019 +0100 fixed BinnedSpikeTrain.bin_edges property (NeuralEnsemble#257) * fixed BinnedSpikeTrain.bin_edges property * bin_edges: include t_stop if spiketrain duration is divisible by binsize * BinnedSpikeTrain: warn if binning discards the spikes in the last (excluded) bin * reuse is_binary func from utility.py commit bd684f5 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 10:14:34 2019 +0100 Refactored unitary event (NeuralEnsemble#251) * completely removed misleading N arg - num. of neurons, which is always extracted from the data * fixed inverse_hash_from_pattern for long N > int64; added test_hash_inverse_longpattern * UE: deprecation warning commit 2967dac Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Fri Nov 8 16:38:27 2019 +0100 Calculate spike train auto-correlation time ('timescale') (NeuralEnsemble#266) * Implementation of timescale calculation commit 092d2fb Author: Michael Denker <m.denker@fz-juelich.de> Date: Fri Nov 8 12:33:32 2019 +0100 Added a simplified to correctly handle division with Compound Units (NeuralEnsemble#270) .magnitude -> .simplified.magnitude commit 5dfa042 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 31 15:54:30 2019 +0100 elephant v0.6.4 [workshop] (NeuralEnsemble#268) commit 5b5a023 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Oct 22 16:10:16 2019 +0200 neo v0.8.0 compatible (NeuralEnsemble#256) * fixes for neo v0.8.0 * upd array_annotation comment * added todo for neo empty array_annotations commit eac914c Author: PaulinaDabrowska <p.dabrowska@fz-juelich.de> Date: Tue Oct 22 16:09:41 2019 +0200 Improved doc string of covariance() and corrcoef() (NeuralEnsemble#260) * corrected explanation of the formula used in corrcoef and covariance (in the doc string) * pep8 commit 434cacd Author: Robin Gutzen <robin.gutzen@live.de> Date: Mon Oct 21 13:56:38 2019 +0200 keep array_annotations in the output of signal_processing functions (NeuralEnsemble#258) * pass array_annotations on to output of signal_processing functions * test keep array_annotations for zscore, butter, hilbert commit 15acc90 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 17 12:19:51 2019 +0200 removed unused debug code in spade/fast_fca (NeuralEnsemble#249) * removed dead code * small speedup commit ce9e7ac Author: Björn Müller <bj.mueller@fz-juelich.de> Date: Wed Oct 16 10:25:09 2019 +0200 Performance and API enhancement for STTC (NeuralEnsemble#244) * Speedup for STTC * Implemented spike search in other sptr with numpy search Memory complexity: O(max(n1, n2)) (exactly max(n1, n2) is used additionally at peak) Runtime complexity: O(max(n1, n2)*log(max(n1, n2))), because search happens n1 times with log(n2) for binary search (or n2 times with log(n1)) commit a833a96 Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Wed Oct 9 15:21:41 2019 +0200 [spade] fixed the calculation of the duration of a pattern in the output (NeuralEnsemble#254) * fixed the calculation of the duration of a pattern, in the output of spade. Previously, the calculation of the pattern duration was not always correct, depending on the order of lags and neurons. We fixed the bug by accounting the duration independently from the order of neurons; * added raise ValueErrors in code when the output_format is not one of the two parameters allowed * fixed bug of undefined spectrum type when running spade without surrogates (only FIM) commit c78af7e Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Thu Sep 12 16:15:43 2019 +0200 Updated authors list - alessandra and peter
pbouss
pushed a commit
to INM-6/elephant
that referenced
this pull request
Dec 3, 2019
commit b116804 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 25 12:31:36 2019 +0100 Fast covariance() and Pearson corrcoef() (NeuralEnsemble#274) clarified the difference in the implementation of CrossCorrHist.cross_corr_coef() and the reference book "Analysis of parallel spike trains", 2010 commit 59e6412 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Wed Nov 20 09:46:54 2019 +0100 Feature/spade duration (NeuralEnsemble#263) * Corrected error message * Doc. for Holm-Bonferroni * Changed calculation of duration for output of patterns * fixed calculation of duration for the signature * Pass spectrum to concepts_to_output function * Bug fixed in vairable assignment: concept commit de077d6 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Nov 19 16:17:52 2019 +0100 Use fftconvolve instead of np.correlate in cross_correlation_histogram (NeuralEnsemble#273) * cch np.correlate -> fftconvolve * moved internal functions in cross_correlation_histogram to a separate class * added BinnedSpikeTrain.get_num_of_spikes() function * added BinnedSpikeTrain.binarize() function commit 059ccf5 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Nov 15 16:24:58 2019 +0100 Homogeneous Poisson Process with refr. period (NeuralEnsemble#261) commit f378f07 Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Tue Nov 12 10:23:29 2019 +0100 Feature timescale (NeuralEnsemble#271) * Implementation of timescale calculation * Use cross_corr_coef of CCH function * Check that tau_max is an integer multiple of binsize commit 82f307f Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 11:09:44 2019 +0100 fixed BinnedSpikeTrain.bin_edges property (NeuralEnsemble#257) * fixed BinnedSpikeTrain.bin_edges property * bin_edges: include t_stop if spiketrain duration is divisible by binsize * BinnedSpikeTrain: warn if binning discards the spikes in the last (excluded) bin * reuse is_binary func from utility.py commit bd684f5 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 10:14:34 2019 +0100 Refactored unitary event (NeuralEnsemble#251) * completely removed misleading N arg - num. of neurons, which is always extracted from the data * fixed inverse_hash_from_pattern for long N > int64; added test_hash_inverse_longpattern * UE: deprecation warning commit 2967dac Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Fri Nov 8 16:38:27 2019 +0100 Calculate spike train auto-correlation time ('timescale') (NeuralEnsemble#266) * Implementation of timescale calculation commit 092d2fb Author: Michael Denker <m.denker@fz-juelich.de> Date: Fri Nov 8 12:33:32 2019 +0100 Added a simplified to correctly handle division with Compound Units (NeuralEnsemble#270) .magnitude -> .simplified.magnitude commit 5dfa042 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 31 15:54:30 2019 +0100 elephant v0.6.4 [workshop] (NeuralEnsemble#268) commit 5b5a023 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Oct 22 16:10:16 2019 +0200 neo v0.8.0 compatible (NeuralEnsemble#256) * fixes for neo v0.8.0 * upd array_annotation comment * added todo for neo empty array_annotations commit eac914c Author: PaulinaDabrowska <p.dabrowska@fz-juelich.de> Date: Tue Oct 22 16:09:41 2019 +0200 Improved doc string of covariance() and corrcoef() (NeuralEnsemble#260) * corrected explanation of the formula used in corrcoef and covariance (in the doc string) * pep8 commit 434cacd Author: Robin Gutzen <robin.gutzen@live.de> Date: Mon Oct 21 13:56:38 2019 +0200 keep array_annotations in the output of signal_processing functions (NeuralEnsemble#258) * pass array_annotations on to output of signal_processing functions * test keep array_annotations for zscore, butter, hilbert commit 15acc90 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 17 12:19:51 2019 +0200 removed unused debug code in spade/fast_fca (NeuralEnsemble#249) * removed dead code * small speedup commit ce9e7ac Author: Björn Müller <bj.mueller@fz-juelich.de> Date: Wed Oct 16 10:25:09 2019 +0200 Performance and API enhancement for STTC (NeuralEnsemble#244) * Speedup for STTC * Implemented spike search in other sptr with numpy search Memory complexity: O(max(n1, n2)) (exactly max(n1, n2) is used additionally at peak) Runtime complexity: O(max(n1, n2)*log(max(n1, n2))), because search happens n1 times with log(n2) for binary search (or n2 times with log(n1)) commit a833a96 Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Wed Oct 9 15:21:41 2019 +0200 [spade] fixed the calculation of the duration of a pattern in the output (NeuralEnsemble#254) * fixed the calculation of the duration of a pattern, in the output of spade. Previously, the calculation of the pattern duration was not always correct, depending on the order of lags and neurons. We fixed the bug by accounting the duration independently from the order of neurons; * added raise ValueErrors in code when the output_format is not one of the two parameters allowed * fixed bug of undefined spectrum type when running spade without surrogates (only FIM) commit c78af7e Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Thu Sep 12 16:15:43 2019 +0200 Updated authors list - alessandra and peter
dizcza
added a commit
to INM-6/elephant
that referenced
this pull request
Dec 10, 2019
commit 0d44cd7 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Dec 5 10:25:48 2019 +0100 unitary event decorate_deprecated_N refactoring (NeuralEnsemble#278) commit b116804 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 25 12:31:36 2019 +0100 Fast covariance() and Pearson corrcoef() (NeuralEnsemble#274) clarified the difference in the implementation of CrossCorrHist.cross_corr_coef() and the reference book "Analysis of parallel spike trains", 2010 commit 59e6412 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Wed Nov 20 09:46:54 2019 +0100 Feature/spade duration (NeuralEnsemble#263) * Corrected error message * Doc. for Holm-Bonferroni * Changed calculation of duration for output of patterns * fixed calculation of duration for the signature * Pass spectrum to concepts_to_output function * Bug fixed in vairable assignment: concept commit de077d6 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Nov 19 16:17:52 2019 +0100 Use fftconvolve instead of np.correlate in cross_correlation_histogram (NeuralEnsemble#273) * cch np.correlate -> fftconvolve * moved internal functions in cross_correlation_histogram to a separate class * added BinnedSpikeTrain.get_num_of_spikes() function * added BinnedSpikeTrain.binarize() function commit 059ccf5 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Nov 15 16:24:58 2019 +0100 Homogeneous Poisson Process with refr. period (NeuralEnsemble#261) commit f378f07 Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Tue Nov 12 10:23:29 2019 +0100 Feature timescale (NeuralEnsemble#271) * Implementation of timescale calculation * Use cross_corr_coef of CCH function * Check that tau_max is an integer multiple of binsize commit 82f307f Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 11:09:44 2019 +0100 fixed BinnedSpikeTrain.bin_edges property (NeuralEnsemble#257) * fixed BinnedSpikeTrain.bin_edges property * bin_edges: include t_stop if spiketrain duration is divisible by binsize * BinnedSpikeTrain: warn if binning discards the spikes in the last (excluded) bin * reuse is_binary func from utility.py commit bd684f5 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 10:14:34 2019 +0100 Refactored unitary event (NeuralEnsemble#251) * completely removed misleading N arg - num. of neurons, which is always extracted from the data * fixed inverse_hash_from_pattern for long N > int64; added test_hash_inverse_longpattern * UE: deprecation warning commit 2967dac Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Fri Nov 8 16:38:27 2019 +0100 Calculate spike train auto-correlation time ('timescale') (NeuralEnsemble#266) * Implementation of timescale calculation commit 092d2fb Author: Michael Denker <m.denker@fz-juelich.de> Date: Fri Nov 8 12:33:32 2019 +0100 Added a simplified to correctly handle division with Compound Units (NeuralEnsemble#270) .magnitude -> .simplified.magnitude commit 5dfa042 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 31 15:54:30 2019 +0100 elephant v0.6.4 [workshop] (NeuralEnsemble#268) commit 5b5a023 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Oct 22 16:10:16 2019 +0200 neo v0.8.0 compatible (NeuralEnsemble#256) * fixes for neo v0.8.0 * upd array_annotation comment * added todo for neo empty array_annotations commit eac914c Author: PaulinaDabrowska <p.dabrowska@fz-juelich.de> Date: Tue Oct 22 16:09:41 2019 +0200 Improved doc string of covariance() and corrcoef() (NeuralEnsemble#260) * corrected explanation of the formula used in corrcoef and covariance (in the doc string) * pep8 commit 434cacd Author: Robin Gutzen <robin.gutzen@live.de> Date: Mon Oct 21 13:56:38 2019 +0200 keep array_annotations in the output of signal_processing functions (NeuralEnsemble#258) * pass array_annotations on to output of signal_processing functions * test keep array_annotations for zscore, butter, hilbert commit 15acc90 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 17 12:19:51 2019 +0200 removed unused debug code in spade/fast_fca (NeuralEnsemble#249) * removed dead code * small speedup commit ce9e7ac Author: Björn Müller <bj.mueller@fz-juelich.de> Date: Wed Oct 16 10:25:09 2019 +0200 Performance and API enhancement for STTC (NeuralEnsemble#244) * Speedup for STTC * Implemented spike search in other sptr with numpy search Memory complexity: O(max(n1, n2)) (exactly max(n1, n2) is used additionally at peak) Runtime complexity: O(max(n1, n2)*log(max(n1, n2))), because search happens n1 times with log(n2) for binary search (or n2 times with log(n1)) commit a833a96 Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Wed Oct 9 15:21:41 2019 +0200 [spade] fixed the calculation of the duration of a pattern in the output (NeuralEnsemble#254) * fixed the calculation of the duration of a pattern, in the output of spade. Previously, the calculation of the pattern duration was not always correct, depending on the order of lags and neurons. We fixed the bug by accounting the duration independently from the order of neurons; * added raise ValueErrors in code when the output_format is not one of the two parameters allowed * fixed bug of undefined spectrum type when running spade without surrogates (only FIM) commit c78af7e Author: Alessandra Stella <a.stella@fz-juelich.de> Date: Thu Sep 12 16:15:43 2019 +0200 Updated authors list - alessandra and peter
dizcza
added a commit
to INM-6/elephant
that referenced
this pull request
Jan 9, 2020
commit 08e23a8 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Dec 20 15:20:01 2019 +0100 SPADE refactored if-else statements (NeuralEnsemble#285) Simplified code complexity by move if-else blocks here and there commit fce24ad Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Tue Dec 17 12:23:00 2019 +0100 Bug fixed while dealing with the units of the refr. period (NeuralEnsemble#283) commit 8200eef Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Dec 13 16:11:22 2019 +0100 Feature/spade speed up (NeuralEnsemble#280) * Remove while loop, change mpi-use in pvalue spec * Use sparse representation of the BinnedSpikeTrain * refactored false discovery rate * reorganized errors for test_signature_significance * Implement filtering for pattern subsets when not performing pattern set reduction check for subset patterns trivially explained by longer patterns with the same n_occ due to the moving window * Adapt tests to new filtering proceidure * Always apply subset filtering commit 160e8e2 Author: Danylo Ulianych <dizcza@gmail.com> Date: Wed Dec 11 15:44:04 2019 +0100 fixed bug in BinnedSpikeTrain.sparsity (NeuralEnsemble#281) commit 0d44cd7 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Dec 5 10:25:48 2019 +0100 unitary event decorate_deprecated_N refactoring (NeuralEnsemble#278) commit b116804 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 25 12:31:36 2019 +0100 Fast covariance() and Pearson corrcoef() (NeuralEnsemble#274) clarified the difference in the implementation of CrossCorrHist.cross_corr_coef() and the reference book "Analysis of parallel spike trains", 2010 commit 59e6412 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Wed Nov 20 09:46:54 2019 +0100 Feature/spade duration (NeuralEnsemble#263) * Corrected error message * Doc. for Holm-Bonferroni * Changed calculation of duration for output of patterns * fixed calculation of duration for the signature * Pass spectrum to concepts_to_output function * Bug fixed in vairable assignment: concept commit de077d6 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Nov 19 16:17:52 2019 +0100 Use fftconvolve instead of np.correlate in cross_correlation_histogram (NeuralEnsemble#273) * cch np.correlate -> fftconvolve * moved internal functions in cross_correlation_histogram to a separate class * added BinnedSpikeTrain.get_num_of_spikes() function * added BinnedSpikeTrain.binarize() function commit 059ccf5 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Nov 15 16:24:58 2019 +0100 Homogeneous Poisson Process with refr. period (NeuralEnsemble#261) commit f378f07 Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Tue Nov 12 10:23:29 2019 +0100 Feature timescale (NeuralEnsemble#271) * Implementation of timescale calculation * Use cross_corr_coef of CCH function * Check that tau_max is an integer multiple of binsize commit 82f307f Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 11:09:44 2019 +0100 fixed BinnedSpikeTrain.bin_edges property (NeuralEnsemble#257) * fixed BinnedSpikeTrain.bin_edges property * bin_edges: include t_stop if spiketrain duration is divisible by binsize * BinnedSpikeTrain: warn if binning discards the spikes in the last (excluded) bin * reuse is_binary func from utility.py commit bd684f5 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 10:14:34 2019 +0100 Refactored unitary event (NeuralEnsemble#251) * completely removed misleading N arg - num. of neurons, which is always extracted from the data * fixed inverse_hash_from_pattern for long N > int64; added test_hash_inverse_longpattern * UE: deprecation warning commit 2967dac Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Fri Nov 8 16:38:27 2019 +0100 Calculate spike train auto-correlation time ('timescale') (NeuralEnsemble#266) * Implementation of timescale calculation commit 092d2fb Author: Michael Denker <m.denker@fz-juelich.de> Date: Fri Nov 8 12:33:32 2019 +0100 Added a simplified to correctly handle division with Compound Units (NeuralEnsemble#270) .magnitude -> .simplified.magnitude commit 5dfa042 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 31 15:54:30 2019 +0100 elephant v0.6.4 [workshop] (NeuralEnsemble#268) commit 5b5a023 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Oct 22 16:10:16 2019 +0200 neo v0.8.0 compatible (NeuralEnsemble#256) * fixes for neo v0.8.0 * upd array_annotation comment * added todo for neo empty array_annotations commit eac914c Author: PaulinaDabrowska <p.dabrowska@fz-juelich.de> Date: Tue Oct 22 16:09:41 2019 +0200 Improved doc string of covariance() and corrcoef() (NeuralEnsemble#260) * corrected explanation of the formula used in corrcoef and covariance (in the doc string) * pep8 commit 434cacd Author: Robin Gutzen <robin.gutzen@live.de> Date: Mon Oct 21 13:56:38 2019 +0200 keep array_annotations in the output of signal_processing functions (NeuralEnsemble#258) * pass array_annotations on to output of signal_processing functions * test keep array_annotations for zscore, butter, hilbert
dizcza
added a commit
to INM-6/elephant
that referenced
this pull request
Jan 10, 2020
commit 731bbe4 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Jan 10 12:15:45 2020 +0100 Feature/joint isi dithering (NeuralEnsemble#275) commit 08e23a8 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Dec 20 15:20:01 2019 +0100 SPADE refactored if-else statements (NeuralEnsemble#285) Simplified code complexity by move if-else blocks here and there commit fce24ad Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Tue Dec 17 12:23:00 2019 +0100 Bug fixed while dealing with the units of the refr. period (NeuralEnsemble#283) commit 8200eef Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Dec 13 16:11:22 2019 +0100 Feature/spade speed up (NeuralEnsemble#280) * Remove while loop, change mpi-use in pvalue spec * Use sparse representation of the BinnedSpikeTrain * refactored false discovery rate * reorganized errors for test_signature_significance * Implement filtering for pattern subsets when not performing pattern set reduction check for subset patterns trivially explained by longer patterns with the same n_occ due to the moving window * Adapt tests to new filtering proceidure * Always apply subset filtering commit 160e8e2 Author: Danylo Ulianych <dizcza@gmail.com> Date: Wed Dec 11 15:44:04 2019 +0100 fixed bug in BinnedSpikeTrain.sparsity (NeuralEnsemble#281) commit 0d44cd7 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Dec 5 10:25:48 2019 +0100 unitary event decorate_deprecated_N refactoring (NeuralEnsemble#278) commit b116804 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 25 12:31:36 2019 +0100 Fast covariance() and Pearson corrcoef() (NeuralEnsemble#274) clarified the difference in the implementation of CrossCorrHist.cross_corr_coef() and the reference book "Analysis of parallel spike trains", 2010 commit 59e6412 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Wed Nov 20 09:46:54 2019 +0100 Feature/spade duration (NeuralEnsemble#263) * Corrected error message * Doc. for Holm-Bonferroni * Changed calculation of duration for output of patterns * fixed calculation of duration for the signature * Pass spectrum to concepts_to_output function * Bug fixed in vairable assignment: concept commit de077d6 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Nov 19 16:17:52 2019 +0100 Use fftconvolve instead of np.correlate in cross_correlation_histogram (NeuralEnsemble#273) * cch np.correlate -> fftconvolve * moved internal functions in cross_correlation_histogram to a separate class * added BinnedSpikeTrain.get_num_of_spikes() function * added BinnedSpikeTrain.binarize() function commit 059ccf5 Author: pbouss <34713558+pbouss@users.noreply.github.com> Date: Fri Nov 15 16:24:58 2019 +0100 Homogeneous Poisson Process with refr. period (NeuralEnsemble#261) commit f378f07 Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Tue Nov 12 10:23:29 2019 +0100 Feature timescale (NeuralEnsemble#271) * Implementation of timescale calculation * Use cross_corr_coef of CCH function * Check that tau_max is an integer multiple of binsize commit 82f307f Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 11:09:44 2019 +0100 fixed BinnedSpikeTrain.bin_edges property (NeuralEnsemble#257) * fixed BinnedSpikeTrain.bin_edges property * bin_edges: include t_stop if spiketrain duration is divisible by binsize * BinnedSpikeTrain: warn if binning discards the spikes in the last (excluded) bin * reuse is_binary func from utility.py commit bd684f5 Author: Danylo Ulianych <dizcza@gmail.com> Date: Mon Nov 11 10:14:34 2019 +0100 Refactored unitary event (NeuralEnsemble#251) * completely removed misleading N arg - num. of neurons, which is always extracted from the data * fixed inverse_hash_from_pattern for long N > int64; added test_hash_inverse_longpattern * UE: deprecation warning commit 2967dac Author: Alexander van Meegen <32771370+AlexVanMeegen@users.noreply.github.com> Date: Fri Nov 8 16:38:27 2019 +0100 Calculate spike train auto-correlation time ('timescale') (NeuralEnsemble#266) * Implementation of timescale calculation commit 092d2fb Author: Michael Denker <m.denker@fz-juelich.de> Date: Fri Nov 8 12:33:32 2019 +0100 Added a simplified to correctly handle division with Compound Units (NeuralEnsemble#270) .magnitude -> .simplified.magnitude commit 5dfa042 Author: Danylo Ulianych <dizcza@gmail.com> Date: Thu Oct 31 15:54:30 2019 +0100 elephant v0.6.4 [workshop] (NeuralEnsemble#268) commit 5b5a023 Author: Danylo Ulianych <dizcza@gmail.com> Date: Tue Oct 22 16:10:16 2019 +0200 neo v0.8.0 compatible (NeuralEnsemble#256) * fixes for neo v0.8.0 * upd array_annotation comment * added todo for neo empty array_annotations commit eac914c Author: PaulinaDabrowska <p.dabrowska@fz-juelich.de> Date: Tue Oct 22 16:09:41 2019 +0200 Improved doc string of covariance() and corrcoef() (NeuralEnsemble#260) * corrected explanation of the formula used in corrcoef and covariance (in the doc string) * pep8 commit 434cacd Author: Robin Gutzen <robin.gutzen@live.de> Date: Mon Oct 21 13:56:38 2019 +0200 keep array_annotations in the output of signal_processing functions (NeuralEnsemble#258) * pass array_annotations on to output of signal_processing functions * test keep array_annotations for zscore, butter, hilbert
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
np.correlatehas the time complexity ofO(N)whilefftconvolvehas just a constant. For small arraysnp.correlateis preferred, but the main problem, of course, with large arrays (see picture).CrossCorrHistclass that encapsulates the "private" functions in the former implementation.get_num_of_spikes()inBinnedSpikeTrain..spike_indicesproperty inBinnedSpikeTrain.binarize()function inBinnedSpikeTrain.