Functional

[ljwolf]

This is a bit of metaprogramming that re-rexpresses most of the esda stats classes as functions that just return the statistic and a simulated p-value.

Still needs work with the LOSH and Spatial_Pearson statistics, but this is one of the two APIs I think we should adopt. The other is the LOSH/Spatial_Pearson-style sklearn estimator.

[codecov-commenter]

Codecov Report

Merging #189 (fc7a4d5) into master (4a63e0b) will decrease coverage by 2.71%. The diff coverage is 21.60%.

@@            Coverage Diff             @@
##           master     #189      +/-   ##
==========================================
- Coverage   80.59%   77.88%   -2.72%     
==========================================
  Files          42       45       +3     
  Lines        4277     4444     +167     
==========================================
+ Hits         3447     3461      +14     
- Misses        830      983     +153     

Impacted Files	Coverage Δ
esda/crand_perf_nogit.py	0.00% <0.00%> (ø)
esda/crand_perf_sim.py	0.00% <0.00%> (ø)
esda/functions.py	0.00% <0.00%> (ø)
esda/lee.py	21.87% <60.00%> (+1.44%)	:arrow_up:
esda/losh.py	88.09% <76.92%> (-1.65%)	:arrow_down:
esda/geary_local.py	86.66% <80.00%> (-1.06%)	:arrow_down:
esda/join_counts_local.py	98.00% <80.00%> (-2.00%)	:arrow_down:
esda/join_counts_local_bv.py	96.29% <80.00%> (-1.14%)	:arrow_down:
esda/join_counts_local_mv.py	98.11% <80.00%> (-1.89%)	:arrow_down:
esda/geary_local_mv.py	98.46% <93.33%> (-1.54%)	:arrow_down:
... and 3 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 4a63e0b...fc7a4d5. Read the comment docs.