Skip to content

Instantly share code, notes, and snippets.

@bradyneal
Last active April 4, 2023 16:29
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save bradyneal/4e6c2aabab1d6759f42655fad85ccedb to your computer and use it in GitHub Desktop.
Save bradyneal/4e6c2aabab1d6759f42655fad85ccedb to your computer and use it in GitHub Desktop.
Output of autorank (including Nemenyi test) of the positivity experiments for only the methods that Nemenyi could not statistically distinguish from the best method
method meanrank median mad ci_lower ci_upper effect_size magnitude
238 prop clip_xgboost_flaml_0.005 78.99107142857143 0.9145 0.08550000000000002 0.8045 0.988 -10.202546836008775 large
239 prop clip percentile_decision_tree_tuned_0.1 73.58035714285714 0.971 0.029000000000000026 0.906 0.989 -31.93833221842974 large
240 prop clip percentile_decision_tree_tuned_0.05 72.91428571428571 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
241 prop clip percentile_decision_tree_tuned_0.01 72.89285714285714 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
242 prop clip percentile_decision_tree_tuned_0.0001 72.89285714285714 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
243 prop clip percentile_decision_tree_tuned_0.0005 72.89285714285714 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
244 prop clip percentile_decision_tree_tuned_0.001 72.89285714285714 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
245 prop clip percentile_decision_tree_tuned_0.005 72.89285714285714 0.971 0.029000000000000026 0.908 0.989 -31.93833221842974 large
246 prop clip_lgbm_calib_0.01 70.975 0.999 0.0010000000000000009 0.992 1.0 -952.9201057673821 large
247 prop clip_xgboost_flaml_calib_0.01 69.30178571428571 0.9985 0.0014999999999999458 0.971 1.0 -634.9621125183626 large
248 prop clip_lgbm_0.001 68.72142857142858 0.99275 0.007249999999999979 0.78 1.0 -130.61495081296087 large
249 prop clip_xgboost_calib_0.01 68.66428571428571 0.9995 0.0004999999999999449 0.947 1.0 -1906.7940855147233 large
250 prop clip_lgbm_0.0005 66.82321428571429 0.9910000000000001 0.008999999999999897 0.8315 1.0 -105.0321237699352 large
251 prop clip_lgbm_0.0001 66.72678571428571 0.984 0.016000000000000014 0.913 1.0 -58.663249754448444 large
252 prop clip_decision_tree_tuned_0.005 64.05714285714286 0.985 0.015000000000000013 0.9425 0.997 -62.63772467006149 large
253 prop clip_rf_sklearn_calib_0.005 63.69642857142857 1.0 0.0 0.984 1.0 -inf large
254 prop clip_lgbm_flaml_calib_0.005 60.95178571428571 0.9995 0.0004999999999999449 0.9965 1.0 -1906.7940855147233 large
255 prop clip_rf_flaml_calib_0.005 58.39821428571429 1.0 0.0 0.9925 1.0 -inf large
256 prop clip_decision_tree_tuned_0.0001 58.06785714285714 0.9904999999999999 0.009500000000000064 0.972 0.999 -99.45391336205537 large
257 prop clip_decision_tree_tuned_0.001 57.80535714285714 0.991 0.009000000000000008 0.9725 0.999 -105.03212376993392 large
258 prop clip_decision_tree_tuned_0.0005 57.425 0.991 0.009000000000000008 0.9725 0.999 -105.03212376993392 large
259 prop clip_lgbm_calib_0.005 55.52321428571429 1.0 0.0 0.9985 1.0 -inf large
260 prop clip_xgboost_flaml_calib_0.005 50.00535714285714 1.0 0.0 0.999 1.0 -inf large
261 prop clip_xgboost_calib_0.005 49.33571428571429 1.0 0.0 0.9995 1.0 -inf large
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment