mirror of
https://github.com/hanzoai/papers.git
synced 2026-10-22 17:20:17 +00:00
Kai: capability cells filled from the committed runs, and Enso read where it writes
results-data.tex regenerated from hanzoai/benchmarks de99c3e: cardinality, invariance, joint programs, sensors and the original rows for Laya and Jev (729 keys). The training guard reads the current manifest: 2,383,278 training rows, 4,099 excluded. runs.json now names $BENCH/enso/results, where the Enso harness writes. 494 cells still print XXX: Kai's trained checkpoint, speed at 1 to 1,000 questions, and the Enso studies. Co-authored-by: Hanzo Dev <[email protected]>
This commit is contained in:
+1
-1
@@ -6,5 +6,5 @@
|
||||
"mix": "$PAPER/data/mix.json",
|
||||
"parity": "$BENCH/decision/results/laya/parity.json",
|
||||
"cap": "$BENCH/decision/capability/results",
|
||||
"enso": "$BENCH/decision/enso/results"
|
||||
"enso": "$BENCH/enso/results"
|
||||
}
|
||||
|
||||
+731
-4
@@ -2949,8 +2949,8 @@
|
||||
\setres{kj/jev/errors}{1}
|
||||
\setres{kj/guard/commit}{e95c6ee}
|
||||
\setres{kj/guard/units}{11,129}
|
||||
\setres{kj/guard/excluded}{4,086}
|
||||
\setres{kj/guard/train}{961,597}
|
||||
\setres{kj/guard/excluded}{4,099}
|
||||
\setres{kj/guard/train}{2,383,278}
|
||||
\setres{kj/agnews/kai/trained}{split}
|
||||
\setres{kj/agnews/kai/trainrows}{119,918}
|
||||
\setres{kj/emotion/kai/trained}{split}
|
||||
@@ -2975,6 +2975,44 @@
|
||||
\setres{kj/typeddecisions/kai/trainrows}{1,185}
|
||||
\setres{kj/massive/kai/trained}{split}
|
||||
\setres{kj/massive/kai/trainrows}{584,562}
|
||||
\setres{kj/done/kai/trained}{split}
|
||||
\setres{kj/done/kai/trainrows}{3,640}
|
||||
\setres{kj/repo/kai/trained}{split}
|
||||
\setres{kj/repo/kai/trainrows}{686}
|
||||
\setres{kj/risk/kai/trained}{split}
|
||||
\setres{kj/risk/kai/trainrows}{41,070}
|
||||
\setres{kj/tool/kai/trained}{split}
|
||||
\setres{kj/tool/kai/trainrows}{90,213}
|
||||
\setres{kj/function/kai/trained}{split}
|
||||
\setres{kj/function/kai/trainrows}{121,548}
|
||||
\setres{kj/clinc150/kai/trained}{split}
|
||||
\setres{kj/clinc150/kai/trainrows}{15,240}
|
||||
\setres{kj/xnli/kai/trained}{split}
|
||||
\setres{kj/xnli/kai/trainrows}{299,965}
|
||||
\setres{kj/sst5/kai/trained}{split}
|
||||
\setres{kj/sst5/kai/trainrows}{8,532}
|
||||
\setres{kj/boolq/kai/trained}{split}
|
||||
\setres{kj/boolq/kai/trainrows}{9,427}
|
||||
\setres{kj/mnli/kai/trained}{split}
|
||||
\setres{kj/mnli/kai/trainrows}{39,998}
|
||||
\setres{kj/action/kai/trained}{split}
|
||||
\setres{kj/action/kai/trainrows}{288,418}
|
||||
\setres{kj/resolve/kai/trained}{split}
|
||||
\setres{kj/resolve/kai/trainrows}{209,672}
|
||||
\setres{kj/nemotron/kai/trained}{split}
|
||||
\setres{kj/nemotron/kai/trainrows}{170,392}
|
||||
\setres{kj/alias/kai/trained}{split}
|
||||
\setres{kj/alias/kai/trainrows}{20,429}
|
||||
\setres{kj/near/kai/trained}{split}
|
||||
\setres{kj/near/kai/trainrows}{20,000}
|
||||
\setres{kj/shuffle/kai/trained}{split}
|
||||
\setres{kj/shuffle/kai/trainrows}{33,097}
|
||||
\setres{kj/subset/kai/trained}{split}
|
||||
\setres{kj/subset/kai/trainrows}{20,000}
|
||||
\setres{kj/swap/kai/trained}{split}
|
||||
\setres{kj/swap/kai/trainrows}{20,429}
|
||||
\setres{kj/codeswitch/kai/trained}{split}
|
||||
\setres{kj/codeswitch/kai/trainrows}{49,999}
|
||||
\setres{kj/agnews/laya/trained}{mix}
|
||||
\setres{kj/emotion/laya/trained}{no}
|
||||
\setres{kj/banking77full/laya/trained}{no}
|
||||
@@ -3036,5 +3074,694 @@
|
||||
\setres{td/agent/qt/noul}{0.857}
|
||||
\setres{td/agent/qt/score}{0.723}
|
||||
% absent: $BENCH/decision/results/laya/parity.json
|
||||
% absent: $BENCH/decision/capability/results
|
||||
% absent: $BENCH/decision/enso/results
|
||||
\setres{cap/cardinality/jev/k1000.accall}{0.000}
|
||||
\setres{cap/cardinality/jev/k1000.recall1}{0.000}
|
||||
\setres{cap/cardinality/jev/k1000.recall20}{0.000}
|
||||
\setres{cap/cardinality/jev/k1000.recall5}{0.000}
|
||||
\setres{cap/cardinality/jev/k1000.rejected}{1.000}
|
||||
\setres{cap/cardinality/jev/k1000.sl.accall}{0.060}
|
||||
\setres{cap/cardinality/jev/k1000.sl.accuracy}{0.060}
|
||||
\setres{cap/cardinality/jev/k1000.sl.brier}{1.394}
|
||||
\setres{cap/cardinality/jev/k1000.sl.ece}{0.566}
|
||||
\setres{cap/cardinality/jev/k1000.sl.recall1}{0.060}
|
||||
\setres{cap/cardinality/jev/k1000.sl.recall20}{0.905}
|
||||
\setres{cap/cardinality/jev/k1000.sl.recall5}{0.300}
|
||||
\setres{cap/cardinality/jev/k1000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k1000.unanswered}{200}
|
||||
\setres{cap/cardinality/jev/k1000.usd}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.accall}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.recall1}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.recall20}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.recall5}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.rejected}{1.000}
|
||||
\setres{cap/cardinality/jev/k10000.sl.accall}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.sl.accuracy}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.sl.brier}{1.487}
|
||||
\setres{cap/cardinality/jev/k10000.sl.ece}{0.609}
|
||||
\setres{cap/cardinality/jev/k10000.sl.recall1}{0.000}
|
||||
\setres{cap/cardinality/jev/k10000.sl.recall20}{0.890}
|
||||
\setres{cap/cardinality/jev/k10000.sl.recall5}{0.255}
|
||||
\setres{cap/cardinality/jev/k10000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k10000.unanswered}{200}
|
||||
\setres{cap/cardinality/jev/k10000.usd}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.accall}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.recall1}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.recall20}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.recall5}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.rejected}{1.000}
|
||||
\setres{cap/cardinality/jev/k100000.sl.accall}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.sl.accuracy}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.sl.brier}{1.502}
|
||||
\setres{cap/cardinality/jev/k100000.sl.ece}{0.637}
|
||||
\setres{cap/cardinality/jev/k100000.sl.recall1}{0.000}
|
||||
\setres{cap/cardinality/jev/k100000.sl.recall20}{0.980}
|
||||
\setres{cap/cardinality/jev/k100000.sl.recall5}{0.260}
|
||||
\setres{cap/cardinality/jev/k100000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k100000.unanswered}{50}
|
||||
\setres{cap/cardinality/jev/k100000.usd}{0.000}
|
||||
\setres{cap/cardinality/jev/k150.accall}{0.940}
|
||||
\setres{cap/cardinality/jev/k150.accuracy}{0.940}
|
||||
\setres{cap/cardinality/jev/k150.brier}{0.106}
|
||||
\setres{cap/cardinality/jev/k150.ece}{0.032}
|
||||
\setres{cap/cardinality/jev/k150.p50ms}{267}
|
||||
\setres{cap/cardinality/jev/k150.p95ms}{360}
|
||||
\setres{cap/cardinality/jev/k150.recall1}{0.940}
|
||||
\setres{cap/cardinality/jev/k150.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k150.recall5}{0.998}
|
||||
\setres{cap/cardinality/jev/k150.rejected}{0.000}
|
||||
\setres{cap/cardinality/jev/k150.sl.accall}{0.193}
|
||||
\setres{cap/cardinality/jev/k150.sl.accuracy}{0.193}
|
||||
\setres{cap/cardinality/jev/k150.sl.brier}{1.268}
|
||||
\setres{cap/cardinality/jev/k150.sl.ece}{0.551}
|
||||
\setres{cap/cardinality/jev/k150.sl.recall1}{0.193}
|
||||
\setres{cap/cardinality/jev/k150.sl.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k150.sl.recall5}{0.502}
|
||||
\setres{cap/cardinality/jev/k150.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k150.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k150.usd}{0.041}
|
||||
\setres{cap/cardinality/jev/k16.accall}{0.855}
|
||||
\setres{cap/cardinality/jev/k16.accuracy}{0.855}
|
||||
\setres{cap/cardinality/jev/k16.brier}{0.219}
|
||||
\setres{cap/cardinality/jev/k16.ece}{0.075}
|
||||
\setres{cap/cardinality/jev/k16.p50ms}{262}
|
||||
\setres{cap/cardinality/jev/k16.p95ms}{387}
|
||||
\setres{cap/cardinality/jev/k16.recall1}{0.858}
|
||||
\setres{cap/cardinality/jev/k16.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k16.recall5}{0.998}
|
||||
\setres{cap/cardinality/jev/k16.rejected}{0.000}
|
||||
\setres{cap/cardinality/jev/k16.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k16.usd}{0.010}
|
||||
\setres{cap/cardinality/jev/k4.accall}{0.902}
|
||||
\setres{cap/cardinality/jev/k4.accuracy}{0.902}
|
||||
\setres{cap/cardinality/jev/k4.brier}{0.139}
|
||||
\setres{cap/cardinality/jev/k4.ece}{0.050}
|
||||
\setres{cap/cardinality/jev/k4.p50ms}{260}
|
||||
\setres{cap/cardinality/jev/k4.p95ms}{379}
|
||||
\setres{cap/cardinality/jev/k4.recall1}{0.905}
|
||||
\setres{cap/cardinality/jev/k4.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k4.recall5}{1.000}
|
||||
\setres{cap/cardinality/jev/k4.rejected}{0.000}
|
||||
\setres{cap/cardinality/jev/k4.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k4.usd}{0.006}
|
||||
\setres{cap/cardinality/jev/k77.accall}{0.838}
|
||||
\setres{cap/cardinality/jev/k77.accuracy}{0.838}
|
||||
\setres{cap/cardinality/jev/k77.brier}{0.256}
|
||||
\setres{cap/cardinality/jev/k77.ece}{0.081}
|
||||
\setres{cap/cardinality/jev/k77.p50ms}{263}
|
||||
\setres{cap/cardinality/jev/k77.p95ms}{366}
|
||||
\setres{cap/cardinality/jev/k77.recall1}{0.838}
|
||||
\setres{cap/cardinality/jev/k77.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k77.recall5}{0.990}
|
||||
\setres{cap/cardinality/jev/k77.rejected}{0.000}
|
||||
\setres{cap/cardinality/jev/k77.sl.accall}{0.077}
|
||||
\setres{cap/cardinality/jev/k77.sl.accuracy}{0.077}
|
||||
\setres{cap/cardinality/jev/k77.sl.brier}{1.494}
|
||||
\setres{cap/cardinality/jev/k77.sl.ece}{0.667}
|
||||
\setres{cap/cardinality/jev/k77.sl.recall1}{0.077}
|
||||
\setres{cap/cardinality/jev/k77.sl.recall20}{1.000}
|
||||
\setres{cap/cardinality/jev/k77.sl.recall5}{0.495}
|
||||
\setres{cap/cardinality/jev/k77.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k77.unanswered}{0}
|
||||
\setres{cap/cardinality/jev/k77.usd}{0.027}
|
||||
\setres{cap/cardinality/laya/k1000.accall}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.recall1}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.recall20}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.recall5}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.rejected}{1.000}
|
||||
\setres{cap/cardinality/laya/k1000.sl.accall}{0.040}
|
||||
\setres{cap/cardinality/laya/k1000.sl.accuracy}{0.040}
|
||||
\setres{cap/cardinality/laya/k1000.sl.brier}{1.482}
|
||||
\setres{cap/cardinality/laya/k1000.sl.casems}{28.1}
|
||||
\setres{cap/cardinality/laya/k1000.sl.ece}{0.613}
|
||||
\setres{cap/cardinality/laya/k1000.sl.hit}{0.060}
|
||||
\setres{cap/cardinality/laya/k1000.sl.indexs}{4.872}
|
||||
\setres{cap/cardinality/laya/k1000.sl.msperq}{28.0}
|
||||
\setres{cap/cardinality/laya/k1000.sl.recall1}{0.040}
|
||||
\setres{cap/cardinality/laya/k1000.sl.recall20}{0.060}
|
||||
\setres{cap/cardinality/laya/k1000.sl.recall5}{0.050}
|
||||
\setres{cap/cardinality/laya/k1000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k1000.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.unanswered}{200}
|
||||
\setres{cap/cardinality/laya/k1000.wide.accall}{0.055}
|
||||
\setres{cap/cardinality/laya/k1000.wide.accuracy}{0.055}
|
||||
\setres{cap/cardinality/laya/k1000.wide.brier}{1.209}
|
||||
\setres{cap/cardinality/laya/k1000.wide.ece}{0.414}
|
||||
\setres{cap/cardinality/laya/k1000.wide.msperq}{3458}
|
||||
\setres{cap/cardinality/laya/k1000.wide.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.wide.recall1}{0.055}
|
||||
\setres{cap/cardinality/laya/k1000.wide.recall20}{0.115}
|
||||
\setres{cap/cardinality/laya/k1000.wide.recall5}{0.090}
|
||||
\setres{cap/cardinality/laya/k1000.wide.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k1000.wide.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k10000.accall}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.recall1}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.recall20}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.recall5}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.rejected}{1.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.accall}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.accuracy}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.brier}{1.504}
|
||||
\setres{cap/cardinality/laya/k10000.sl.casems}{49.2}
|
||||
\setres{cap/cardinality/laya/k10000.sl.ece}{0.608}
|
||||
\setres{cap/cardinality/laya/k10000.sl.hit}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.indexs}{33.7}
|
||||
\setres{cap/cardinality/laya/k10000.sl.msperq}{29.0}
|
||||
\setres{cap/cardinality/laya/k10000.sl.recall1}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.recall20}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.recall5}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k10000.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k10000.unanswered}{200}
|
||||
\setres{cap/cardinality/laya/k100000.accall}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.recall1}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.recall20}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.recall5}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.rejected}{1.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.accall}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.accuracy}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.brier}{1.507}
|
||||
\setres{cap/cardinality/laya/k100000.sl.casems}{722}
|
||||
\setres{cap/cardinality/laya/k100000.sl.ece}{0.613}
|
||||
\setres{cap/cardinality/laya/k100000.sl.hit}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.indexs}{384}
|
||||
\setres{cap/cardinality/laya/k100000.sl.msperq}{56.3}
|
||||
\setres{cap/cardinality/laya/k100000.sl.recall1}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.recall20}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.recall5}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k100000.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k100000.unanswered}{50}
|
||||
\setres{cap/cardinality/laya/k150.accall}{0.610}
|
||||
\setres{cap/cardinality/laya/k150.accuracy}{0.610}
|
||||
\setres{cap/cardinality/laya/k150.brier}{0.698}
|
||||
\setres{cap/cardinality/laya/k150.ece}{0.326}
|
||||
\setres{cap/cardinality/laya/k150.msperq}{186}
|
||||
\setres{cap/cardinality/laya/k150.optionscut}{1.000}
|
||||
\setres{cap/cardinality/laya/k150.recall1}{0.610}
|
||||
\setres{cap/cardinality/laya/k150.recall20}{0.975}
|
||||
\setres{cap/cardinality/laya/k150.recall5}{0.920}
|
||||
\setres{cap/cardinality/laya/k150.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k150.sl.accall}{0.193}
|
||||
\setres{cap/cardinality/laya/k150.sl.accuracy}{0.193}
|
||||
\setres{cap/cardinality/laya/k150.sl.brier}{1.384}
|
||||
\setres{cap/cardinality/laya/k150.sl.casems}{39.0}
|
||||
\setres{cap/cardinality/laya/k150.sl.ece}{0.629}
|
||||
\setres{cap/cardinality/laya/k150.sl.hit}{0.200}
|
||||
\setres{cap/cardinality/laya/k150.sl.indexs}{0.404}
|
||||
\setres{cap/cardinality/laya/k150.sl.msperq}{38.4}
|
||||
\setres{cap/cardinality/laya/k150.sl.recall1}{0.193}
|
||||
\setres{cap/cardinality/laya/k150.sl.recall20}{0.200}
|
||||
\setres{cap/cardinality/laya/k150.sl.recall5}{0.200}
|
||||
\setres{cap/cardinality/laya/k150.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k150.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k150.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k150.wide.accall}{0.657}
|
||||
\setres{cap/cardinality/laya/k150.wide.accuracy}{0.657}
|
||||
\setres{cap/cardinality/laya/k150.wide.brier}{0.620}
|
||||
\setres{cap/cardinality/laya/k150.wide.ece}{0.288}
|
||||
\setres{cap/cardinality/laya/k150.wide.msperq}{192}
|
||||
\setres{cap/cardinality/laya/k150.wide.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k150.wide.recall1}{0.657}
|
||||
\setres{cap/cardinality/laya/k150.wide.recall20}{0.983}
|
||||
\setres{cap/cardinality/laya/k150.wide.recall5}{0.950}
|
||||
\setres{cap/cardinality/laya/k150.wide.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k150.wide.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k16.accall}{0.670}
|
||||
\setres{cap/cardinality/laya/k16.accuracy}{0.670}
|
||||
\setres{cap/cardinality/laya/k16.brier}{0.574}
|
||||
\setres{cap/cardinality/laya/k16.ece}{0.250}
|
||||
\setres{cap/cardinality/laya/k16.msperq}{35.5}
|
||||
\setres{cap/cardinality/laya/k16.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k16.recall1}{0.670}
|
||||
\setres{cap/cardinality/laya/k16.recall20}{1.000}
|
||||
\setres{cap/cardinality/laya/k16.recall5}{0.892}
|
||||
\setres{cap/cardinality/laya/k16.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k16.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k16.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k4.accall}{0.777}
|
||||
\setres{cap/cardinality/laya/k4.accuracy}{0.777}
|
||||
\setres{cap/cardinality/laya/k4.brier}{0.323}
|
||||
\setres{cap/cardinality/laya/k4.ece}{0.060}
|
||||
\setres{cap/cardinality/laya/k4.msperq}{22.2}
|
||||
\setres{cap/cardinality/laya/k4.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k4.recall1}{0.777}
|
||||
\setres{cap/cardinality/laya/k4.recall20}{1.000}
|
||||
\setres{cap/cardinality/laya/k4.recall5}{1.000}
|
||||
\setres{cap/cardinality/laya/k4.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k4.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k4.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k77.accall}{0.425}
|
||||
\setres{cap/cardinality/laya/k77.accuracy}{0.425}
|
||||
\setres{cap/cardinality/laya/k77.brier}{0.925}
|
||||
\setres{cap/cardinality/laya/k77.ece}{0.381}
|
||||
\setres{cap/cardinality/laya/k77.msperq}{110}
|
||||
\setres{cap/cardinality/laya/k77.optionscut}{1.000}
|
||||
\setres{cap/cardinality/laya/k77.recall1}{0.425}
|
||||
\setres{cap/cardinality/laya/k77.recall20}{0.877}
|
||||
\setres{cap/cardinality/laya/k77.recall5}{0.698}
|
||||
\setres{cap/cardinality/laya/k77.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k77.sl.accall}{0.075}
|
||||
\setres{cap/cardinality/laya/k77.sl.accuracy}{0.075}
|
||||
\setres{cap/cardinality/laya/k77.sl.brier}{1.368}
|
||||
\setres{cap/cardinality/laya/k77.sl.casems}{31.8}
|
||||
\setres{cap/cardinality/laya/k77.sl.ece}{0.550}
|
||||
\setres{cap/cardinality/laya/k77.sl.hit}{0.080}
|
||||
\setres{cap/cardinality/laya/k77.sl.indexs}{0.277}
|
||||
\setres{cap/cardinality/laya/k77.sl.msperq}{47.6}
|
||||
\setres{cap/cardinality/laya/k77.sl.recall1}{0.075}
|
||||
\setres{cap/cardinality/laya/k77.sl.recall20}{0.080}
|
||||
\setres{cap/cardinality/laya/k77.sl.recall5}{0.080}
|
||||
\setres{cap/cardinality/laya/k77.sl.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k77.statecut}{0.000}
|
||||
\setres{cap/cardinality/laya/k77.unanswered}{0}
|
||||
\setres{cap/cardinality/laya/k77.wide.accall}{0.532}
|
||||
\setres{cap/cardinality/laya/k77.wide.accuracy}{0.532}
|
||||
\setres{cap/cardinality/laya/k77.wide.brier}{0.765}
|
||||
\setres{cap/cardinality/laya/k77.wide.ece}{0.302}
|
||||
\setres{cap/cardinality/laya/k77.wide.msperq}{141}
|
||||
\setres{cap/cardinality/laya/k77.wide.optionscut}{0.000}
|
||||
\setres{cap/cardinality/laya/k77.wide.recall1}{0.532}
|
||||
\setres{cap/cardinality/laya/k77.wide.recall20}{0.912}
|
||||
\setres{cap/cardinality/laya/k77.wide.recall5}{0.828}
|
||||
\setres{cap/cardinality/laya/k77.wide.rejected}{0.000}
|
||||
\setres{cap/cardinality/laya/k77.wide.unanswered}{0}
|
||||
\setres{cap/invariance/jev/alias.code.acc}{0.850}
|
||||
\setres{cap/invariance/jev/alias.code.flip}{0.040}
|
||||
\setres{cap/invariance/jev/alias.snake.acc}{0.850}
|
||||
\setres{cap/invariance/jev/alias.snake.flip}{0.015}
|
||||
\setres{cap/invariance/jev/alias.text.acc}{0.860}
|
||||
\setres{cap/invariance/jev/perm.acc.max}{0.870}
|
||||
\setres{cap/invariance/jev/perm.acc.min}{0.850}
|
||||
\setres{cap/invariance/jev/perm.compared}{200}
|
||||
\setres{cap/invariance/jev/perm.flip}{0.055}
|
||||
\setres{cap/invariance/jev/perm.maxdp}{0.450}
|
||||
\setres{cap/invariance/jev/perm.tv}{0.032}
|
||||
\setres{cap/invariance/jev/swap.acc}{0.823}
|
||||
\setres{cap/invariance/jev/swap.desc.acc}{0.200}
|
||||
\setres{cap/invariance/jev/swap.desc.compared}{400}
|
||||
\setres{cap/invariance/jev/swap.desc.gap}{0.552}
|
||||
\setres{cap/invariance/jev/swap.desc.maxgap}{0.970}
|
||||
\setres{cap/invariance/jev/swap.desc.mirror}{0.058}
|
||||
\setres{cap/invariance/jev/swap.neg.acc}{0.835}
|
||||
\setres{cap/invariance/jev/swap.neg.compared}{400}
|
||||
\setres{cap/invariance/jev/swap.neg.gap}{0.091}
|
||||
\setres{cap/invariance/jev/swap.neg.maxgap}{0.470}
|
||||
\setres{cap/invariance/jev/swap.neg.mirror}{0.887}
|
||||
\setres{cap/invariance/jev/unanswered}{0}
|
||||
\setres{cap/invariance/laya/alias.code.acc}{0.650}
|
||||
\setres{cap/invariance/laya/alias.code.flip}{0.205}
|
||||
\setres{cap/invariance/laya/alias.snake.acc}{0.610}
|
||||
\setres{cap/invariance/laya/alias.snake.flip}{0.210}
|
||||
\setres{cap/invariance/laya/alias.text.acc}{0.625}
|
||||
\setres{cap/invariance/laya/perm.acc.max}{0.650}
|
||||
\setres{cap/invariance/laya/perm.acc.min}{0.620}
|
||||
\setres{cap/invariance/laya/perm.compared}{200}
|
||||
\setres{cap/invariance/laya/perm.flip}{0.255}
|
||||
\setres{cap/invariance/laya/perm.maxdp}{0.996}
|
||||
\setres{cap/invariance/laya/perm.tv}{0.144}
|
||||
\setres{cap/invariance/laya/swap.acc}{0.815}
|
||||
\setres{cap/invariance/laya/swap.desc.acc}{0.310}
|
||||
\setres{cap/invariance/laya/swap.desc.compared}{400}
|
||||
\setres{cap/invariance/laya/swap.desc.gap}{0.510}
|
||||
\setres{cap/invariance/laya/swap.desc.maxgap}{1.000}
|
||||
\setres{cap/invariance/laya/swap.desc.mirror}{0.205}
|
||||
\setres{cap/invariance/laya/swap.neg.acc}{0.410}
|
||||
\setres{cap/invariance/laya/swap.neg.compared}{400}
|
||||
\setres{cap/invariance/laya/swap.neg.gap}{0.464}
|
||||
\setres{cap/invariance/laya/swap.neg.maxgap}{1.000}
|
||||
\setres{cap/invariance/laya/swap.neg.mirror}{0.300}
|
||||
\setres{cap/invariance/laya/unanswered}{0}
|
||||
\setres{cap/joint/jev/acc}{0.839}
|
||||
\setres{cap/joint/jev/acc.action}{0.760}
|
||||
\setres{cap/joint/jev/acc.evidence}{0.920}
|
||||
\setres{cap/joint/jev/acc.fault}{1.000}
|
||||
\setres{cap/joint/jev/acc.option}{0.975}
|
||||
\setres{cap/joint/jev/acc.readiness}{0.835}
|
||||
\setres{cap/joint/jev/acc.review}{0.603}
|
||||
\setres{cap/joint/jev/acc.risk}{0.930}
|
||||
\setres{cap/joint/jev/acc.severity}{0.930}
|
||||
\setres{cap/joint/jev/exact}{0.443}
|
||||
\setres{cap/joint/jev/maintenance.acc}{0.821}
|
||||
\setres{cap/joint/jev/maintenance.exact}{0.430}
|
||||
\setres{cap/joint/jev/maintenance.proj.acc}{0.887}
|
||||
\setres{cap/joint/jev/maintenance.proj.exact}{0.540}
|
||||
\setres{cap/joint/jev/maintenance.violate}{0.300}
|
||||
\setres{cap/joint/jev/proj.acc}{0.883}
|
||||
\setres{cap/joint/jev/proj.acc.action}{0.960}
|
||||
\setres{cap/joint/jev/proj.acc.evidence}{0.940}
|
||||
\setres{cap/joint/jev/proj.acc.fault}{1.000}
|
||||
\setres{cap/joint/jev/proj.acc.option}{0.980}
|
||||
\setres{cap/joint/jev/proj.acc.readiness}{0.925}
|
||||
\setres{cap/joint/jev/proj.acc.review}{0.618}
|
||||
\setres{cap/joint/jev/proj.acc.risk}{0.910}
|
||||
\setres{cap/joint/jev/proj.acc.severity}{0.960}
|
||||
\setres{cap/joint/jev/proj.exact}{0.570}
|
||||
\setres{cap/joint/jev/proj.violate}{0.000}
|
||||
\setres{cap/joint/jev/trade.acc}{0.857}
|
||||
\setres{cap/joint/jev/trade.exact}{0.455}
|
||||
\setres{cap/joint/jev/trade.proj.acc}{0.880}
|
||||
\setres{cap/joint/jev/trade.proj.exact}{0.600}
|
||||
\setres{cap/joint/jev/trade.violate}{0.195}
|
||||
\setres{cap/joint/jev/unanswered}{0}
|
||||
\setres{cap/joint/jev/violate}{0.247}
|
||||
\setres{cap/joint/laya/acc}{0.336}
|
||||
\setres{cap/joint/laya/acc.action}{0.280}
|
||||
\setres{cap/joint/laya/acc.evidence}{0.300}
|
||||
\setres{cap/joint/laya/acc.fault}{0.325}
|
||||
\setres{cap/joint/laya/acc.option}{0.315}
|
||||
\setres{cap/joint/laya/acc.readiness}{0.328}
|
||||
\setres{cap/joint/laya/acc.review}{0.482}
|
||||
\setres{cap/joint/laya/acc.risk}{0.260}
|
||||
\setres{cap/joint/laya/acc.severity}{0.260}
|
||||
\setres{cap/joint/laya/en.acc}{0.320}
|
||||
\setres{cap/joint/laya/en.acc.action}{0.100}
|
||||
\setres{cap/joint/laya/en.acc.evidence}{0.290}
|
||||
\setres{cap/joint/laya/en.acc.fault}{0.300}
|
||||
\setres{cap/joint/laya/en.acc.option}{0.275}
|
||||
\setres{cap/joint/laya/en.acc.readiness}{0.352}
|
||||
\setres{cap/joint/laya/en.acc.review}{0.482}
|
||||
\setres{cap/joint/laya/en.acc.risk}{0.280}
|
||||
\setres{cap/joint/laya/en.acc.severity}{0.285}
|
||||
\setres{cap/joint/laya/en.exact}{0.000}
|
||||
\setres{cap/joint/laya/en.maintenance.acc}{0.305}
|
||||
\setres{cap/joint/laya/en.maintenance.exact}{0.000}
|
||||
\setres{cap/joint/laya/en.maintenance.proj.acc}{0.354}
|
||||
\setres{cap/joint/laya/en.maintenance.proj.exact}{0.005}
|
||||
\setres{cap/joint/laya/en.maintenance.violate}{0.910}
|
||||
\setres{cap/joint/laya/en.proj.acc}{0.344}
|
||||
\setres{cap/joint/laya/en.proj.acc.action}{0.320}
|
||||
\setres{cap/joint/laya/en.proj.acc.evidence}{0.290}
|
||||
\setres{cap/joint/laya/en.proj.acc.fault}{0.275}
|
||||
\setres{cap/joint/laya/en.proj.acc.option}{0.275}
|
||||
\setres{cap/joint/laya/en.proj.acc.readiness}{0.360}
|
||||
\setres{cap/joint/laya/en.proj.acc.review}{0.482}
|
||||
\setres{cap/joint/laya/en.proj.acc.risk}{0.280}
|
||||
\setres{cap/joint/laya/en.proj.acc.severity}{0.320}
|
||||
\setres{cap/joint/laya/en.proj.exact}{0.003}
|
||||
\setres{cap/joint/laya/en.proj.violate}{0.000}
|
||||
\setres{cap/joint/laya/en.trade.acc}{0.335}
|
||||
\setres{cap/joint/laya/en.trade.exact}{0.000}
|
||||
\setres{cap/joint/laya/en.trade.proj.acc}{0.335}
|
||||
\setres{cap/joint/laya/en.trade.proj.exact}{0.000}
|
||||
\setres{cap/joint/laya/en.trade.violate}{0.030}
|
||||
\setres{cap/joint/laya/en.unanswered}{0}
|
||||
\setres{cap/joint/laya/en.violate}{0.470}
|
||||
\setres{cap/joint/laya/exact}{0.005}
|
||||
\setres{cap/joint/laya/maintenance.acc}{0.329}
|
||||
\setres{cap/joint/laya/maintenance.exact}{0.000}
|
||||
\setres{cap/joint/laya/maintenance.proj.acc}{0.302}
|
||||
\setres{cap/joint/laya/maintenance.proj.exact}{0.070}
|
||||
\setres{cap/joint/laya/maintenance.violate}{0.960}
|
||||
\setres{cap/joint/laya/proj.acc}{0.328}
|
||||
\setres{cap/joint/laya/proj.acc.action}{0.220}
|
||||
\setres{cap/joint/laya/proj.acc.evidence}{0.300}
|
||||
\setres{cap/joint/laya/proj.acc.fault}{0.300}
|
||||
\setres{cap/joint/laya/proj.acc.option}{0.315}
|
||||
\setres{cap/joint/laya/proj.acc.readiness}{0.335}
|
||||
\setres{cap/joint/laya/proj.acc.review}{0.482}
|
||||
\setres{cap/joint/laya/proj.acc.risk}{0.285}
|
||||
\setres{cap/joint/laya/proj.acc.severity}{0.220}
|
||||
\setres{cap/joint/laya/proj.exact}{0.040}
|
||||
\setres{cap/joint/laya/proj.violate}{0.000}
|
||||
\setres{cap/joint/laya/trade.acc}{0.343}
|
||||
\setres{cap/joint/laya/trade.exact}{0.010}
|
||||
\setres{cap/joint/laya/trade.proj.acc}{0.353}
|
||||
\setres{cap/joint/laya/trade.proj.exact}{0.010}
|
||||
\setres{cap/joint/laya/trade.violate}{0.315}
|
||||
\setres{cap/joint/laya/unanswered}{0}
|
||||
\setres{cap/joint/laya/violate}{0.637}
|
||||
\setres{cap/orig/jev/agnews.accuracy}{0.860}
|
||||
\setres{cap/orig/jev/agnews.aurc}{0.083}
|
||||
\setres{cap/orig/jev/agnews.brier}{0.232}
|
||||
\setres{cap/orig/jev/agnews.ece}{0.102}
|
||||
\setres{cap/orig/jev/agnews.macrof1}{0.849}
|
||||
\setres{cap/orig/jev/agnews.nll}{1.805}
|
||||
\setres{cap/orig/jev/agnews.unanswered}{0}
|
||||
\setres{cap/orig/jev/agnews.zeroprob}{0.058}
|
||||
\setres{cap/orig/jev/banking77full.accuracy}{0.835}
|
||||
\setres{cap/orig/jev/banking77full.aurc}{0.058}
|
||||
\setres{cap/orig/jev/banking77full.brier}{0.257}
|
||||
\setres{cap/orig/jev/banking77full.ece}{0.073}
|
||||
\setres{cap/orig/jev/banking77full.macrof1}{0.314}
|
||||
\setres{cap/orig/jev/banking77full.nll}{1.347}
|
||||
\setres{cap/orig/jev/banking77full.unanswered}{0}
|
||||
\setres{cap/orig/jev/banking77full.zeroprob}{0.035}
|
||||
\setres{cap/orig/jev/beats.laya}{56}
|
||||
\setres{cap/orig/jev/emailspam.accuracy}{0.978}
|
||||
\setres{cap/orig/jev/emailspam.aurc}{$7.0 \times 10^{-4}$}
|
||||
\setres{cap/orig/jev/emailspam.brier}{0.038}
|
||||
\setres{cap/orig/jev/emailspam.ece}{0.060}
|
||||
\setres{cap/orig/jev/emailspam.macrof1}{0.977}
|
||||
\setres{cap/orig/jev/emailspam.nll}{0.097}
|
||||
\setres{cap/orig/jev/emailspam.unanswered}{0}
|
||||
\setres{cap/orig/jev/emailspam.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/emotion.accuracy}{0.603}
|
||||
\setres{cap/orig/jev/emotion.aurc}{0.231}
|
||||
\setres{cap/orig/jev/emotion.brier}{0.646}
|
||||
\setres{cap/orig/jev/emotion.ece}{0.280}
|
||||
\setres{cap/orig/jev/emotion.macrof1}{0.515}
|
||||
\setres{cap/orig/jev/emotion.nll}{4.683}
|
||||
\setres{cap/orig/jev/emotion.unanswered}{0}
|
||||
\setres{cap/orig/jev/emotion.zeroprob}{0.142}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.accuracy}{0.940}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.aurc}{0.012}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.brier}{0.104}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.ece}{0.047}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.macrof1}{0.918}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.nll}{0.189}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.unanswered}{0}
|
||||
\setres{cap/orig/jev/guardrailsjailbreak.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/massive.above3xrandom}{51}
|
||||
\setres{cap/orig/jev/massive.languages}{51}
|
||||
\setres{cap/orig/jev/massive.macroaccuracy}{0.890}
|
||||
\setres{cap/orig/jev/massive.macroece}{0.064}
|
||||
\setres{cap/orig/jev/massive.nonenglishmacro}{0.890}
|
||||
\setres{cap/orig/jev/modelroutingdomain.accuracy}{0.977}
|
||||
\setres{cap/orig/jev/modelroutingdomain.aurc}{0.003}
|
||||
\setres{cap/orig/jev/modelroutingdomain.brier}{0.040}
|
||||
\setres{cap/orig/jev/modelroutingdomain.ece}{0.014}
|
||||
\setres{cap/orig/jev/modelroutingdomain.macrof1}{0.592}
|
||||
\setres{cap/orig/jev/modelroutingdomain.nll}{0.085}
|
||||
\setres{cap/orig/jev/modelroutingdomain.unanswered}{0}
|
||||
\setres{cap/orig/jev/modelroutingdomain.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/moderationtoxicity.accuracy}{0.662}
|
||||
\setres{cap/orig/jev/moderationtoxicity.aurc}{0.138}
|
||||
\setres{cap/orig/jev/moderationtoxicity.brier}{0.432}
|
||||
\setres{cap/orig/jev/moderationtoxicity.ece}{0.177}
|
||||
\setres{cap/orig/jev/moderationtoxicity.macrof1}{0.628}
|
||||
\setres{cap/orig/jev/moderationtoxicity.nll}{0.619}
|
||||
\setres{cap/orig/jev/moderationtoxicity.unanswered}{0}
|
||||
\setres{cap/orig/jev/moderationtoxicity.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/phishing.accuracy}{0.900}
|
||||
\setres{cap/orig/jev/phishing.aurc}{0.021}
|
||||
\setres{cap/orig/jev/phishing.brier}{0.152}
|
||||
\setres{cap/orig/jev/phishing.ece}{0.039}
|
||||
\setres{cap/orig/jev/phishing.macrof1}{0.887}
|
||||
\setres{cap/orig/jev/phishing.nll}{0.249}
|
||||
\setres{cap/orig/jev/phishing.unanswered}{0}
|
||||
\setres{cap/orig/jev/phishing.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/ragrelevance.accuracy}{0.620}
|
||||
\setres{cap/orig/jev/ragrelevance.aurc}{0.331}
|
||||
\setres{cap/orig/jev/ragrelevance.brier}{0.628}
|
||||
\setres{cap/orig/jev/ragrelevance.ece}{0.279}
|
||||
\setres{cap/orig/jev/ragrelevance.macrof1}{0.592}
|
||||
\setres{cap/orig/jev/ragrelevance.nll}{1.113}
|
||||
\setres{cap/orig/jev/ragrelevance.unanswered}{0}
|
||||
\setres{cap/orig/jev/ragrelevance.zeroprob}{0.000}
|
||||
\setres{cap/orig/jev/supporttriage.accuracy}{0.365}
|
||||
\setres{cap/orig/jev/supporttriage.aurc}{0.487}
|
||||
\setres{cap/orig/jev/supporttriage.brier}{1.056}
|
||||
\setres{cap/orig/jev/supporttriage.ece}{0.483}
|
||||
\setres{cap/orig/jev/supporttriage.macrof1}{0.368}
|
||||
\setres{cap/orig/jev/supporttriage.nll}{10.1}
|
||||
\setres{cap/orig/jev/supporttriage.unanswered}{0}
|
||||
\setres{cap/orig/jev/supporttriage.zeroprob}{0.335}
|
||||
\setres{cap/orig/jev/typeddecisions.accuracy}{0.736}
|
||||
\setres{cap/orig/jev/typeddecisions.aurc}{0.128}
|
||||
\setres{cap/orig/jev/typeddecisions.brier}{0.366}
|
||||
\setres{cap/orig/jev/typeddecisions.ece}{0.047}
|
||||
\setres{cap/orig/jev/typeddecisions.macrof1}{0.738}
|
||||
\setres{cap/orig/jev/typeddecisions.nll}{0.707}
|
||||
\setres{cap/orig/jev/typeddecisions.unanswered}{0}
|
||||
\setres{cap/orig/jev/typeddecisions.zeroprob}{0.003}
|
||||
\setres{cap/orig/laya/agnews.accuracy}{0.950}
|
||||
\setres{cap/orig/laya/agnews.aurc}{0.007}
|
||||
\setres{cap/orig/laya/agnews.brier}{0.081}
|
||||
\setres{cap/orig/laya/agnews.ece}{0.032}
|
||||
\setres{cap/orig/laya/agnews.macrof1}{0.944}
|
||||
\setres{cap/orig/laya/agnews.nll}{0.164}
|
||||
\setres{cap/orig/laya/agnews.unanswered}{0}
|
||||
\setres{cap/orig/laya/agnews.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/banking77full.accuracy}{0.425}
|
||||
\setres{cap/orig/laya/banking77full.aurc}{0.356}
|
||||
\setres{cap/orig/laya/banking77full.brier}{0.925}
|
||||
\setres{cap/orig/laya/banking77full.ece}{0.381}
|
||||
\setres{cap/orig/laya/banking77full.macrof1}{0.112}
|
||||
\setres{cap/orig/laya/banking77full.nll}{4.797}
|
||||
\setres{cap/orig/laya/banking77full.unanswered}{0}
|
||||
\setres{cap/orig/laya/banking77full.zeroprob}{0.098}
|
||||
\setres{cap/orig/laya/beats.jev}{6}
|
||||
\setres{cap/orig/laya/emailspam.accuracy}{0.998}
|
||||
\setres{cap/orig/laya/emailspam.aurc}{0.000}
|
||||
\setres{cap/orig/laya/emailspam.brier}{0.005}
|
||||
\setres{cap/orig/laya/emailspam.ece}{0.011}
|
||||
\setres{cap/orig/laya/emailspam.macrof1}{0.998}
|
||||
\setres{cap/orig/laya/emailspam.nll}{0.013}
|
||||
\setres{cap/orig/laya/emailspam.unanswered}{0}
|
||||
\setres{cap/orig/laya/emailspam.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/emotion.accuracy}{0.595}
|
||||
\setres{cap/orig/laya/emotion.aurc}{0.266}
|
||||
\setres{cap/orig/laya/emotion.brier}{0.696}
|
||||
\setres{cap/orig/laya/emotion.ece}{0.306}
|
||||
\setres{cap/orig/laya/emotion.macrof1}{0.475}
|
||||
\setres{cap/orig/laya/emotion.nll}{2.019}
|
||||
\setres{cap/orig/laya/emotion.unanswered}{0}
|
||||
\setres{cap/orig/laya/emotion.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.accuracy}{0.705}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.aurc}{0.230}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.brier}{0.543}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.ece}{0.262}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.macrof1}{0.682}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.nll}{1.912}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.unanswered}{0}
|
||||
\setres{cap/orig/laya/guardrailsjailbreak.zeroprob}{0.003}
|
||||
\setres{cap/orig/laya/massive.above3xrandom}{48}
|
||||
\setres{cap/orig/laya/massive.languages}{51}
|
||||
\setres{cap/orig/laya/massive.macroaccuracy}{0.382}
|
||||
\setres{cap/orig/laya/massive.macroece}{0.418}
|
||||
\setres{cap/orig/laya/massive.nonenglishmacro}{0.373}
|
||||
\setres{cap/orig/laya/modelroutingdomain.accuracy}{0.639}
|
||||
\setres{cap/orig/laya/modelroutingdomain.aurc}{0.176}
|
||||
\setres{cap/orig/laya/modelroutingdomain.brier}{0.494}
|
||||
\setres{cap/orig/laya/modelroutingdomain.ece}{0.089}
|
||||
\setres{cap/orig/laya/modelroutingdomain.macrof1}{0.320}
|
||||
\setres{cap/orig/laya/modelroutingdomain.nll}{1.089}
|
||||
\setres{cap/orig/laya/modelroutingdomain.unanswered}{0}
|
||||
\setres{cap/orig/laya/modelroutingdomain.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/moderationtoxicity.accuracy}{0.530}
|
||||
\setres{cap/orig/laya/moderationtoxicity.aurc}{0.281}
|
||||
\setres{cap/orig/laya/moderationtoxicity.brier}{0.602}
|
||||
\setres{cap/orig/laya/moderationtoxicity.ece}{0.297}
|
||||
\setres{cap/orig/laya/moderationtoxicity.macrof1}{0.400}
|
||||
\setres{cap/orig/laya/moderationtoxicity.nll}{0.886}
|
||||
\setres{cap/orig/laya/moderationtoxicity.unanswered}{0}
|
||||
\setres{cap/orig/laya/moderationtoxicity.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/phishing.accuracy}{0.983}
|
||||
\setres{cap/orig/laya/phishing.aurc}{$5.0 \times 10^{-4}$}
|
||||
\setres{cap/orig/laya/phishing.brier}{0.026}
|
||||
\setres{cap/orig/laya/phishing.ece}{0.010}
|
||||
\setres{cap/orig/laya/phishing.macrof1}{0.982}
|
||||
\setres{cap/orig/laya/phishing.nll}{0.044}
|
||||
\setres{cap/orig/laya/phishing.unanswered}{0}
|
||||
\setres{cap/orig/laya/phishing.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/ragrelevance.accuracy}{0.625}
|
||||
\setres{cap/orig/laya/ragrelevance.aurc}{0.280}
|
||||
\setres{cap/orig/laya/ragrelevance.brier}{0.466}
|
||||
\setres{cap/orig/laya/ragrelevance.ece}{0.117}
|
||||
\setres{cap/orig/laya/ragrelevance.macrof1}{0.623}
|
||||
\setres{cap/orig/laya/ragrelevance.nll}{0.671}
|
||||
\setres{cap/orig/laya/ragrelevance.unanswered}{0}
|
||||
\setres{cap/orig/laya/ragrelevance.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/supporttriage.accuracy}{0.502}
|
||||
\setres{cap/orig/laya/supporttriage.aurc}{0.321}
|
||||
\setres{cap/orig/laya/supporttriage.brier}{0.644}
|
||||
\setres{cap/orig/laya/supporttriage.ece}{0.091}
|
||||
\setres{cap/orig/laya/supporttriage.macrof1}{0.480}
|
||||
\setres{cap/orig/laya/supporttriage.nll}{1.424}
|
||||
\setres{cap/orig/laya/supporttriage.unanswered}{0}
|
||||
\setres{cap/orig/laya/supporttriage.zeroprob}{0.000}
|
||||
\setres{cap/orig/laya/typeddecisions.accuracy}{0.766}
|
||||
\setres{cap/orig/laya/typeddecisions.aurc}{0.099}
|
||||
\setres{cap/orig/laya/typeddecisions.brier}{0.400}
|
||||
\setres{cap/orig/laya/typeddecisions.ece}{0.213}
|
||||
\setres{cap/orig/laya/typeddecisions.macrof1}{0.750}
|
||||
\setres{cap/orig/laya/typeddecisions.nll}{0.707}
|
||||
\setres{cap/orig/laya/typeddecisions.unanswered}{0}
|
||||
\setres{cap/orig/laya/typeddecisions.zeroprob}{0.000}
|
||||
\setres{cap/sensors/jev/cmapss.accuracy}{0.270}
|
||||
\setres{cap/sensors/jev/cmapss.brier}{0.776}
|
||||
\setres{cap/sensors/jev/cmapss.ece}{0.077}
|
||||
\setres{cap/sensors/jev/cmapss.errors}{0}
|
||||
\setres{cap/sensors/jev/cmapss.macrof1}{0.224}
|
||||
\setres{cap/sensors/jev/cmapss.p50ms}{259}
|
||||
\setres{cap/sensors/jev/cmapss.unanswered}{0}
|
||||
\setres{cap/sensors/jev/cmapss.usd}{0.008}
|
||||
\setres{cap/sensors/jev/cmapss.zeroprob}{0.000}
|
||||
\setres{cap/sensors/jev/har.accuracy}{0.150}
|
||||
\setres{cap/sensors/jev/har.brier}{1.099}
|
||||
\setres{cap/sensors/jev/har.ece}{0.448}
|
||||
\setres{cap/sensors/jev/har.errors}{0}
|
||||
\setres{cap/sensors/jev/har.macrof1}{0.046}
|
||||
\setres{cap/sensors/jev/har.p50ms}{260}
|
||||
\setres{cap/sensors/jev/har.unanswered}{0}
|
||||
\setres{cap/sensors/jev/har.usd}{0.025}
|
||||
\setres{cap/sensors/jev/har.zeroprob}{0.000}
|
||||
\setres{cap/sensors/jev/mimii.accuracy}{0.840}
|
||||
\setres{cap/sensors/jev/mimii.auc}{0.872}
|
||||
\setres{cap/sensors/jev/mimii.brier}{0.345}
|
||||
\setres{cap/sensors/jev/mimii.ece}{0.194}
|
||||
\setres{cap/sensors/jev/mimii.errors}{0}
|
||||
\setres{cap/sensors/jev/mimii.macrof1}{0.839}
|
||||
\setres{cap/sensors/jev/mimii.p50ms}{257}
|
||||
\setres{cap/sensors/jev/mimii.unanswered}{0}
|
||||
\setres{cap/sensors/jev/mimii.usd}{0.006}
|
||||
\setres{cap/sensors/jev/mimii.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/cmapss.accuracy}{0.170}
|
||||
\setres{cap/sensors/laya/cmapss.brier}{0.785}
|
||||
\setres{cap/sensors/laya/cmapss.ece}{0.130}
|
||||
\setres{cap/sensors/laya/cmapss.macrof1}{0.114}
|
||||
\setres{cap/sensors/laya/cmapss.msperq}{74.0}
|
||||
\setres{cap/sensors/laya/cmapss.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/cmapss.statecut}{1.000}
|
||||
\setres{cap/sensors/laya/cmapss.unanswered}{0}
|
||||
\setres{cap/sensors/laya/cmapss.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/har.accuracy}{0.203}
|
||||
\setres{cap/sensors/laya/har.brier}{0.880}
|
||||
\setres{cap/sensors/laya/har.ece}{0.143}
|
||||
\setres{cap/sensors/laya/har.macrof1}{0.056}
|
||||
\setres{cap/sensors/laya/har.msperq}{76.4}
|
||||
\setres{cap/sensors/laya/har.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/har.statecut}{1.000}
|
||||
\setres{cap/sensors/laya/har.unanswered}{0}
|
||||
\setres{cap/sensors/laya/har.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/long.cmapss.accuracy}{0.140}
|
||||
\setres{cap/sensors/laya/long.cmapss.brier}{0.960}
|
||||
\setres{cap/sensors/laya/long.cmapss.ece}{0.409}
|
||||
\setres{cap/sensors/laya/long.cmapss.macrof1}{0.061}
|
||||
\setres{cap/sensors/laya/long.cmapss.msperq}{154}
|
||||
\setres{cap/sensors/laya/long.cmapss.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/long.cmapss.statecut}{0.000}
|
||||
\setres{cap/sensors/laya/long.cmapss.unanswered}{0}
|
||||
\setres{cap/sensors/laya/long.cmapss.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/long.har.accuracy}{0.163}
|
||||
\setres{cap/sensors/laya/long.har.brier}{1.057}
|
||||
\setres{cap/sensors/laya/long.har.ece}{0.423}
|
||||
\setres{cap/sensors/laya/long.har.macrof1}{0.052}
|
||||
\setres{cap/sensors/laya/long.har.msperq}{164}
|
||||
\setres{cap/sensors/laya/long.har.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/long.har.statecut}{0.000}
|
||||
\setres{cap/sensors/laya/long.har.unanswered}{0}
|
||||
\setres{cap/sensors/laya/long.har.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/long.mimii.accuracy}{0.500}
|
||||
\setres{cap/sensors/laya/long.mimii.auc}{0.486}
|
||||
\setres{cap/sensors/laya/long.mimii.brier}{0.848}
|
||||
\setres{cap/sensors/laya/long.mimii.ece}{0.417}
|
||||
\setres{cap/sensors/laya/long.mimii.macrof1}{0.333}
|
||||
\setres{cap/sensors/laya/long.mimii.msperq}{78.4}
|
||||
\setres{cap/sensors/laya/long.mimii.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/long.mimii.statecut}{0.000}
|
||||
\setres{cap/sensors/laya/long.mimii.unanswered}{0}
|
||||
\setres{cap/sensors/laya/long.mimii.zeroprob}{0.000}
|
||||
\setres{cap/sensors/laya/mimii.accuracy}{0.500}
|
||||
\setres{cap/sensors/laya/mimii.auc}{0.537}
|
||||
\setres{cap/sensors/laya/mimii.brier}{0.512}
|
||||
\setres{cap/sensors/laya/mimii.ece}{0.115}
|
||||
\setres{cap/sensors/laya/mimii.macrof1}{0.333}
|
||||
\setres{cap/sensors/laya/mimii.msperq}{75.3}
|
||||
\setres{cap/sensors/laya/mimii.rejected}{0.000}
|
||||
\setres{cap/sensors/laya/mimii.statecut}{1.000}
|
||||
\setres{cap/sensors/laya/mimii.unanswered}{0}
|
||||
\setres{cap/sensors/laya/mimii.zeroprob}{0.000}
|
||||
|
||||
Reference in New Issue
Block a user