[JOSS Review] Broken example notebooks

[espenhgn]

There are several errors and issues with the provided example notebooks:

• There are many abbreviations like IMU, PSS, PANAS, HRV that are not explained. Perhaps these are standard terms in the field, but for a broader acceptance of this tool, these abbreviations should be defined.
• ECG_Processing_Example.ipynb: throws an error at the line fig, axs = ecg.plotting.ecg_plot(ep, key='Data', figsize=(10,5)); NameError: name 'ecg' is not defined
• ECG_Analysis_Example.ipynb fails on line display_dict_structure(study_data_dict) because of the above error in ECG_Processing_Example.ipynb (no results to load).
• EEG_Example.ipynb: Please add some documentation of what functionality this notebook is supposed to demonstrate. Some time series are shown but it’s not very clear what is going on.
• Log_Data_Example.ipynb; ditto.
• Protocol_Example.ipynb fails because of the above error in ECG_Processing_Example.ipynb (no results to load)
• Questionnaire_Example.ipynb: Incorporating purposely failing code (pss = bp.questionnaires.pss(data_pss) resulting in ValueRangeError … ) in the notebook, I don’t think that’s a good idea whatsoever (prevents running all cells and such). One could capture the error in a more controlled manner.
• SklearnPipelinePermuter_Example.ipynb: Throws an error NameError: name 'datasets' is not defined on line breast_cancer = datasets.load_breast_cancer()
• StatsPipeline_Plotting_Example.ipynb: Typo “cortsol”
• STROOP_Example.ipynb: Crashes at line stroop.hr_ensemble_plot(data=dict_phase,figsize=(10,8),ylims=(40,120)) with TypeError: hr_ensemble_plot() missing 1 required positional argument: 'ensemble_id'

Linked issue: https://github.com/openjournals/joss-reviews/issues/3702

[richrobe]

Thank you for your comments! We updated the example notebooks in 1725a2706f730aedd3328febd0f7c9cbd6b0e174.

[espenhgn]

Thank you for your comments! We updated the example notebooks in 1725a27.

Sounds good!

I can now (SHA:f97f3ce9f8cc4d8d976715a738d069532efae3df) run all the example notebooks without crashes and similar.
But in ECG_Analysis_Example.ipynb I do get a bunch of NaN values in the output dataframes around here:

Is that the expected output due to missing datapoints or similar?

[richrobe]

Hi @espenhgn , yes, that's the expected output because the ECG sequences are too short to compute certain HRV parameter. The same warning is also produced for the example in the documentation.

[espenhgn]

Hi @espenhgn , yes, that's the expected output because the ECG sequences are too short to compute certain HRV parameter. The same warning is also produced for the example in the documentation.

Ok, thanks for the clarification. Then I think this issue can be closed!