Christof Leutenegger
Christof Leutenegger
When running the tests we get around 500 warnings, that will results in unstable code. We should try to handle as much errors as possible.
In #431 we discovered an error in calculating the mean angle for a location. We calculate the average coordinates with `.centroid`. This function assumes that the coordinate system is projected...
When a new person discovers our framework, it may be useful for them to compare it to similar packages such as scikit-mobility and movingpandas. This allows us to present what...
read_locations_csv and write_locations_csv both use extent as keyword but there is no test assuring it works.
Implementation of a reader of the "Mikrozensus Mobilität und Verkehr 2015" around 95% finished. It is not the cleanest implementation as the data needs a lot of massaging until we...
- [x] I have checked that this issue has not already been reported. - [x] I have confirmed this bug exists on the latest version of geopandas. - [ ]...
Currently input data in `merge_staypoints` is not validated. Should we add some validation here?
Running the tutorial lead to some plotting result that could be improved. Plotting positionfixes with `ti.plot` lead to a plot that seems to almost cut of some positionfixes. `plot_module_split` by...
I think we should allow empty Positionfixes, Staypoints, etc. And thus remove the check in `validate`. With this we should also remove the empty checks in the `generate` functions
Functions like [`get_speed_positionfixes`](https://github.com/mie-lab/trackintel/blob/963fb9e28f5679ad87c98373c7c3d86ce6a49880/trackintel/geogr/distances.py#L289), [`get_speed_triplegs`](https://github.com/mie-lab/trackintel/blob/963fb9e28f5679ad87c98373c7c3d86ce6a49880/trackintel/geogr/distances.py#L289), [`create_activity_flag`](https://github.com/mie-lab/trackintel/blob/963fb9e28f5679ad87c98373c7c3d86ce6a49880/trackintel/analysis/labelling.py#L8), [`predict_transport_mode`](https://github.com/mie-lab/trackintel/blob/963fb9e28f5679ad87c98373c7c3d86ce6a49880/trackintel/analysis/labelling.py#L47) take a DataFrame, copy it, potentially override a column with the result and then return the new DataFrame. I think it would be cleaner...