Gaitmap

- This release **removes** Python 3.7 support!

Added

- Data Transforms are added as a new algorithm class.
These are simple algorithms that can take a single sensor data as input and provide a `transformed_data_` output.
Transforms can be used for normalisations or feature transforms.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/148)
- A new helper function `gaitmap.utils.array_handling.iterate_region_data` is added.
It allows to iterate over the data of multiple strides or ROIs cut from a dataset.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/149)
- A new helper (`gaitmap.utils.datatype_helper.to_dict_multi_sensor_data`) to convert DataFrame based MultiSensor data
to Dict-based MultiSensor data is added.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/149)
- A new example (`stride_segmentation/barth_dtw_custom_template.py`) is added, which explains how to create custom
dtw-templates from data.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/149)

Changed

- All Algorithm classes now build on `tpcp.Algorithm`.
This is a major change and might result in some unexpected incompatibilities with older code.
The core functionality of tpcp should still work as expected.
If you were using any of the base classes and algorithm helpers, check if they are still available in gaitmap.
If not, there is likely a replacement in tpcp.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/173)
- The way DTW - templates are generated has been changed completely.
Generation is now handled via BaseClasses instead of helper functions.
These baseclasses follow the "optimizable" interface introduced in `tpcp`.
Further, all templates now use the new data transforms instead of a fixed scaling factor.
In combination, it should be much easier to create new templates and use more complex data normalisations.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/149)
- Spatial gait parameters can be calculated, even if some gait parameters are missing.

Removed

- We removed the entire `gaitmap.future` package.
This was marked experimental anyway.
All functionality has been moved into the `tpcp` package.
For most methods and functions, it should be enough to just change the import statement.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/173)
- We removed `create_dtw_template` and `create_interpolated_dtw_template` as ways to create templates.
Instead, you should now use `DtwTemplate` and `InterpolatedDtwTemplate` classes directly.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/149)
- tslearn is dropped as dependency. Instead we have vendored the set of functions we need.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/181)
- Python 3.7 support is dropped.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/181)

Migration Guide

- If you were creating custom templates, you should now use the `DtwTemplate` and `InterpolatedDtwTemplate` classes.
Check the new example (`stride_segmentation/barth_dtw_custom_template`) for more information, on how to use these new classes.
- If you were using scaling factors for DtwTemplates you should now use the new transformers in `tpcp.data_transforms`
instead.
To replace a fixed scaling factor, use `FixedScaler(scale=500)`.
However, in many cases you might be better off using one of the Trainable scalars to automatically learn the scaling
factors from the template.
- If you worked with the raw data of the `BarthDtwTemplate`, be aware that the stored data is now **unscaled**
(i.e. multiplied by 500 compared to the old version).


Scientific Changes

Scientific Changes

- We fixed a bug with the sign of the TC and IC angle.
They were flipped for real sensor data.
Now they correctly follow the conventions [Kanzler2015](https://pubmed.ncbi.nlm.nih.gov/26737518/)
This is a breaking change and your angle results will be different!
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/180)

Added

- A new submodule for ZUPT-Detection Algorithms and a first implementation in form of the `NormZuptDetector`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/167)
- New sub module in preprocessing for sensor alignment functions
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/161)
- New algorithm `PcaAlignment` for sensor heading alignment using 2D PCA to align sensor frame to medio-lateral plane
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/161)
- New position method `PieceWiseLinearDedriftedIntegration` using piece wise linear dedrifted integration utilizing zupt
updates.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/166)
- A new algorithm (`ForwardDirectionSignAlignment`) to detect the walking direction to automatically align the sensor
attachment orientation.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/171)
- An example that explains how to automatically align a sensor to the gaitmap coordinate system, when it was attached in
an arbitrary orientation.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/171)
- The `HerzerEventDetection` - Algorithm, specifically designed to work with stair gait.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/158)

Changed

- Cloning is now more robust and falls back to deep cloning objects that are not algorithms.
This prevents issues when mutable objects (other than nested algorithms) are used in parameters
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/170)
- We now handle mutable defaults to prevent strange edge cases! Read more about it in the `project_structure` guide
under the development-guide tab in the Documentation.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/168)
- The constrained DTW only tracks a single counter now.
By using the positive and the negative value range, we can save 1/3 of the overall RAM.
This should improve performance.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/155)
- `RtsKalman.find_zupts` now gets passed the entire data **before** any unit conversions.
Before it only got the gyro data already converted in rad/s.
This change should make it easier to overwrite the ZUPT detection with custom methods.
This is a breaking change! See the migration guide for more details.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/159)
- The Zupt-Detection of `RtsKalman` is now handled via dependency injection.
Instead of specifying a list of parameters for the Zupt detection you can simply pass a instance of a Zupt detector.
(see `gaitmap.zupt_detection`).
The old parameters a deprecated.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/169)
- The `memory` option for `RtsKalman` was removed.
This is a breaking change!
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/159)

Deprecated

- The parameters `"zupt_threshold_dps", "zupt_window_length_s", "zupt_window_overlap_s"` for `RtsKalman` are now
deprecated.
Instead, a Zupt-Detector instance should be used.
When converting from the current parameters to the new `NormZuptDetector` class, note that the overlap is now
specified as a fraction of the window length and not in seconds!
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/159)


Removed

- The `memory` option for `RtsKalman` was removed.
This is a breaking change!
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/159)

Fixed

- Bugfix for `gaitmap.gait_detection.ullrich_gait_sequence_detection`: Now proper handling of cases where margin
should be added but no gait sequences were detected
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/157)

Migration Guide

- We removed the orientation update feature of the `RtsKalman` class.
This feature was off by default.
Unless you explicitly enabled it, you should be fine.
In case you used the feature, remove all references to the respective keywords.
- The call signature and the units of the gyro data passed `RtsKalman.find_zupts` has changed.
In case you where using custom subclasses of `RtsKalman` to implement custom zupt methods, you need to update your
zupt methods for the new input structure.
However, due to the new changes to ZuptDetectors, it is likely that you do not need a custom subclass anymore, but can
simply pass a instance of a ZuptDetector to the Kalman class
- The `memory` option for `RtsKalman` was removed, as it was not particularly useful. In case you need caching we
recommend implementing it externally.

Scientific Changes

- The "ori_update" feature of the Kalman filter was removed again.
The feature was experimental anyway, but in turns out the implementation was actually wrong.
The RTSKalman filter tracks the change in error state open-loop.
This means errors are not applied during ZUPT-measurements, but simply added up and corrected at the very end.
This works well for position and velocity, but the equations for orientation assume, that the orientation error
remains small enough so that a linearization and separation of the angle error on the different axis is possible.
When the filter is run open loop, it is very likely that the orientation error will quickly reach ranges, where this
doesn't hold anymore.
We concluded that the used method to track the orientation error (and hence, correct it directly) is not suitable for
open-loop Kalmanfilter designs.
It is unlikely that the feature will be added again in the future.
However, prior versions can be used as reference to implement a closed loop Kalmanfilter with direct observations of
the orientation error.
This is a breaking change!
See the Migration section for more details, if you used this feature before.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/159)

With 1.4. we finally realize the longterm goal to make the comparison of algorithms easy and unify code across dataset
and algorithms.
The highlights of this release are the new experimental `Dataset` and `Pipeline` classes and first-class support for
trainable algorithms.

Added

- New module for experimental features. Things implemented there might be changed or deleted without deprecation
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/122)
- New dataset class that will serve as the base class for all future datasets in `gaitmap.future`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/122,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/132)
- New example on how to create custom datasets with the new dataset-class
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/132)
- The new gaitmap logo is now shown everywhere :)
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/129)
- New parameter `zero_division` added to all methods in `gaitmap.evaluation_utils.scores` allowing for setting the
return value in case of a zero_division happening.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/134)
- A new interpolation method `gaitmap.utils.array_handling.multi_array_interpolation` to interpolate multiple 2D arrays
at once to the same length.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/138)
- Baseclasses to build custom pipelines are added
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/128)
- A GridSearch optimizer (similar to GridSearchCV in sklearn, but without the CV part) with a respective example is
added. (https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/128)
- A generic wrapper for algorithms and pipelines that can be optimized/trained with respective example.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/128)
- Added a warning about providing multiple methods when using any trajectory wrapper class to prevent user error.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/142)
- Added a new `safe_run` method to pipelines that should be used over the normal `run` method.
`safe_run` performance multiple checks to catch potential implementation errors in user created pipelines.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/144)
- New parameter `enforce_consistency` added to `gaitmap.event_detection.RamppEventDetection` to make event detection
postprocessing optional.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/136)
- A GridSearchCV optimizer (similar to GridSearchCV in sklearn) with respective example is added.
This version is specifically implemented for gaitmap and has some fancy ways of performance optimization.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/139)
- A cross-validation function is added to evaluate optimizable pipelines of any kind.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/139)
- `gaitmap.gait_detection.ullrich_gait_sequence_detection.UllrichGaitSequenceDetection` now has the option to add a
margin that will be symmetrically added around the detected gait sequences. This should help to included intervals
before and after steady state gait, that can be missed by the algorithm.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/152)

Changed

- Changed the output format of `gaitmap.evaluation_utils.scores.precision_recall_f1_score` so it will
be easier to use with the pipelines and optimizer.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/133)
- Made parameters for `gaitmap.evaluation_utils` keyword only to avoid input confusion.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/135)
- Improved the performance of dtw template interpolation by a factor of 15-30 (depending on the interpolation type).
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/138)
- File based dtw templates do not implicitly cache the template data anymore.
This means calling `get_data` on the template does not change the object anymore, which is desirable in the context of
caching.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/143)

Deprecated

Removed

- `gaitmap.utils.array_handling.interpolate1d` was removed as it is not needed anymore internally.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/138)

Fixed
- After internal updates of scipy>=1.6 we now manually handle empty orientation inputs in
`gaitmap.parameters.spatial_parameters._calc_turning_angle` and
`gaitmap.parameters.spatial_parameters._compute_sole_angle_course`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/153,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/154)

Migration Guide

- If you used `gaitmap.evaluation_utils.scores.precision_recall_f1_score`, the output is now a dictionary instead of a
tuple. Migration should be straight forward.
- Some parameters are now keyword only. In case you get an unexpected error, telling you that "an algorithm only
receives n positional arguments, but you provided n+1", double-check the call signature.

Scientific Changes

Added

- New evaluation function to compare the output of temporal or spatial parameter calculation with ground truth.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/105)
- New function for merging intervals that are overlapping and that are within a specific distance from each other in
`utils.array_handling`. (https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/120)
- Some algorithms now support caching intermediate results with `joblib`.
There is also a new example explaining caching in more detail.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/126 (mostly reverted),
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/127)
- Experimental support for ZUPT based orientation updates in the `RtsKalman` filter.
This approach is not fully validated and should be used with care.
If the feature is turned on, a runtime warning will indicate that as well.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/113)
- `RtsKalman` has a new results `zupts_` that can be used to check which zero-velocity region were used by the filer.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/113)
- A new guide on how to evaluate (gait)parameters using parameter optimization and cross validation.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/124)

Changed

- `evaluation_utils.stride_segmentation._match_label_lists` now works for n-D and should be faster for large stride
length. (https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/110)
- Internal refactoring of how the algorithms combine the results of the single sensors to the final output.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/116)
- Starting from this release the name "sensor data" instead of "dataset" is used to refer to data from multiple sensors
within one recording.
The term "dataset" will be used in the future to describe entire sets of multiple recordings potentially including
multiple participants and further information besides raw IMU data.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/117)
- Internal refactoring that should improve typing. `mypy` now only returns a couple of issues that are not easily
fixable or actual `mypy` bugs.
They will be resolved step by step in the future.
mypy checking can be done using the new command `dodo type_check` and developer should try to not introduce further
typing issues.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/118)
The refactoring also required some small user facing changes:
- The `regions_of_interest` parameter of `RoiStrideSegmentation.segment` is now keyword only to be compatible with the
`StrideSegmentation` base class.
- The `UllrichGaitSequenceDetection` now returns a dict also in the case of merged gait sequences.
The gait sequences stored for all sensor are set equal.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/121)
- `gait_detection.ullrich_gait_sequence_detection` now uses `utils.array_handling.merge_intervals` and should therefore
be able to merge sequences faster.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/120)
- The `regions_of_interest` parameter of `RoiStrideSegmentation.segment` is now keyword only to be compatible with the
`StrideSegmentation` base class.
- `RtsKalman.covariance` now has the shape `(len(data), 9 * 9)` instead of `(len(data) * 9, 9)` which should make it
easier to plot.


Deprecated

Removed

- `gait_detection.ullrich_gait_sequence_detection._gait_sequences_to_boolean_single` has been made redundant by
`utils.array_handling.merge_intervals` and was therefore removed.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/120)
- Removed old serialization format for DTW templates that was deprecated a couple of releases ago.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/issues/127)

Fixed

- In `stride_segmentation.roi_stride_segmentation` the assignment of the offset corrected strides to the
`combined_stride_list` was fixed.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/114)
- Added a note to BarthDTW to clarify the calculation of the distance matrix.
- Changed the Chebyshev distance function in `evaluation_utils.stride_segmentation._match_label_lists` to a general one
that can use n-D input. (https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/115)
- Fixed bug in `utils.static_moment_detection.find_static_samples` where extremely short sequences, which would only fit
a single window, would fail.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/109)

Migration Guide

- Due to the renaming of dataset to sensordata, a couple of names and import path are changed:
- All imports from `gaitmap.utils.dataset_helper` should be changed to `gaitmap.utils.datatype_helper`
- `SingleSensorDataset`, `MultiSensorDataset` and `Dataset` are now called, `SingleSensorData`, `MultiSensorData` and
`SensorData`, respectively.
- The functions `is_single_sensor_dataset`, `is_multi_sensor_dataset`, `is_dataset`, and
`get_multi_sensor_dataset_names` are renamed to `is_single_sensor_data`, `is_multi_sensor_data`, `is_sensor_data`,
and `get_multi_sensor_names`.

Scientific Changes

Added

- An Error-State-Kalman-Filter with rts smoothing (`RtsKalman`) was added that can be used to estimate orientation and
position over longer regions by detecting its own ZUPTs.
This can (should) be used in combination with `RegionLevelTrajectory`.
- A `RegionLevelTrajectory` class that calculates trajectories for entire ROIs/gait sequences.
Otherwise the interface is very similar to `StrideLevelTrajectory`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/87,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/98,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/85)
- Added a new example for `RegionLevelTrajectory`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/87)
- Support for trajectory methods that can calculate orientation and position in one go is added for
`StrideLevelTrajectory` (https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/87)
- Implemented local warping constraints for DTW. This should help in cases were only parts of a sequence are matched
instead of the entire sequence.
These constraints are available for `BaseDTW` and `BarthDTW`, but are **off** by default.
A new class `ConstrainedBarthDTW` was added, that has the constraints **on** by default.
- Added support for `MultiSensorStrideLists` for all evaluation functions in `evaluation_utils.stride_segmentation`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/91)
- Global validation errors for region of interest list
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/88)
- Global validation errors for orientation/position/velocity lists
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/88,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/95)
- New evaluation functions to compare the results of event detection methods with ground truth
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/92)
- Added the functionality to merge gait sequences from individual sensors in `UllrichGaitSequenceDetection`
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/93)

Changed

- The initial orientation for the Stride and Region Level trajectory now uses the correct number of samples as input.
This might change the output of the integration method slightly!
However, it also allows to pass 0 for `align_window_length` and hence, just use the first sample of the integration
region to estimate the initial orientation.
(mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/98)
- It is now ensured that in **all** in gaitmap where a start and an end sample is provided, the end sample is
**exclusive**.
Meaning if the respective region should be extracted from a dataarray, you can simply do `data[start:end]`.
All functions that used different assumptions are changed.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/97~,
https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/103)
- The default check for orientation/position/velocity lists now only expects a "sample" column and not a "sample" and a
"s_id" column by default.
The `s_id` parameter was removed and replaced with a more generic `{position, velocity, orientation}_list_type`
parameter than can be used to check for either stride or region based lists.
See the migration guide for more information.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/88)
- Internal restructuring of the evaluation utils module.
In case you used direct imports of the functions from the submodule, you need to update these.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/92)

Deprecated

Removed

Fixed

- Fixed possible `ZeroDivisionError` when using `evaluation_utils.scores`.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/104)

Migration Guide

- In case your algorithm uses `is_{single,multi}_sensor_{position, velocity, orientation}_list` all usages must be
updated to include `{position, velocity, orientation}_list_type="stride"` to restore the default behaviour before the
update.
If you were using the function with `s_id=False`, you can update to `{position, velocity, orientation}_list_type=None`
to get the same behaviour.

Scientific Changes

- All Dtw based methods now produce outputs that follow our index guidelines.
This means that they correctly produce matches, which end index is **exclusive**.
In some cases this might change existing results by adding 1 to the end index.
(https://mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/103)
- The initial orientation in the StrideLevelTrajectory now has a window with exactly the number of samples specified
(instead of one less due to Python indexing).
This will lead to very small changes in the calculated trajectory.
(mad-srv.informatik.uni-erlangen.de/MadLab/GaitAnalysis/gaitmap/-/merge_requests/98)