Ongoing developments

[philmain28]

Branch contains changes proposed here as they are implemented:
#19

Along with enhancements/issues fixes.

[philmain28]

Added functionality to exclude patients who have a DoD elsewhere in the UKRDC for example as a patient record created by TRACING. I have also included a switch to easily turn this on or off.

[jtc42]

Added functionality to exclude patients who have a DoD elsewhere in the UKRDC for example as a patient record created by TRACING. I have also included a switch to easily turn this on or off.

I'm assuming there's a fairly substantial performance hit associated with this?

[philmain28]

Yeah there is but the demographics stuff calculates pretty quickly anyway. The performance can be improved by uncommenting the line which restricts it to looking in tracing.

[codecov]

Codecov Report

Base: 73.21% // Head: 69.15% // Decreases project coverage by -4.06% ⚠️

Coverage data is based on head (9088e1b) compared to base (ff77a25).
Patch coverage: 88.37% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #20      +/-   ##
==========================================
- Coverage   73.21%   69.15%   -4.07%     
==========================================
  Files          11       13       +2     
  Lines         336      389      +53     
==========================================
+ Hits          246      269      +23     
- Misses         90      120      +30     

Flag	Coverage Δ
3.10	69.15% <88.37%> (-4.07%)	⬇️
3.11	69.15% <88.37%> (-4.07%)	⬇️
3.7	69.15% <88.37%> (-4.07%)	⬇️
3.8	69.15% <88.37%> (-4.07%)	⬇️
3.9	69.15% <88.37%> (-4.07%)	⬇️
unittests	69.15% <88.37%> (-4.07%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
ukrdc_stats/biomarkers.py	0.00% <ø> (ø)
ukrdc_stats/calculators/abc.py	100.00% <ø> (ø)
ukrdc_stats/descriptions.py	100.00% <ø> (ø)
ukrdc_stats/utils.py	60.00% <33.33%> (ø)
ukrdc_stats/calculators/demographics.py	88.40% <76.47%> (-4.93%)	⬇️
ukrdc_stats/calculators/dialysis.py	92.91% <96.77%> (+1.32%)	⬆️
ukrdc_stats/models/generic_2d.py	100.00% <100.00%> (ø)
ukrdc_stats/models/networks.py	0.00% <0.00%> (-100.00%)	⬇️
ukrdc_stats/__init__.py	100.00% <0.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[philmain28]

This is nearly ready to merge and contains a fairly substantial set of changes (major release?). I suspect there is going to be a big performance hit associated with it.

Still left to do:

• Profile the speed
• Update the descriptions to accurately reflect the current status of the code

[jtc42]

If you could profile the performance hit before merging that'd be ideal. We can't afford a substantial hit to the dialysis stats without potentially requiring pre-computing. Which we can do, but I'd love not to have to for now until we have a better idea of how people will use this.

[philmain28]

dialysis stats now calculates in 2-20 secs (on my desktop but it would probably be more on the app server). There also seems to be quite a lot of variation based on the postgres optimizations and how heavily the database is being used.

I will think about ways to speed up the code but any suggestions would be welcome.

[philmain28]

Summary of changes

dialysis calculator

• Sankey plots removed and replaced with pie charts (or more precisely the underlying pydantic data structures have been switched out)
• Stats are now calculated for each of the satellite units
• Replaced some of the hard coded things with lookups

demographics

• Included look up against staging

demo

• Updated dialysis notebook to demo new data structure
• Included Benchmarks in notebook and added various switches into the calculators to allow benchmarking