Ethene

SystemEthene (C2H4)
Atoms—1
Runs502
Cap—3

1 The problem file does not state an atom count.

2 The problem file does not state a run count. The n2p2 rows record 50 runs.

3 The problem file does not state a cap.

The main metric is Mean epochs. Scoring rules.

Charts

Success rate on Ethene, in rank order. The rate is the share of runs that finished. 6 ranked rows did not record a success rate and are not drawn.
Success rate on Ethene, in rank order. The rate is the share of runs that finished. 6 ranked rows did not record a success rate and are not drawn. PDF.
Force calls on each recorded start for Ethene. The box is the quartiles of those counts. A blue point is a success and a vermillion cross is a recorded failure (90). The horizontal position is the recorded count. The vertical offset only keeps marks from covering one another.
Force calls on each recorded start for Ethene. The box is the quartiles of those counts. A blue point is a success and a vermillion cross is a recorded failure (90). The horizontal position is the recorded count. The vertical offset only keeps marks from covering one another. PDF.
Cost distribution for Ethene0117397.1 to 38698.6: 138698.6 to 60000: 117397.160000
Cost distribution of 2 ranked rows. Linear bins, from 17397.1 to 60000.

No performance profile. One start has no finite cost.

Cost and success SVG

Cost and success for Ethene
Cost and success. Select the plot for the full SVG.
Best cost for C2H4C2H410⁴10⁵cost (force calls, log scale)n2p2: 17397.1n2p217397.1
A shorter bar is a lower cost on a log scale shared by the codes of this problem.

Leaderboard

A shorter bar is a lower cost on the log scale for that problem. Select a column heading to sort.
RankCodeBestMedianSuccessPAR-2
1 n2p2 17397.1 38698.6 20.0% 17397.1

The tables below list every ranked row, by method family and overall.

These archived rows rank on the value each one published, which leaves out any failure. The failure rule has no score for them, for the reason given with each.

  • Pyamff (8037eb98) · pyamff-rprop10-7-fcoeff0_1 (the problem states no cap)
  • Pyamff (8037eb98) · pyamff-adam10-7-fcoeff0_1 (the problem states no cap)
  • Pyamff (8037eb98) · pyamff-lbfgs10-7-fcoeff0_1 (the problem states no cap)

Minimizers

Minimizers. One row is one code, one version, and one method. A lower cost is better. PAR-2 counts a finished run as the calls it used and a failed run as twice the cap. An archived row shows the average it published, and its PAR-2 applies that rule to the failures the entry records. Hessians is the mean analytic Hessian count. FD calls is that count times one force call per coordinate.
RankRowCostUnitHessiansFD callsRuns
1n2p2 (2.3.0) · Kalman energy17397.1loss-gradient evaluations0050
2n2p2 (2.3.0) · Kalman force60000loss-gradient evaluations0050

Overall

One row is one code, one version, and one method. A lower cost is better. PAR-2 counts a finished run as the calls it used and a failed run as twice the cap. An archived row shows the average it published, and its PAR-2 applies that rule to the failures the entry records. Hessians is the mean analytic Hessian count. FD calls is that count times one force call per coordinate.
RowFamilyFamily rankUnitPAR-2PublishedMedian callsMean callsHessiansFD callsFailuresRuns
n2p2 (2.3.0) · Kalman energyMinimizers1loss-gradient evaluations17397.1100009397.1200crash 0, divergence 0, cap 40, wrong_stationary 050
n2p2 (2.3.0) · Kalman forceMinimizers2loss-gradient evaluations60000300003000000crash 0, divergence 0, cap 50, wrong_stationary 050

A row that counts another unit is ranked in its own table after the overall table.

Ranked in epochs

These rows count epochs. The tables above count loss-gradient evaluations, so these rows are ranked only among themselves.

The archived mean_epoch is the benchmark website's Mean Epochs: mlrf0_optb.py -s averages the last epoch in each model's pyamff.log. optbench/n2p2_ml.py charges one loss-gradient evaluation before each Kalman epoch. No source equates a pyamff epoch with that count.

One row is one code, one version, and one method. A lower cost is better. PAR-2 counts a finished run as the calls it used and a failed run as twice the cap. An archived row shows the average it published, and its PAR-2 applies that rule to the failures the entry records. Hessians is the mean analytic Hessian count. FD calls is that count times one force call per coordinate.
RowFamilyFamily rankUnitPAR-2PublishedMedian callsMean callsHessiansFD callsFailuresRuns
Pyamff (8037eb98) · pyamff-rprop10-7-fcoeff0Minimizers1epochs184.56000 failed50
Pyamff (8037eb98) · pyamff-lbfgs10-7-fcoeff0Minimizers2epochs2157.14000 failed50
Pyamff (8037eb98) · pyamff-adam10-7-fcoeff0Minimizers3epochs4395.3000 failed50
Pyamff (8037eb98) · pyamff-rprop10-7-fcoeff0_1Minimizers4epochs12926.90018 failed, class not recorded50
Pyamff (8037eb98) · pyamff-adam10-7-fcoeff0_1Minimizers5epochs13876.70010 failed, class not recorded50
Pyamff (8037eb98) · pyamff-lbfgs10-7-fcoeff0_1Minimizers6epochs23314.50019 failed, class not recorded50