# PyntBCI
The Python Noise-Tagging Brain-Computer Interfacing (PyntBCI) library is a specialized Python toolbox developed for the noise-tagging brain-computer interfacing (BCI) project at the Donders Institute for Brain, Cognition, and Behaviour at Radboud University in Nijmegen, the Netherlands. PyntBCI offers a suite of signal processing tools and machine learning algorithms tailored for BCIs using evoked responses, such as those recorded by electroencephalography (EEG). It is particularly focused on supporting code-modulated responses like the code-modulated visual evoked potential (c-VEP).
For detailed documentation as wel as tutorials and examples, see:
- https://thijor.github.io/pyntbci/
For a constructive review of the c-VEP BCI field, see:
- Martínez-Cagigal, V., Thielen, J., Santamaría-Vázquez, E., Pérez-Velasco, S., Desain, P., & Hornero, R. (2021). Brain–computer interfaces based on code-modulated visual evoked potentials (c-VEP): a literature review. Journal of Neural Engineering. DOI: [10.1088/1741-2552/ac38cf](https://doi.org/10.1088/1741-2552/ac38cf)
For an extensive literature overview, also see:
- https://github.com/thijor/cvep-literature.
For an example of an online BCI with PyntBCI, see our Dareplane implementation:
- https://github.com/thijor/dp-cvep
- https://github.com/thijor/dp-cvep-speller
- https://github.com/thijor/dp-cvep-decoder
## Installation
To install PyntBCI, use:
pip install pyntbci
## Getting started
Various tutorials and example analysis pipelines are provided in the `tutorials/` and `examples/` folder, which operate on limited preprocessed data as provided with PyntBCI. Furthermore, please find various pipelines for several open-access datasets below in the `pipelines/` folder.
## Referencing
When using PyntBCI, please reference at least one of the following:
* Thielen, J., van den Broek, P., Farquhar, J., & Desain, P. (2015). Broad-Band visually evoked potentials: re(con)volution in brain-computer interfacing. PLOS ONE, 10(7), e0133797. DOI: [10.1371/journal.pone.0133797](https://doi.org/10.1371/journal.pone.0133797)
* Thielen, J., Marsman, P., Farquhar, J., & Desain, P. (2021). From full calibration to zero training for a code-modulated visual evoked potentials for brain–computer interface. Journal of Neural Engineering, 18(5), 056007. DOI: [10.1088/1741-2552/abecef](https://doi.org/10.1088/1741-2552/abecef)
## Contact
* Jordy Thielen (jordy.thielen@donders.ru.nl)
## Licensing
PyntBCI is licensed by the BSD 3-Clause License:
Copyright (c) 2021, Jordy Thielen All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
# Changelog
## Version 1.8.0 (08-11-2024)
### Added
- Added `min_time` to stopping methods in `stopping`
- Added `max_time` to `CriterionStopping` in `stopping`
### Changed
### Fixed
- Fixed fit exception in `DistributionStopping` in `stopping`
## Version 1.7.0 (22-10-2024)
### Added
- Added `tmin` to `encoding_matrix` in `utilities`
- Added `tmin` to `rCCA` in `classifiers`
### Changed
### Fixed
## Version 1.6.1 (10-10-2024)
### Added
- Added `find_neighbours` and `find_worst_neighbour` to `utilities`
- Added `optimize_subset_clustering` to `stimulus`
- Added `optimize_layout_incremental` to `stimulus`
- Added `stimplot` to `plotting`
### Changed
- Changed order of tutorials and examples
### Fixed
- Fixed `max_time` in all `stopping` classes to deal with "partial" segments
## Version 1.5.0 (30-09-2024)
### Added
- Added `ValueStopping` to `stopping`
- Added parameter `distribution` to `DistributionStopping` in `stopping`
### Changed
- Changed `envelope_rms` to `rms` in `envelope`
- Changed `envelope_gammatone` to `gammatone` in `envelope`
- Changed `BetaStopping` in `stopping` to `DistributionStopping`
### Fixed
- Fixed default `CCA` in `transformers` to `inv`, not `pinv`
- Fixed `seed` for `make_m_sequence` and `make_gold_codes` in `stimulus` to not be full zeros
## Version 1.4.1 (19-07-2024)
### Added
### Changed
### Fixed
- Fixed default `CCA` in `transformers` to `inv`, not `pinv`
## Version 1.4.0 (15-07-2024)
### Added
- Added `pinv` to `utilities`
- Added `alpha_x` to `CCA` in `tranformers`
- Added `alpha_y` to `CCA` in `tranformers`
- Added `alpha_x` to `eCCA` in `classifiers`
- Added `alpha_t` to `eCCA` in `classifiers`
- Added `alpha_x` to `rCCA` in `classifiers`
- Added `alpha_m` to `rCCA` in `classifiers`
- Added `squeeze_components` to `rCCA`, `eCCA`, `eTRCA` in `classifiers'
### Changed
- Changed `numpy` typing of `np.ndarray` to `NDArray`
- Changed `cca_` and `trca_` attributes to be `list` always in `eCCA`, `rCCA` and `eTRCA`
- Changed `scipy.linalg.inv` to `pyntbci.utilities.pinv` in `CCA` of `transformers`
- Changed `decision_function` and `predict` of `classifiers` to return without additional dimension for components if `n_components=1` and `squeeze_components=True`, both of which are defaults
### Fixed
## Version 1.3.3 (01-07-2024)
### Added
### Changed
### Fixed
- Fixed components bug in `decision_function` of `eCCA` in `classifiers`
## Version 1.3.2 (23-06-2024)
### Added
- Added `cov_estimator_t` to `eCCA` in `classifiers`
### Changed
- Changed separate covariance estimators for data and templates in `eCCA` of `classifiers`
### Fixed
## Version 1.3.1 (23-06-2024)
### Added
### Changed
### Fixed
- Fixed zero division `eventplot` in `plotting`
- Fixed event order duration event `event_matrix` in `utilities`
## Version 1.3.0 (18-06-2024)
### Added
- Removed `gating` of `rCCA` in `classifiers`
- Removed `_score` methods in `classifiers`
- Added `n_components` in `eCCA` in `classifiers`
- Added `n_components` in `eTRCA` in `classifiers`
### Changed
- Changed "bes" to "bds" in `BayesStopping` in `stopping` in line with publication
- Changed `lx` and `ly` to `gamma_x` and `gamma_y` iof `eCCA` in `classifiers`
- Changed `gating` to `gates`
- Changed `TRCA` in `transformers` to deal with one-class data only
- Changed `_get_T` to `get_T` in all `classifiers`
### Fixed
## Version 1.2.0 (18-04-2024)
### Added
### Changed
- Changed `lx` of `rCCA` in `classifiers` to `gamma_x`, which ranges between 0-1, such that the parameter represents shrinkage regularization
- Changed `ly` of `rCCA` in `classifiers` to `gamma_m`, which ranges between 0-1, such that the parameter represents shrinkage regularization
- Changed `lx` of `CCA` in `transformers` to `gamma_x`, which ranges between 0-1, such that the parameter represents shrinkage regularization
- Changed `ly` of `CCA` in `transformers` to `gamma_y`, which ranges between 0-1, such that the parameter represents shrinkage regularization
### Fixed
## Version 1.1.0 (17-04-2024)
### Added
- Added `envelope` module containing `envelope_gammatone` and `envelope_rms` functions
- Added `CriterionStopping` to `stopping` for some static stopping methods
### Changed
- Changed default value of `encoding_length` in `rCCA` of `classifiers` of 0.3 to None, which is equivalent to 1 / fs
### Fixed
- Fixed variable `fs` of type np.ndarray instead of int in examples, tutorials, and pipelines
- Fixed double call to `decoding_matrix` in `fit` of `rCCA` in `classifiers`
## Version 1.0.1 (26-03-2024)
### Added
- Added `set_stimulus_amplitudes` for `rCCA` in `classifiers`
### Changed
### Fixed
- Fixed dependency between `stimulus` and `amplitudes` in `rCCA` of `classifiers`
## Version 1.0.0 (22-03-2024)
### Added
- Added variable `decoding_length` of `rCCA` in `classifier` controlling the length of a learned spectral filter
- Added variable `decoding_stride` of `rCCA` in `classifier` controlling the stride of a learned spectral filter
- Added function `decoding_matrix` in `utilities` to phase-shit the EEG data maintaining channel-prime ordering
- Added variable `encoding_stride` of `rCCA` in `classifier` controlling the stride of a learned temporal response
- Added module `gating` with gating functions, for instance for multi-component or filterbank analysis
- Added variable `gating` of `rCCA` in `classifier` to deal with multiple CCA components
- Added variable `gating` of `Ensemble` in `classifier`, for example to deal with a filterbank
### Changed
- Changed variable `codes` of `rCCA` in `classifiers` to `stimulus`
- Changed variable `transient_size` of `rCCA` in `classifiers` to `encoding_length`
- Changed class `FilterBank` in `classifiers` to `Ensemble`
- Changed function `structure_matrix` in `utilities` to `encoding_matrix`
### Fixed
- Fixed several documentation issues
## Version 0.2.5 (29-02-2024)
### Added
- Added function `eventplot` in `plotting` to visualize an event matrix
- Added variable `running` of `covariance` in `utilities` to do incremental running covariance updates
- Added variable `running` of `CCA` in `transformers` to use a running covariance for CCA
- Added variable `cov_estimator_x` and `cov_estimator_m` of `rCCA` in `classifiers` to change the covariance estimator
- Added event definitions "on", "off" and "onoff" for `event_matrix` in `utilities`
### Changed
- Changed the CCA optimization to contain separate computations for Cxx, Cyy and Cxy
- Changed the CCA to allow separate BaseEstimators for Cxx and Cyy
### Fixed
- Fixed zero-division in `itr` in `utilities`
## Version 0.2.4
### Added
- Added CCA cumulative/incremental average and covariance
- Added `amplitudes` (e.g. envelopes) in `structure_matrix` of `utilities`
- Added `max_time` to classes in `stopping` to allow a maximum stopping time for stopping methods
- Added brainamp64.loc to capfiles
- Added plt.show() in all examples
### Changed
### Fixed
## Version 0.2.3
### Added
### Changed
- Changed example pipelines to include more examples and explanation
- Changed tutorial pipelines to include more examples and explanation
### Fixed
- Fixed several documentation issues
## Version 0.2.2
### Added
- Added class `TRCA` to `transformers`
- Added class `eTRCA` to `classifiers`
- Added parameter `ensemble` to classes in `classifiers` to allow a separate spatial filter per class
### Changed
- Changed package name from PyNT to PyntBCI to avoid clash with existing pynt library
- Changed filter order in `filterbank` of `utilities` to be optimized given input parameters
### Fixed
- Fixed issue in `rCCA` of `classifiers` causing novel events in structure matrix when "cutting cycles"
- Fixed `correlation` to not contain mutable input variables
## Version 0.2.1
### Added
- Added `tests`
- Added tutorials
### Changed
- Changed `rCCA` to work with non-binary events instead of binary only
### Fixed
## Version 0.2.0
### Added
- Added dynamic stopping: classes `MarginStopping`, `BetaStopping`, and `BayesStopping` in module `stopping`
- Added value inner for variable `score_metric` in 'classifiers'
### Changed
- Changed all data shapes from (channels, samples, trials) to (trials, channels, samples)
- Changed all codes shapes from (samples, classes) to (classes, samples)
- Changed all decision functions to similarity, not distance (e.g., Euclidean), to always maximize
### Fixed
- Fixed zero-mean templates in `eCCA` and `rCCA` of `classifiers`
## Version 0.1.0
### Added
- Added `Filterbank` to `classifiers`
### Changed
- Changed classifiers all have `predict` and `decision_function` methods in `classifiers`
### Fixed
## Version 0.0.2
### Added
### Changed
- Changed CCA method from sklearn to custom covariance method
### Fixed
## Version 0.0.1
### Added
- Added `eCCA` template metrics: average, median, OCSVM
- Added `eCCA` spatial filter options: all channels or subset
### Changed
### Fixed
## Version 0.0.0
### Added
- Added `CCA` in `transformers`
- Added `rCCA` in `classifiers`
- Added `eCCA` in `classifier`
### Changed
### Fixed
Raw data
{
"_id": null,
"home_page": "https://github.com/thijor/pyntbci",
"name": "pyntbci",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "bci, classification, cvep, c-vep, eeg",
"author": "Jordy Thielen",
"author_email": "jordy.thielen@donders.ru.nl",
"download_url": "https://files.pythonhosted.org/packages/fc/ec/0943f5765d165363ccd5ab41effaba0f0f85d1cbca23120e1d7cebe018f4/pyntbci-1.8.0.tar.gz",
"platform": null,
"description": "# PyntBCI\n\nThe Python Noise-Tagging Brain-Computer Interfacing (PyntBCI) library is a specialized Python toolbox developed for the noise-tagging brain-computer interfacing (BCI) project at the Donders Institute for Brain, Cognition, and Behaviour at Radboud University in Nijmegen, the Netherlands. PyntBCI offers a suite of signal processing tools and machine learning algorithms tailored for BCIs using evoked responses, such as those recorded by electroencephalography (EEG). It is particularly focused on supporting code-modulated responses like the code-modulated visual evoked potential (c-VEP). \n\nFor detailed documentation as wel as tutorials and examples, see:\n- https://thijor.github.io/pyntbci/\n\nFor a constructive review of the c-VEP BCI field, see:\n- Mart\u00ednez-Cagigal, V., Thielen, J., Santamar\u00eda-V\u00e1zquez, E., P\u00e9rez-Velasco, S., Desain, P., & Hornero, R. (2021). Brain\u2013computer interfaces based on code-modulated visual evoked potentials (c-VEP): a literature review. Journal of Neural Engineering. DOI: [10.1088/1741-2552/ac38cf](https://doi.org/10.1088/1741-2552/ac38cf)\n\nFor an extensive literature overview, also see:\n- https://github.com/thijor/cvep-literature.\n\nFor an example of an online BCI with PyntBCI, see our Dareplane implementation:\n- https://github.com/thijor/dp-cvep\n- https://github.com/thijor/dp-cvep-speller\n- https://github.com/thijor/dp-cvep-decoder\n\n## Installation\n\nTo install PyntBCI, use:\n\n\tpip install pyntbci\n\n## Getting started\n\nVarious tutorials and example analysis pipelines are provided in the `tutorials/` and `examples/` folder, which operate on limited preprocessed data as provided with PyntBCI. Furthermore, please find various pipelines for several open-access datasets below in the `pipelines/` folder.\n\n## Referencing\n\nWhen using PyntBCI, please reference at least one of the following:\n\n* Thielen, J., van den Broek, P., Farquhar, J., & Desain, P. (2015). Broad-Band visually evoked potentials: re(con)volution in brain-computer interfacing. PLOS ONE, 10(7), e0133797. DOI: [10.1371/journal.pone.0133797](https://doi.org/10.1371/journal.pone.0133797)\n* Thielen, J., Marsman, P., Farquhar, J., & Desain, P. (2021). From full calibration to zero training for a code-modulated visual evoked potentials for brain\u2013computer interface. Journal of Neural Engineering, 18(5), 056007. DOI: [10.1088/1741-2552/abecef](https://doi.org/10.1088/1741-2552/abecef)\n\n## Contact\n\n* Jordy Thielen (jordy.thielen@donders.ru.nl)\n\n## Licensing\n\nPyntBCI is licensed by the BSD 3-Clause License:\n\nCopyright (c) 2021, Jordy Thielen All rights reserved.\n\nRedistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:\n\n1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.\n2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.\n3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.\n\nTHIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS \"AS IS\" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.\n\n# Changelog\n\n## Version 1.8.0 (08-11-2024)\n\n### Added\n- Added `min_time` to stopping methods in `stopping`\n- Added `max_time` to `CriterionStopping` in `stopping`\n\n### Changed\n\n### Fixed\n- Fixed fit exception in `DistributionStopping` in `stopping`\n\n## Version 1.7.0 (22-10-2024)\n\n### Added\n- Added `tmin` to `encoding_matrix` in `utilities`\n- Added `tmin` to `rCCA` in `classifiers`\n\n### Changed\n\n### Fixed\n\n## Version 1.6.1 (10-10-2024)\n\n### Added\n- Added `find_neighbours` and `find_worst_neighbour` to `utilities`\n- Added `optimize_subset_clustering` to `stimulus`\n- Added `optimize_layout_incremental` to `stimulus`\n- Added `stimplot` to `plotting`\n\n### Changed\n- Changed order of tutorials and examples\n\n### Fixed\n- Fixed `max_time` in all `stopping` classes to deal with \"partial\" segments\n\n## Version 1.5.0 (30-09-2024)\n\n### Added\n- Added `ValueStopping` to `stopping`\n- Added parameter `distribution` to `DistributionStopping` in `stopping` \n\n### Changed\n- Changed `envelope_rms` to `rms` in `envelope`\n- Changed `envelope_gammatone` to `gammatone` in `envelope`\n- Changed `BetaStopping` in `stopping` to `DistributionStopping`\n\n### Fixed\n- Fixed default `CCA` in `transformers` to `inv`, not `pinv`\n- Fixed `seed` for `make_m_sequence` and `make_gold_codes` in `stimulus` to not be full zeros\n\n## Version 1.4.1 (19-07-2024)\n\n### Added\n\n### Changed\n\n### Fixed\n- Fixed default `CCA` in `transformers` to `inv`, not `pinv`\n\n## Version 1.4.0 (15-07-2024)\n\n### Added\n- Added `pinv` to `utilities`\n- Added `alpha_x` to `CCA` in `tranformers`\n- Added `alpha_y` to `CCA` in `tranformers`\n- Added `alpha_x` to `eCCA` in `classifiers`\n- Added `alpha_t` to `eCCA` in `classifiers`\n- Added `alpha_x` to `rCCA` in `classifiers`\n- Added `alpha_m` to `rCCA` in `classifiers`\n- Added `squeeze_components` to `rCCA`, `eCCA`, `eTRCA` in `classifiers'\n\n### Changed\n- Changed `numpy` typing of `np.ndarray` to `NDArray`\n- Changed `cca_` and `trca_` attributes to be `list` always in `eCCA`, `rCCA` and `eTRCA`\n- Changed `scipy.linalg.inv` to `pyntbci.utilities.pinv` in `CCA` of `transformers`\n- Changed `decision_function` and `predict` of `classifiers` to return without additional dimension for components if `n_components=1` and `squeeze_components=True`, both of which are defaults\n\n### Fixed\n\n## Version 1.3.3 (01-07-2024)\n\n### Added\n\n### Changed\n\n### Fixed\n- Fixed components bug in `decision_function` of `eCCA` in `classifiers` \n\n## Version 1.3.2 (23-06-2024)\n\n### Added\n- Added `cov_estimator_t` to `eCCA` in `classifiers`\n\n### Changed\n- Changed separate covariance estimators for data and templates in `eCCA` of `classifiers`\n\n### Fixed\n\n## Version 1.3.1 (23-06-2024)\n\n### Added\n\n### Changed\n\n### Fixed\n- Fixed zero division `eventplot` in `plotting`\n- Fixed event order duration event `event_matrix` in `utilities` \n\n## Version 1.3.0 (18-06-2024)\n\n### Added\n- Removed `gating` of `rCCA` in `classifiers`\n- Removed `_score` methods in `classifiers`\n- Added `n_components` in `eCCA` in `classifiers`\n- Added `n_components` in `eTRCA` in `classifiers`\n\n### Changed\n- Changed \"bes\" to \"bds\" in `BayesStopping` in `stopping` in line with publication\n- Changed `lx` and `ly` to `gamma_x` and `gamma_y` iof `eCCA` in `classifiers`\n- Changed `gating` to `gates`\n- Changed `TRCA` in `transformers` to deal with one-class data only\n- Changed `_get_T` to `get_T` in all `classifiers`\n\n### Fixed\n\n## Version 1.2.0 (18-04-2024)\n\n### Added\n\n### Changed\n\n- Changed `lx` of `rCCA` in `classifiers` to `gamma_x`, which ranges between 0-1, such that the parameter represents shrinkage regularization\n- Changed `ly` of `rCCA` in `classifiers` to `gamma_m`, which ranges between 0-1, such that the parameter represents shrinkage regularization\n- Changed `lx` of `CCA` in `transformers` to `gamma_x`, which ranges between 0-1, such that the parameter represents shrinkage regularization\n- Changed `ly` of `CCA` in `transformers` to `gamma_y`, which ranges between 0-1, such that the parameter represents shrinkage regularization\n\n### Fixed\n\n## Version 1.1.0 (17-04-2024)\n\n### Added\n\n- Added `envelope` module containing `envelope_gammatone` and `envelope_rms` functions\n- Added `CriterionStopping` to `stopping` for some static stopping methods \n\n### Changed\n\n- Changed default value of `encoding_length` in `rCCA` of `classifiers` of 0.3 to None, which is equivalent to 1 / fs\n\n### Fixed\n\n- Fixed variable `fs` of type np.ndarray instead of int in examples, tutorials, and pipelines \n- Fixed double call to `decoding_matrix` in `fit` of `rCCA` in `classifiers`\n\n## Version 1.0.1 (26-03-2024)\n\n### Added\n\n- Added `set_stimulus_amplitudes` for `rCCA` in `classifiers`\n\n### Changed\n\n### Fixed\n\n- Fixed dependency between `stimulus` and `amplitudes` in `rCCA` of `classifiers`\n\n## Version 1.0.0 (22-03-2024)\n\n### Added\n\n- Added variable `decoding_length` of `rCCA` in `classifier` controlling the length of a learned spectral filter\n- Added variable `decoding_stride` of `rCCA` in `classifier` controlling the stride of a learned spectral filter\n- Added function `decoding_matrix` in `utilities` to phase-shit the EEG data maintaining channel-prime ordering\n- Added variable `encoding_stride` of `rCCA` in `classifier` controlling the stride of a learned temporal response\n- Added module `gating` with gating functions, for instance for multi-component or filterbank analysis\n- Added variable `gating` of `rCCA` in `classifier` to deal with multiple CCA components\n- Added variable `gating` of `Ensemble` in `classifier`, for example to deal with a filterbank\n\n### Changed\n\n- Changed variable `codes` of `rCCA` in `classifiers` to `stimulus`\n- Changed variable `transient_size` of `rCCA` in `classifiers` to `encoding_length`\n- Changed class `FilterBank` in `classifiers` to `Ensemble`\n- Changed function `structure_matrix` in `utilities` to `encoding_matrix`\n\n### Fixed\n\n- Fixed several documentation issues\n\n## Version 0.2.5 (29-02-2024)\n\n### Added\n\n- Added function `eventplot` in `plotting` to visualize an event matrix\n- Added variable `running` of `covariance` in `utilities` to do incremental running covariance updates\n- Added variable `running` of `CCA` in `transformers` to use a running covariance for CCA \n- Added variable `cov_estimator_x` and `cov_estimator_m` of `rCCA` in `classifiers` to change the covariance estimator \n- Added event definitions \"on\", \"off\" and \"onoff\" for `event_matrix` in `utilities`\n\n### Changed\n\n- Changed the CCA optimization to contain separate computations for Cxx, Cyy and Cxy\n- Changed the CCA to allow separate BaseEstimators for Cxx and Cyy\n\n### Fixed\n\n- Fixed zero-division in `itr` in `utilities` \n\n## Version 0.2.4\n\n### Added\n\n- Added CCA cumulative/incremental average and covariance\n- Added `amplitudes` (e.g. envelopes) in `structure_matrix` of `utilities`\n- Added `max_time` to classes in `stopping` to allow a maximum stopping time for stopping methods\n- Added brainamp64.loc to capfiles\n- Added plt.show() in all examples\n\n### Changed\n\n### Fixed\n\n## Version 0.2.3\n\n### Added\n\n### Changed\n\n- Changed example pipelines to include more examples and explanation\n- Changed tutorial pipelines to include more examples and explanation\n\n### Fixed\n\n- Fixed several documentation issues\n\n## Version 0.2.2\n\n### Added\n\n- Added class `TRCA` to `transformers`\n- Added class `eTRCA` to `classifiers`\n- Added parameter `ensemble` to classes in `classifiers` to allow a separate spatial filter per class\n\n### Changed\n\n- Changed package name from PyNT to PyntBCI to avoid clash with existing pynt library\n- Changed filter order in `filterbank` of `utilities` to be optimized given input parameters\n\n### Fixed\n\n- Fixed issue in `rCCA` of `classifiers` causing novel events in structure matrix when \"cutting cycles\"\n- Fixed `correlation` to not contain mutable input variables\n\n## Version 0.2.1\n\n### Added\n\n- Added `tests`\n- Added tutorials\n\n### Changed\n\n- Changed `rCCA` to work with non-binary events instead of binary only\n\n### Fixed\n\n## Version 0.2.0\n\n### Added\n\n- Added dynamic stopping: classes `MarginStopping`, `BetaStopping`, and `BayesStopping` in module `stopping`\n- Added value inner for variable `score_metric` in 'classifiers'\n\n### Changed\n\n- Changed all data shapes from (channels, samples, trials) to (trials, channels, samples)\n- Changed all codes shapes from (samples, classes) to (classes, samples)\n- Changed all decision functions to similarity, not distance (e.g., Euclidean), to always maximize\n\n### Fixed\n\n- Fixed zero-mean templates in `eCCA` and `rCCA` of `classifiers`\n\n## Version 0.1.0\n\n### Added\n\n- Added `Filterbank` to `classifiers`\n\n### Changed\n\n- Changed classifiers all have `predict` and `decision_function` methods in `classifiers`\n\n### Fixed\n\n## Version 0.0.2\n\n### Added\n\n### Changed\n\n- Changed CCA method from sklearn to custom covariance method\n\n### Fixed\n\n## Version 0.0.1\n\n### Added\n\n- Added `eCCA` template metrics: average, median, OCSVM\n- Added `eCCA` spatial filter options: all channels or subset\n\n### Changed\n\n### Fixed\n\n## Version 0.0.0\n\n### Added\n\n- Added `CCA` in `transformers`\n- Added `rCCA` in `classifiers`\n- Added `eCCA` in `classifier`\n\n### Changed\n\n### Fixed\n",
"bugtrack_url": null,
"license": "BSD-3-Clause",
"summary": "Python Noise-Tagging Brain-Computer Interface (PyntBCI)",
"version": "1.8.0",
"project_urls": {
"Homepage": "https://github.com/thijor/pyntbci"
},
"split_keywords": [
"bci",
" classification",
" cvep",
" c-vep",
" eeg"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8336f48043ec080434c843c74fdb39436801d3b68f266f5ff5937b84fcfdfc01",
"md5": "3f04f2ba25983437f4f9f626944f3b7e",
"sha256": "aad4089901160dd6a46dfae12ece55f029e675b308fc591f80da32357f107e78"
},
"downloads": -1,
"filename": "pyntbci-1.8.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "3f04f2ba25983437f4f9f626944f3b7e",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 45234406,
"upload_time": "2024-11-08T14:35:04",
"upload_time_iso_8601": "2024-11-08T14:35:04.599582Z",
"url": "https://files.pythonhosted.org/packages/83/36/f48043ec080434c843c74fdb39436801d3b68f266f5ff5937b84fcfdfc01/pyntbci-1.8.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fcec0943f5765d165363ccd5ab41effaba0f0f85d1cbca23120e1d7cebe018f4",
"md5": "be9bb98f625ddc7d4d71eed11186a5c5",
"sha256": "991ecbef8a23f7f1c02e7d58210b12bc3780e5bd8283feaa2871980ff120a2c1"
},
"downloads": -1,
"filename": "pyntbci-1.8.0.tar.gz",
"has_sig": false,
"md5_digest": "be9bb98f625ddc7d4d71eed11186a5c5",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 45233735,
"upload_time": "2024-11-08T14:35:41",
"upload_time_iso_8601": "2024-11-08T14:35:41.128829Z",
"url": "https://files.pythonhosted.org/packages/fc/ec/0943f5765d165363ccd5ab41effaba0f0f85d1cbca23120e1d7cebe018f4/pyntbci-1.8.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-08 14:35:41",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "thijor",
"github_project": "pyntbci",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "pyntbci"
}
JSON
