LDLT / UCI_N6 /wilcoxon_pairwise_all.json
SuperComputer's picture
Add files using upload-large-folder tool
a1efe59 verified
[
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":1,
"wins_b":120,
"ties":0,
"win_rate_a_over_b":0.0082644628,
"mean_diff_a_minus_b":-0.2739230915,
"median_diff_a_minus_b":-0.2317528344,
"W_stat":45.0,
"p_two_sided":4.173048145e-21,
"z_equiv":-9.4281702496,
"effect_size_r":0.8571063863,
"p_holm_global":3.046325146e-19,
"p_holm_within_metric":6.259572217e-20
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":120,
"wins_a":3,
"wins_b":117,
"ties":1,
"win_rate_a_over_b":0.0289256198,
"mean_diff_a_minus_b":-0.2489971982,
"median_diff_a_minus_b":-0.2098822109,
"W_stat":60.0,
"p_two_sided":8.927572293e-21,
"z_equiv":-9.3480532511,
"effect_size_r":0.8533566057,
"p_holm_global":6.249300605e-19,
"p_holm_within_metric":1.249860121e-19
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":5,
"wins_b":116,
"ties":0,
"win_rate_a_over_b":0.041322314,
"mean_diff_a_minus_b":-0.2999183728,
"median_diff_a_minus_b":-0.2735294178,
"W_stat":82.0,
"p_two_sided":1.034358853e-20,
"z_equiv":-9.3324659151,
"effect_size_r":0.8484059923,
"p_holm_global":7.033640204e-19,
"p_holm_within_metric":1.34466651e-19
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":117,
"wins_a":1,
"wins_b":116,
"ties":4,
"win_rate_a_over_b":0.0247933884,
"mean_diff_a_minus_b":-0.2222316045,
"median_diff_a_minus_b":-0.1720663907,
"W_stat":34.0,
"p_two_sided":1.49193415e-20,
"z_equiv":-9.2935734145,
"effect_size_r":0.8591911661,
"p_holm_global":9.995958805e-19,
"p_holm_within_metric":1.79032098e-19
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":14,
"wins_b":107,
"ties":0,
"win_rate_a_over_b":0.1157024793,
"mean_diff_a_minus_b":-0.1491411817,
"median_diff_a_minus_b":-0.0818716213,
"W_stat":450.0,
"p_two_sided":5.266046065e-17,
"z_equiv":-8.3805957774,
"effect_size_r":0.7618723434,
"p_holm_global":0.0,
"p_holm_within_metric":5.792650671e-16
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":96,
"wins_a":6,
"wins_b":90,
"ties":25,
"win_rate_a_over_b":0.152892562,
"mean_diff_a_minus_b":-0.1507771911,
"median_diff_a_minus_b":-0.0671557561,
"W_stat":113.0,
"p_two_sided":5.846618541e-16,
"z_equiv":-8.0924712901,
"effect_size_r":0.8259343925,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":25,
"wins_b":96,
"ties":0,
"win_rate_a_over_b":0.2066115702,
"mean_diff_a_minus_b":-0.1231459003,
"median_diff_a_minus_b":-0.0714981307,
"W_stat":733.0,
"p_two_sided":0.0,
"z_equiv":-7.6485869487,
"effect_size_r":0.6953260862,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":18,
"wins_b":102,
"ties":1,
"win_rate_a_over_b":0.152892562,
"mean_diff_a_minus_b":-0.0509211747,
"median_diff_a_minus_b":-0.0375725031,
"W_stat":945.0,
"p_two_sided":0.0,
"z_equiv":-7.0303540979,
"effect_size_r":0.6417805878,
"p_holm_global":0.0000000001,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":93,
"wins_b":28,
"ties":0,
"win_rate_a_over_b":0.7685950413,
"mean_diff_a_minus_b":0.0776867683,
"median_diff_a_minus_b":0.043483858,
"W_stat":995.0,
"p_two_sided":0.0,
"z_equiv":6.9708967963,
"effect_size_r":0.6337178906,
"p_holm_global":0.0000000001,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":120,
"wins_a":32,
"wins_b":88,
"ties":1,
"win_rate_a_over_b":0.2685950413,
"mean_diff_a_minus_b":-0.098220007,
"median_diff_a_minus_b":-0.035191671,
"W_stat":1203.0,
"p_two_sided":0.0000000002,
"z_equiv":-6.3546858702,
"effect_size_r":0.5801007995,
"p_holm_global":0.0000000077,
"p_holm_within_metric":0.0000000013
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":117,
"wins_a":35,
"wins_b":82,
"ties":4,
"win_rate_a_over_b":0.305785124,
"mean_diff_a_minus_b":-0.0714544134,
"median_diff_a_minus_b":-0.0293723642,
"W_stat":1335.0,
"p_two_sided":0.0000000087,
"z_equiv":-5.7551072125,
"effect_size_r":0.53205985,
"p_holm_global":0.0000002598,
"p_holm_within_metric":0.0000000433
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":89,
"wins_b":32,
"ties":0,
"win_rate_a_over_b":0.7355371901,
"mean_diff_a_minus_b":0.0516914869,
"median_diff_a_minus_b":0.0233454705,
"W_stat":1603.0,
"p_two_sided":0.0000000673,
"z_equiv":5.3982417863,
"effect_size_r":0.4907492533,
"p_holm_global":0.0000018843,
"p_holm_within_metric":0.0000002692
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":81,
"wins_b":40,
"ties":0,
"win_rate_a_over_b":0.6694214876,
"mean_diff_a_minus_b":0.0249258933,
"median_diff_a_minus_b":0.0182105228,
"W_stat":2277.0,
"p_two_sided":0.0002573114,
"z_equiv":3.6548709363,
"effect_size_r":0.3322609942,
"p_holm_global":0.004631605,
"p_holm_within_metric":0.0007719342
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":54,
"wins_b":67,
"ties":0,
"win_rate_a_over_b":0.4462809917,
"mean_diff_a_minus_b":-0.0259952814,
"median_diff_a_minus_b":-0.0096504689,
"W_stat":2867.0,
"p_two_sided":0.0332728975,
"z_equiv":-2.1287747916,
"effect_size_r":0.1935249811,
"p_holm_global":0.3078706868,
"p_holm_within_metric":0.066545795
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":67,
"wins_b":54,
"ties":0,
"win_rate_a_over_b":0.5537190083,
"mean_diff_a_minus_b":0.0267655936,
"median_diff_a_minus_b":0.0061113834,
"W_stat":2930.0,
"p_two_sided":0.0493195497,
"z_equiv":1.9658187626,
"effect_size_r":0.1787107966,
"p_holm_global":0.3945563974,
"p_holm_within_metric":0.066545795
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":110,
"wins_a":0,
"wins_b":110,
"ties":11,
"win_rate_a_over_b":0.0454545455,
"mean_diff_a_minus_b":-0.1337634204,
"median_diff_a_minus_b":-0.0716657937,
"W_stat":0.0,
"p_two_sided":8.869456124e-20,
"z_equiv":-9.1019863804,
"effect_size_r":0.8678403502,
"p_holm_global":5.587757358e-18,
"p_holm_within_metric":1.330418419e-18
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":114,
"wins_a":2,
"wins_b":112,
"ties":7,
"win_rate_a_over_b":0.0454545455,
"mean_diff_a_minus_b":-0.1626829197,
"median_diff_a_minus_b":-0.0885858461,
"W_stat":74.0,
"p_two_sided":1.351396746e-19,
"z_equiv":-9.0561471899,
"effect_size_r":0.8481858966,
"p_holm_global":8.378659827e-18,
"p_holm_within_metric":1.891955445e-18
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":103,
"wins_a":2,
"wins_b":101,
"ties":18,
"win_rate_a_over_b":0.0909090909,
"mean_diff_a_minus_b":-0.1051036161,
"median_diff_a_minus_b":-0.0429337709,
"W_stat":22.0,
"p_two_sided":2.407012276e-18,
"z_equiv":-8.7363791132,
"effect_size_r":0.8608210125,
"p_holm_global":1.468277489e-16,
"p_holm_within_metric":3.129115959e-17
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":105,
"wins_a":2,
"wins_b":103,
"ties":16,
"win_rate_a_over_b":0.0826446281,
"mean_diff_a_minus_b":-0.1135240161,
"median_diff_a_minus_b":-0.0488128485,
"W_stat":71.0,
"p_two_sided":4.454047917e-18,
"z_equiv":-8.66655163,
"effect_size_r":0.8457688368,
"p_holm_global":2.67242875e-16,
"p_holm_within_metric":5.344857501e-17
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":114,
"wins_a":10,
"wins_b":104,
"ties":7,
"win_rate_a_over_b":0.1115702479,
"mean_diff_a_minus_b":-0.0879753088,
"median_diff_a_minus_b":-0.0350877196,
"W_stat":415.0,
"p_two_sided":5.869031305e-16,
"z_equiv":-8.0920053879,
"effect_size_r":0.7578857434,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":114,
"wins_a":17,
"wins_b":97,
"ties":7,
"win_rate_a_over_b":0.1694214876,
"mean_diff_a_minus_b":-0.0491589036,
"median_diff_a_minus_b":-0.0273425561,
"W_stat":465.0,
"p_two_sided":0.0,
"z_equiv":-7.9506356222,
"effect_size_r":0.7446452517,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":87,
"wins_a":3,
"wins_b":84,
"ties":34,
"win_rate_a_over_b":0.1652892562,
"mean_diff_a_minus_b":-0.0747076109,
"median_diff_a_minus_b":-0.0172084123,
"W_stat":48.0,
"p_two_sided":0.0,
"z_equiv":-7.8955227184,
"effect_size_r":0.8464888875,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":110,
"wins_a":20,
"wins_b":90,
"ties":11,
"win_rate_a_over_b":0.2107438017,
"mean_diff_a_minus_b":-0.0590558095,
"median_diff_a_minus_b":-0.0295574367,
"W_stat":685.0,
"p_two_sided":0.0,
"z_equiv":-7.0591093586,
"effect_size_r":0.6730596687,
"p_holm_global":0.0000000001,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_255",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":114,
"wins_a":84,
"wins_b":30,
"ties":7,
"win_rate_a_over_b":0.7231404959,
"mean_diff_a_minus_b":0.0575793037,
"median_diff_a_minus_b":0.0163958966,
"W_stat":1043.0,
"p_two_sided":0.0000000003,
"z_equiv":6.3164011309,
"effect_size_r":0.5915851679,
"p_holm_global":0.0000000096,
"p_holm_within_metric":0.0000000019
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":111,
"wins_a":80,
"wins_b":31,
"ties":10,
"win_rate_a_over_b":0.7024793388,
"mean_diff_a_minus_b":0.0286598044,
"median_diff_a_minus_b":0.0146750305,
"W_stat":1338.0,
"p_two_sided":0.0000001926,
"z_equiv":5.2063487426,
"effect_size_r":0.4941647538,
"p_holm_global":0.0000052,
"p_holm_within_metric":0.0000011556
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":103,
"wins_a":27,
"wins_b":76,
"ties":18,
"win_rate_a_over_b":0.2975206612,
"mean_diff_a_minus_b":-0.0303960051,
"median_diff_a_minus_b":-0.0085921446,
"W_stat":1234.0,
"p_two_sided":0.0000020443,
"z_equiv":-4.7489976464,
"effect_size_r":0.4679326423,
"p_holm_global":0.000049814,
"p_holm_within_metric":0.0000102214
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":112,
"wins_a":79,
"wins_b":33,
"ties":9,
"win_rate_a_over_b":0.6900826446,
"mean_diff_a_minus_b":0.0202394043,
"median_diff_a_minus_b":0.0148802027,
"W_stat":1541.0,
"p_two_sided":0.0000024732,
"z_equiv":4.7103245755,
"effect_size_r":0.4450838365,
"p_holm_global":0.0000568842,
"p_holm_within_metric":0.0000102214
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":104,
"wins_a":30,
"wins_b":74,
"ties":17,
"win_rate_a_over_b":0.3181818182,
"mean_diff_a_minus_b":-0.0388164052,
"median_diff_a_minus_b":-0.0039353251,
"W_stat":1287.0,
"p_two_sided":0.0000029004,
"z_equiv":-4.6777497092,
"effect_size_r":0.4586910971,
"p_holm_global":0.0000638089,
"p_holm_within_metric":0.0000102214
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":115,
"wins_a":45,
"wins_b":70,
"ties":6,
"win_rate_a_over_b":0.3966942149,
"mean_diff_a_minus_b":-0.0289194993,
"median_diff_a_minus_b":-0.0067800414,
"W_stat":2292.0,
"p_two_sided":0.003621671,
"z_equiv":-2.9093621184,
"effect_size_r":0.2712994164,
"p_holm_global":0.050703394,
"p_holm_within_metric":0.007243342
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":107,
"wins_a":51,
"wins_b":56,
"ties":14,
"win_rate_a_over_b":0.479338843,
"mean_diff_a_minus_b":0.0084204,
"median_diff_a_minus_b":0.0,
"W_stat":2810.0,
"p_two_sided":0.8072471563,
"z_equiv":0.2439789068,
"effect_size_r":0.0235863312,
"p_holm_global":1.0,
"p_holm_within_metric":0.8072471563
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":3,
"wins_b":118,
"ties":0,
"win_rate_a_over_b":0.0247933884,
"mean_diff_a_minus_b":-0.3159544249,
"median_diff_a_minus_b":-0.2903496213,
"W_stat":53.0,
"p_two_sided":5.081865377e-21,
"z_equiv":-9.4074774205,
"effect_size_r":0.85522522,
"p_holm_global":3.608124418e-19,
"p_holm_within_metric":7.622798066e-20
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":4,
"wins_b":117,
"ties":0,
"win_rate_a_over_b":0.0330578512,
"mean_diff_a_minus_b":-0.2939407802,
"median_diff_a_minus_b":-0.2717649229,
"W_stat":115.0,
"p_two_sided":2.306486848e-20,
"z_equiv":-9.2471079952,
"effect_size_r":0.8406461814,
"p_holm_global":1.52228132e-18,
"p_holm_within_metric":3.229081588e-19
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":9,
"wins_b":112,
"ties":0,
"win_rate_a_over_b":0.0743801653,
"mean_diff_a_minus_b":-0.3362883789,
"median_diff_a_minus_b":-0.3413665891,
"W_stat":135.0,
"p_two_sided":3.736856648e-20,
"z_equiv":-9.1953759225,
"effect_size_r":0.8359432657,
"p_holm_global":2.428956821e-18,
"p_holm_within_metric":4.857913642e-19
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":8,
"wins_b":113,
"ties":0,
"win_rate_a_over_b":0.0661157025,
"mean_diff_a_minus_b":-0.3145207473,
"median_diff_a_minus_b":-0.298160661,
"W_stat":140.0,
"p_two_sided":4.214205424e-20,
"z_equiv":-9.1824429043,
"effect_size_r":0.8347675368,
"p_holm_global":2.697091471e-18,
"p_holm_within_metric":5.057046509e-19
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":22,
"wins_b":99,
"ties":0,
"win_rate_a_over_b":0.1818181818,
"mean_diff_a_minus_b":-0.1593331515,
"median_diff_a_minus_b":-0.0761357992,
"W_stat":770.0,
"p_two_sided":0.0,
"z_equiv":-7.5528826142,
"effect_size_r":0.6866256922,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":117,
"wins_a":22,
"wins_b":95,
"ties":4,
"win_rate_a_over_b":0.1983471074,
"mean_diff_a_minus_b":-0.1769552274,
"median_diff_a_minus_b":-0.1111815423,
"W_stat":872.0,
"p_two_sided":0.0,
"z_equiv":-7.0143768908,
"effect_size_r":0.648479373,
"p_holm_global":0.0000000001,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":34,
"wins_b":87,
"ties":0,
"win_rate_a_over_b":0.2809917355,
"mean_diff_a_minus_b":-0.1389991975,
"median_diff_a_minus_b":-0.0613140762,
"W_stat":1073.0,
"p_two_sided":0.0,
"z_equiv":-6.7691417128,
"effect_size_r":0.6153765193,
"p_holm_global":0.0000000005,
"p_holm_within_metric":0.0000000001
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":38,
"wins_b":83,
"ties":0,
"win_rate_a_over_b":0.3140495868,
"mean_diff_a_minus_b":-0.1169855528,
"median_diff_a_minus_b":-0.0544650611,
"W_stat":1275.0,
"p_two_sided":0.0000000004,
"z_equiv":-6.2466477785,
"effect_size_r":0.5678770708,
"p_holm_global":0.0000000143,
"p_holm_within_metric":0.0000000034
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":40,
"wins_b":81,
"ties":0,
"win_rate_a_over_b":0.3305785124,
"mean_diff_a_minus_b":-0.1375655199,
"median_diff_a_minus_b":-0.058108862,
"W_stat":1346.0,
"p_two_sided":0.0000000013,
"z_equiv":-6.0629989205,
"effect_size_r":0.55118172,
"p_holm_global":0.0000000428,
"p_holm_within_metric":0.0000000094
},
{
"metric":"mean_cert_acc_36",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":89,
"wins_b":32,
"ties":0,
"win_rate_a_over_b":0.7355371901,
"mean_diff_a_minus_b":0.0423475987,
"median_diff_a_minus_b":0.0236002803,
"W_stat":1459.0,
"p_two_sided":0.0000000079,
"z_equiv":5.7707127097,
"effect_size_r":0.5246102463,
"p_holm_global":0.0000002447,
"p_holm_within_metric":0.0000000474
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":33,
"wins_b":87,
"ties":1,
"win_rate_a_over_b":0.2768595041,
"mean_diff_a_minus_b":-0.0217676316,
"median_diff_a_minus_b":-0.0164444894,
"W_stat":1846.0,
"p_two_sided":0.000003001,
"z_equiv":-4.6707530392,
"effect_size_r":0.4263794667,
"p_holm_global":0.0000638089,
"p_holm_within_metric":0.0000150049
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":81,
"wins_b":40,
"ties":0,
"win_rate_a_over_b":0.6694214876,
"mean_diff_a_minus_b":0.0220136447,
"median_diff_a_minus_b":0.0100490861,
"W_stat":2305.0,
"p_two_sided":0.000340392,
"z_equiv":3.5824460345,
"effect_size_r":0.3256769122,
"p_holm_global":0.0054462712,
"p_holm_within_metric":0.0013615678
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":50,
"wins_b":71,
"ties":0,
"win_rate_a_over_b":0.4132231405,
"mean_diff_a_minus_b":-0.020333954,
"median_diff_a_minus_b":-0.0093774051,
"W_stat":2753.0,
"p_two_sided":0.0153655075,
"z_equiv":-2.423647606,
"effect_size_r":0.2203316005,
"p_holm_global":0.1843860902,
"p_holm_within_metric":0.0460965226
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":65,
"wins_b":56,
"ties":0,
"win_rate_a_over_b":0.5371900826,
"mean_diff_a_minus_b":0.0205799671,
"median_diff_a_minus_b":0.0014110952,
"W_stat":2942.0,
"p_two_sided":0.0530173574,
"z_equiv":1.934779519,
"effect_size_r":0.1758890472,
"p_holm_global":0.3945563974,
"p_holm_within_metric":0.1060347149
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":75,
"wins_b":46,
"ties":0,
"win_rate_a_over_b":0.6198347107,
"mean_diff_a_minus_b":0.0014336775,
"median_diff_a_minus_b":0.0064880252,
"W_stat":3172.0,
"p_two_sided":0.1802906425,
"z_equiv":1.3398606829,
"effect_size_r":0.1218055166,
"p_holm_global":1.0,
"p_holm_within_metric":0.1802906425
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":0,
"wins_b":121,
"ties":0,
"win_rate_a_over_b":0.0,
"mean_diff_a_minus_b":-0.2656215986,
"median_diff_a_minus_b":-0.2418231331,
"W_stat":0.0,
"p_two_sided":1.36676811e-21,
"z_equiv":-9.5445674132,
"effect_size_r":0.8676879467,
"p_holm_global":1.025076083e-19,
"p_holm_within_metric":2.050152165e-20
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":1,
"wins_b":120,
"ties":0,
"win_rate_a_over_b":0.0082644628,
"mean_diff_a_minus_b":-0.3045117887,
"median_diff_a_minus_b":-0.2825206725,
"W_stat":25.0,
"p_two_sided":2.545212099e-21,
"z_equiv":-9.4799023223,
"effect_size_r":0.861809302,
"p_holm_global":1.883456953e-19,
"p_holm_within_metric":3.563296938e-20
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":4,
"wins_b":117,
"ties":0,
"win_rate_a_over_b":0.0330578512,
"mean_diff_a_minus_b":-0.2930986073,
"median_diff_a_minus_b":-0.2681985311,
"W_stat":45.0,
"p_two_sided":4.173048145e-21,
"z_equiv":-9.4281702496,
"effect_size_r":0.8571063863,
"p_holm_global":3.046325146e-19,
"p_holm_within_metric":5.424962588e-20
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":6,
"wins_b":115,
"ties":0,
"win_rate_a_over_b":0.0495867769,
"mean_diff_a_minus_b":-0.332593737,
"median_diff_a_minus_b":-0.3329101392,
"W_stat":80.0,
"p_two_sided":9.850602055e-21,
"z_equiv":-9.3376391224,
"effect_size_r":0.8488762839,
"p_holm_global":6.796915418e-19,
"p_holm_within_metric":1.182072247e-19
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":16,
"wins_b":105,
"ties":0,
"win_rate_a_over_b":0.132231405,
"mean_diff_a_minus_b":-0.1617923106,
"median_diff_a_minus_b":-0.0864403248,
"W_stat":509.0,
"p_two_sided":1.903870686e-16,
"z_equiv":-8.227986163,
"effect_size_r":0.7479987421,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":103,
"wins_a":10,
"wins_b":93,
"ties":18,
"win_rate_a_over_b":0.1570247934,
"mean_diff_a_minus_b":-0.1708014264,
"median_diff_a_minus_b":-0.0849279724,
"W_stat":304.0,
"p_two_sided":0.0,
"z_equiv":-7.8086220392,
"effect_size_r":0.7694063917,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":30,
"wins_b":91,
"ties":0,
"win_rate_a_over_b":0.2479338843,
"mean_diff_a_minus_b":-0.1337103623,
"median_diff_a_minus_b":-0.0764530534,
"W_stat":873.0,
"p_two_sided":0.0,
"z_equiv":-7.2864624398,
"effect_size_r":0.6624056763,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":95,
"wins_b":26,
"ties":0,
"win_rate_a_over_b":0.7851239669,
"mean_diff_a_minus_b":0.0669721384,
"median_diff_a_minus_b":0.039064005,
"W_stat":1169.0,
"p_two_sided":0.0000000001,
"z_equiv":6.5208277639,
"effect_size_r":0.592802524,
"p_holm_global":0.0000000028,
"p_holm_within_metric":0.0000000006
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":22,
"wins_b":98,
"ties":1,
"win_rate_a_over_b":0.1859504132,
"mean_diff_a_minus_b":-0.0394951297,
"median_diff_a_minus_b":-0.0318346173,
"W_stat":1154.0,
"p_two_sided":0.0000000001,
"z_equiv":-6.4830104561,
"effect_size_r":0.5918151779,
"p_holm_global":0.0000000035,
"p_holm_within_metric":0.0000000006
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":34,
"wins_b":87,
"ties":0,
"win_rate_a_over_b":0.2809917355,
"mean_diff_a_minus_b":-0.0948201722,
"median_diff_a_minus_b":-0.04369919,
"W_stat":1219.0,
"p_two_sided":0.0000000002,
"z_equiv":-6.3914975821,
"effect_size_r":0.5810452347,
"p_holm_global":0.0000000062,
"p_holm_within_metric":0.000000001
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":36,
"wins_b":85,
"ties":0,
"win_rate_a_over_b":0.2975206612,
"mean_diff_a_minus_b":-0.1222971809,
"median_diff_a_minus_b":-0.0452102409,
"W_stat":1256.0,
"p_two_sided":0.0000000003,
"z_equiv":-6.2957932476,
"effect_size_r":0.5723448407,
"p_holm_global":0.0000000107,
"p_holm_within_metric":0.0000000015
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":89,
"wins_b":32,
"ties":0,
"win_rate_a_over_b":0.7355371901,
"mean_diff_a_minus_b":0.03889019,
"median_diff_a_minus_b":0.0201615543,
"W_stat":1852.0,
"p_two_sided":0.0000019926,
"z_equiv":4.7541774811,
"effect_size_r":0.4321979528,
"p_holm_global":0.000049814,
"p_holm_within_metric":0.0000079702
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":49,
"wins_b":72,
"ties":0,
"win_rate_a_over_b":0.4049586777,
"mean_diff_a_minus_b":-0.0280819483,
"median_diff_a_minus_b":-0.0133254528,
"W_stat":2656.0,
"p_two_sided":0.0074830028,
"z_equiv":-2.6745481586,
"effect_size_r":0.2431407417,
"p_holm_global":0.0972790362,
"p_holm_within_metric":0.0224490084
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":67,
"wins_b":54,
"ties":0,
"win_rate_a_over_b":0.5537190083,
"mean_diff_a_minus_b":0.0274770087,
"median_diff_a_minus_b":0.0045558959,
"W_stat":2784.0,
"p_two_sided":0.0191056626,
"z_equiv":2.3434628933,
"effect_size_r":0.2130420812,
"p_holm_global":0.2101622889,
"p_holm_within_metric":0.0382113252
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":74,
"wins_b":47,
"ties":0,
"win_rate_a_over_b":0.6115702479,
"mean_diff_a_minus_b":0.0114131814,
"median_diff_a_minus_b":0.0088901073,
"W_stat":2855.0,
"p_two_sided":0.0307870687,
"z_equiv":2.1598140352,
"effect_size_r":0.1963467305,
"p_holm_global":0.3078706868,
"p_holm_within_metric":0.0382113252
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":121,
"wins_a":16,
"wins_b":105,
"ties":0,
"win_rate_a_over_b":0.132231405,
"mean_diff_a_minus_b":-0.1195835257,
"median_diff_a_minus_b":-0.0631501061,
"W_stat":409.0,
"p_two_sided":2.126854418e-17,
"z_equiv":-8.4866465265,
"effect_size_r":0.7715133206,
"p_holm_global":0.0,
"p_holm_within_metric":3.190281628e-16
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":23,
"wins_b":98,
"ties":0,
"win_rate_a_over_b":0.1900826446,
"mean_diff_a_minus_b":-0.1113283867,
"median_diff_a_minus_b":-0.0656148695,
"W_stat":698.0,
"p_two_sided":0.0,
"z_equiv":-7.7391180759,
"effect_size_r":0.7035561887,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":25,
"wins_b":96,
"ties":0,
"win_rate_a_over_b":0.2066115702,
"mean_diff_a_minus_b":-0.0986620727,
"median_diff_a_minus_b":-0.0525094743,
"W_stat":908.0,
"p_two_sided":0.0,
"z_equiv":-7.1959313126,
"effect_size_r":0.6541755739,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":24,
"wins_b":97,
"ties":0,
"win_rate_a_over_b":0.1983471074,
"mean_diff_a_minus_b":-0.0961238247,
"median_diff_a_minus_b":-0.0445100629,
"W_stat":1047.0,
"p_two_sided":0.0,
"z_equiv":-6.8363934073,
"effect_size_r":0.6214903098,
"p_holm_global":0.0000000003,
"p_holm_within_metric":0.0000000001
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":27,
"wins_b":94,
"ties":0,
"win_rate_a_over_b":0.2231404959,
"mean_diff_a_minus_b":-0.0920606953,
"median_diff_a_minus_b":-0.0356908378,
"W_stat":1307.0,
"p_two_sided":0.0000000007,
"z_equiv":-6.1638764622,
"effect_size_r":0.5603524057,
"p_holm_global":0.0000000234,
"p_holm_within_metric":0.0000000078
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":86,
"wins_b":35,
"ties":0,
"win_rate_a_over_b":0.7107438017,
"mean_diff_a_minus_b":0.023459701,
"median_diff_a_minus_b":0.0102784065,
"W_stat":1511.0,
"p_two_sided":0.0000000174,
"z_equiv":5.6362093207,
"effect_size_r":0.5123826655,
"p_holm_global":0.0000005041,
"p_holm_within_metric":0.0000001738
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":121,
"wins_a":87,
"wins_b":34,
"ties":0,
"win_rate_a_over_b":0.7190082645,
"mean_diff_a_minus_b":0.0275228304,
"median_diff_a_minus_b":0.0124811598,
"W_stat":1779.0,
"p_two_sided":0.0000007693,
"z_equiv":4.9429995465,
"effect_size_r":0.4493635951,
"p_holm_global":0.0000200017,
"p_holm_within_metric":0.0000069237
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":78,
"wins_b":43,
"ties":0,
"win_rate_a_over_b":0.6446280992,
"mean_diff_a_minus_b":0.020921453,
"median_diff_a_minus_b":0.010356299,
"W_stat":2002.0,
"p_two_sided":0.0000126434,
"z_equiv":4.3661869359,
"effect_size_r":0.3969260851,
"p_holm_global":0.0002528686,
"p_holm_within_metric":0.0001011474
},
{
"metric":"mean_test_acc",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":39,
"wins_b":81,
"ties":1,
"win_rate_a_over_b":0.326446281,
"mean_diff_a_minus_b":-0.012666314,
"median_diff_a_minus_b":-0.0078946893,
"W_stat":2171.0,
"p_two_sided":0.0001336571,
"z_equiv":-3.8196205818,
"effect_size_r":0.348682059,
"p_holm_global":0.0025394851,
"p_holm_within_metric":0.0009355998
},
{
"metric":"mean_test_acc",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":80,
"wins_b":41,
"ties":0,
"win_rate_a_over_b":0.6611570248,
"mean_diff_a_minus_b":0.015204562,
"median_diff_a_minus_b":0.006305895,
"W_stat":2279.0,
"p_two_sided":0.0002625491,
"z_equiv":3.649697729,
"effect_size_r":0.3317907026,
"p_holm_global":0.004631605,
"p_holm_within_metric":0.0015752945
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":47,
"wins_b":74,
"ties":0,
"win_rate_a_over_b":0.3884297521,
"mean_diff_a_minus_b":-0.0192676914,
"median_diff_a_minus_b":-0.01009705,
"W_stat":2411.0,
"p_two_sided":0.0009387558,
"z_equiv":-3.3082660492,
"effect_size_r":0.300751459,
"p_holm_global":0.0140813373,
"p_holm_within_metric":0.0046937791
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":64,
"wins_b":56,
"ties":1,
"win_rate_a_over_b":0.5330578512,
"mean_diff_a_minus_b":0.008255139,
"median_diff_a_minus_b":0.0008221223,
"W_stat":3145.0,
"p_two_sided":0.2044973939,
"z_equiv":1.2688420788,
"effect_size_r":0.1158289047,
"p_holm_global":1.0,
"p_holm_within_metric":0.8179895757
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":54,
"wins_b":67,
"ties":0,
"win_rate_a_over_b":0.4462809917,
"mean_diff_a_minus_b":-0.0066013774,
"median_diff_a_minus_b":-0.0023532391,
"W_stat":3206.0,
"p_two_sided":0.2106004159,
"z_equiv":-1.2519161593,
"effect_size_r":0.1138105599,
"p_holm_global":1.0,
"p_holm_within_metric":0.8179895757
},
{
"metric":"mean_test_acc",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":60,
"wins_b":61,
"ties":0,
"win_rate_a_over_b":0.4958677686,
"mean_diff_a_minus_b":0.0025382481,
"median_diff_a_minus_b":-0.0000065726,
"W_stat":3382.0,
"p_two_sided":0.4256404345,
"z_equiv":0.7966739196,
"effect_size_r":0.0724249018,
"p_holm_global":1.0,
"p_holm_within_metric":0.8512808691
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":58,
"wins_b":63,
"ties":0,
"win_rate_a_over_b":0.479338843,
"mean_diff_a_minus_b":-0.0040631294,
"median_diff_a_minus_b":-0.0009775593,
"W_stat":3627.0,
"p_two_sided":0.8705530556,
"z_equiv":-0.162956029,
"effect_size_r":0.0148141845,
"p_holm_global":1.0,
"p_holm_within_metric":0.8705530556
}
]