LDLT / UCI_N4 /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":-0.2415859185,
"median_diff_a_minus_b":-0.2228611507,
"W_stat":11.0,
"p_two_sided":1.797749072e-21,
"z_equiv":-9.5161147732,
"effect_size_r":0.865101343,
"p_holm_global":1.348311804e-19,
"p_holm_within_metric":2.696623608e-20
},
{
"metric":"mean_cert_acc_108",
"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":-0.2070521511,
"median_diff_a_minus_b":-0.1616377295,
"W_stat":33.0,
"p_two_sided":3.102799787e-21,
"z_equiv":-9.4592094932,
"effect_size_r":0.8599281357,
"p_holm_global":2.296071842e-19,
"p_holm_within_metric":4.343919702e-20
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"ortho",
"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":-0.2224755243,
"median_diff_a_minus_b":-0.1828913777,
"W_stat":54.0,
"p_two_sided":5.208425291e-21,
"z_equiv":-9.4048908169,
"effect_size_r":0.8549900743,
"p_holm_global":3.802150463e-19,
"p_holm_within_metric":6.770952879e-20
},
{
"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":-0.2760772691,
"median_diff_a_minus_b":-0.2530477867,
"W_stat":63.0,
"p_two_sided":6.497205758e-21,
"z_equiv":-9.3816113842,
"effect_size_r":0.8528737622,
"p_holm_global":4.677988146e-19,
"p_holm_within_metric":7.79664691e-20
},
{
"metric":"mean_cert_acc_108",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":117,
"wins_a":15,
"wins_b":102,
"ties":4,
"win_rate_a_over_b":0.1404958678,
"mean_diff_a_minus_b":-0.1459002825,
"median_diff_a_minus_b":-0.0794756578,
"W_stat":265.0,
"p_two_sided":4.503316115e-18,
"z_equiv":-8.6652984778,
"effect_size_r":0.8011071277,
"p_holm_global":2.656956508e-16,
"p_holm_within_metric":4.953647726e-17
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":16,
"wins_b":104,
"ties":1,
"win_rate_a_over_b":0.1363636364,
"mean_diff_a_minus_b":-0.1301769866,
"median_diff_a_minus_b":-0.0911340416,
"W_stat":469.0,
"p_two_sided":1.263853299e-16,
"z_equiv":-8.2769357893,
"effect_size_r":0.7555774065,
"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":121,
"wins_a":15,
"wins_b":106,
"ties":0,
"win_rate_a_over_b":0.1239669421,
"mean_diff_a_minus_b":-0.0536017448,
"median_diff_a_minus_b":-0.0410954431,
"W_stat":634.0,
"p_two_sided":0.0,
"z_equiv":-7.9046607086,
"effect_size_r":0.718605519,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":120,
"wins_a":95,
"wins_b":25,
"ties":1,
"win_rate_a_over_b":0.7892561983,
"mean_diff_a_minus_b":0.069025118,
"median_diff_a_minus_b":0.0391470212,
"W_stat":769.0,
"p_two_sided":0.0,
"z_equiv":7.4912750595,
"effect_size_r":0.6838567224,
"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":28,
"wins_b":93,
"ties":0,
"win_rate_a_over_b":0.2314049587,
"mean_diff_a_minus_b":-0.0956856361,
"median_diff_a_minus_b":-0.0559161572,
"W_stat":854.0,
"p_two_sided":0.0,
"z_equiv":-7.3356079089,
"effect_size_r":0.6668734463,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":39,
"wins_b":82,
"ties":0,
"win_rate_a_over_b":0.3223140496,
"mean_diff_a_minus_b":-0.0765752418,
"median_diff_a_minus_b":-0.0288892186,
"W_stat":1595.0,
"p_two_sided":0.00000006,
"z_equiv":-5.4189346153,
"effect_size_r":0.4926304196,
"p_holm_global":0.0000018586,
"p_holm_within_metric":0.0000003597
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":119,
"wins_a":37,
"wins_b":82,
"ties":2,
"win_rate_a_over_b":0.3140495868,
"mean_diff_a_minus_b":-0.0611518687,
"median_diff_a_minus_b":-0.023088851,
"W_stat":1675.0,
"p_two_sided":0.0000005064,
"z_equiv":-5.0238559266,
"effect_size_r":0.4605361177,
"p_holm_global":0.0000141804,
"p_holm_within_metric":0.0000025322
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":83,
"wins_b":38,
"ties":0,
"win_rate_a_over_b":0.6859504132,
"mean_diff_a_minus_b":0.0345337674,
"median_diff_a_minus_b":0.0187233128,
"W_stat":2059.0,
"p_two_sided":0.000024566,
"z_equiv":4.2187505287,
"effect_size_r":0.3835227753,
"p_holm_global":0.0005158858,
"p_holm_within_metric":0.000098264
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":43,
"wins_b":78,
"ties":0,
"win_rate_a_over_b":0.3553719008,
"mean_diff_a_minus_b":-0.0344913506,
"median_diff_a_minus_b":-0.0212858766,
"W_stat":2138.0,
"p_two_sided":0.000059595,
"z_equiv":-4.0144088415,
"effect_size_r":0.3649462583,
"p_holm_global":0.001044842,
"p_holm_within_metric":0.0001787849
},
{
"metric":"mean_cert_acc_108",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":77,
"wins_b":44,
"ties":0,
"win_rate_a_over_b":0.6363636364,
"mean_diff_a_minus_b":0.0191103942,
"median_diff_a_minus_b":0.0181754008,
"W_stat":2506.0,
"p_two_sided":0.0021946809,
"z_equiv":3.0625387038,
"effect_size_r":0.2784126094,
"p_holm_global":0.0197521283,
"p_holm_within_metric":0.0043893618
},
{
"metric":"mean_cert_acc_108",
"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":0.0154233732,
"median_diff_a_minus_b":0.0037754271,
"W_stat":3069.0,
"p_two_sided":0.1082122144,
"z_equiv":1.6062808573,
"effect_size_r":0.1460255325,
"p_holm_global":0.8656977152,
"p_holm_within_metric":0.1082122144
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":112,
"wins_a":1,
"wins_b":111,
"ties":9,
"win_rate_a_over_b":0.0454545455,
"mean_diff_a_minus_b":-0.1173470639,
"median_diff_a_minus_b":-0.0569680196,
"W_stat":10.0,
"p_two_sided":5.435127869e-20,
"z_equiv":-9.1550129731,
"effect_size_r":0.8650674134,
"p_holm_global":3.587184393e-18,
"p_holm_within_metric":8.152691803e-19
},
{
"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":-0.1497530324,
"median_diff_a_minus_b":-0.0803308832,
"W_stat":62.0,
"p_two_sided":9.896991201e-20,
"z_equiv":-9.0900759336,
"effect_size_r":0.8513636146,
"p_holm_global":6.334074369e-18,
"p_holm_within_metric":1.385578768e-18
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":107,
"wins_a":3,
"wins_b":104,
"ties":14,
"win_rate_a_over_b":0.0826446281,
"mean_diff_a_minus_b":-0.0919241391,
"median_diff_a_minus_b":-0.0476176738,
"W_stat":23.0,
"p_two_sided":5.290354121e-19,
"z_equiv":-8.9060071017,
"effect_size_r":0.8609762037,
"p_holm_global":3.227116014e-17,
"p_holm_within_metric":6.877460357e-18
},
{
"metric":"mean_cert_acc_255",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":110,
"wins_a":5,
"wins_b":105,
"ties":11,
"win_rate_a_over_b":0.0867768595,
"mean_diff_a_minus_b":-0.0971526016,
"median_diff_a_minus_b":-0.036305353,
"W_stat":122.0,
"p_two_sided":2.369681608e-18,
"z_equiv":-8.7381455094,
"effect_size_r":0.8331494843,
"p_holm_global":1.421808965e-16,
"p_holm_within_metric":2.84361793e-17
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":114,
"wins_a":11,
"wins_b":103,
"ties":7,
"win_rate_a_over_b":0.1198347107,
"mean_diff_a_minus_b":-0.0526004308,
"median_diff_a_minus_b":-0.0312724002,
"W_stat":258.0,
"p_two_sided":1.390608438e-17,
"z_equiv":-8.5359064521,
"effect_size_r":0.7994608872,
"p_holm_global":8.065528942e-16,
"p_holm_within_metric":1.529669282e-16
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":114,
"wins_a":12,
"wins_b":102,
"ties":7,
"win_rate_a_over_b":0.1280991736,
"mean_diff_a_minus_b":-0.0844428848,
"median_diff_a_minus_b":-0.0396959223,
"W_stat":342.0,
"p_two_sided":1.05518212e-16,
"z_equiv":-8.2984052458,
"effect_size_r":0.7772168612,
"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":93,
"wins_a":8,
"wins_b":85,
"ties":28,
"win_rate_a_over_b":0.1818181818,
"mean_diff_a_minus_b":-0.0653101477,
"median_diff_a_minus_b":-0.0160309789,
"W_stat":182.0,
"p_two_sided":0.0,
"z_equiv":-7.6746815803,
"effect_size_r":0.7958274071,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_255",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":115,
"wins_a":90,
"wins_b":25,
"ties":6,
"win_rate_a_over_b":0.7685950413,
"mean_diff_a_minus_b":0.0578288934,
"median_diff_a_minus_b":0.0239693746,
"W_stat":977.0,
"p_two_sided":0.0,
"z_equiv":6.5792049823,
"effect_size_r":0.613514028,
"p_holm_global":0.0000000017,
"p_holm_within_metric":0.0000000004
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":112,
"wins_a":23,
"wins_b":89,
"ties":9,
"win_rate_a_over_b":0.2272727273,
"mean_diff_a_minus_b":-0.0520369162,
"median_diff_a_minus_b":-0.0241820474,
"W_stat":927.0,
"p_two_sided":0.0000000001,
"z_equiv":-6.4928449387,
"effect_size_r":0.6135161789,
"p_holm_global":0.000000003,
"p_holm_within_metric":0.0000000006
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":115,
"wins_a":80,
"wins_b":35,
"ties":6,
"win_rate_a_over_b":0.6859504132,
"mean_diff_a_minus_b":0.0254229248,
"median_diff_a_minus_b":0.0174753755,
"W_stat":1603.0,
"p_two_sided":0.0000013504,
"z_equiv":4.8321923338,
"effect_size_r":0.4506042586,
"p_holm_global":0.0000337594,
"p_holm_within_metric":0.0000081023
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":106,
"wins_a":27,
"wins_b":79,
"ties":15,
"win_rate_a_over_b":0.2851239669,
"mean_diff_a_minus_b":-0.0266139914,
"median_diff_a_minus_b":-0.0090909088,
"W_stat":1323.0,
"p_two_sided":0.0000018824,
"z_equiv":-4.76565708,
"effect_size_r":0.4628815347,
"p_holm_global":0.0000451774,
"p_holm_within_metric":0.000009412
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":111,
"wins_a":36,
"wins_b":75,
"ties":10,
"win_rate_a_over_b":0.3388429752,
"mean_diff_a_minus_b":-0.031842454,
"median_diff_a_minus_b":-0.0084677431,
"W_stat":1741.0,
"p_two_sided":0.0000580468,
"z_equiv":-4.0206134822,
"effect_size_r":0.3816197434,
"p_holm_global":0.001044842,
"p_holm_within_metric":0.0002321871
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":117,
"wins_a":77,
"wins_b":40,
"ties":4,
"win_rate_a_over_b":0.652892562,
"mean_diff_a_minus_b":0.0201944623,
"median_diff_a_minus_b":0.0131578948,
"W_stat":1982.0,
"p_two_sided":0.0000645869,
"z_equiv":3.9953934287,
"effect_size_r":0.3693742531,
"p_holm_global":0.001044842,
"p_holm_within_metric":0.0002321871
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":118,
"wins_a":45,
"wins_b":73,
"ties":3,
"win_rate_a_over_b":0.3842975207,
"mean_diff_a_minus_b":-0.0324059685,
"median_diff_a_minus_b":-0.0077698529,
"W_stat":2234.0,
"p_two_sided":0.0006111136,
"z_equiv":-3.426633354,
"effect_size_r":0.3154471691,
"p_holm_global":0.0072473427,
"p_holm_within_metric":0.0012222272
},
{
"metric":"mean_cert_acc_255",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":113,
"wins_a":55,
"wins_b":58,
"ties":8,
"win_rate_a_over_b":0.4876033058,
"mean_diff_a_minus_b":0.0052284626,
"median_diff_a_minus_b":0.0,
"W_stat":3068.0,
"p_two_sided":0.6632312125,
"z_equiv":0.4354563548,
"effect_size_r":0.040964288,
"p_holm_global":1.0,
"p_holm_within_metric":0.6632312125
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"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":-0.2438353373,
"median_diff_a_minus_b":-0.231873706,
"W_stat":84.0,
"p_two_sided":1.086095964e-20,
"z_equiv":-9.3272927078,
"effect_size_r":0.8479357007,
"p_holm_global":7.711281345e-19,
"p_holm_within_metric":1.629143946e-19
},
{
"metric":"mean_cert_acc_36",
"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":-0.2706310854,
"median_diff_a_minus_b":-0.2436474561,
"W_stat":88.0,
"p_two_sided":1.197367966e-20,
"z_equiv":-9.3169462933,
"effect_size_r":0.8469951176,
"p_holm_global":8.381575764e-19,
"p_holm_within_metric":1.676315153e-19
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"sdp",
"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":-0.2215721657,
"median_diff_a_minus_b":-0.1912159696,
"W_stat":145.0,
"p_two_sided":4.751745038e-20,
"z_equiv":-9.1695098861,
"effect_size_r":0.8335918078,
"p_holm_global":3.183669175e-18,
"p_holm_within_metric":6.177268549e-19
},
{
"metric":"mean_cert_acc_36",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":10,
"wins_b":111,
"ties":0,
"win_rate_a_over_b":0.0826446281,
"mean_diff_a_minus_b":-0.2304439908,
"median_diff_a_minus_b":-0.218675293,
"W_stat":196.0,
"p_two_sided":1.60159291e-19,
"z_equiv":-9.0375931007,
"effect_size_r":0.8215993728,
"p_holm_global":1.009003533e-17,
"p_holm_within_metric":1.921911492e-18
},
{
"metric":"mean_cert_acc_36",
"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":-0.1632065656,
"median_diff_a_minus_b":-0.1217573732,
"W_stat":579.0,
"p_two_sided":8.490093064e-16,
"z_equiv":-8.0469239085,
"effect_size_r":0.7315385371,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":28,
"wins_b":93,
"ties":0,
"win_rate_a_over_b":0.2314049587,
"mean_diff_a_minus_b":-0.1074245198,
"median_diff_a_minus_b":-0.0510568023,
"W_stat":884.0,
"p_two_sided":0.0,
"z_equiv":-7.2580097998,
"effect_size_r":0.6598190727,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_36",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":97,
"wins_b":24,
"ties":0,
"win_rate_a_over_b":0.8016528926,
"mean_diff_a_minus_b":0.0490589197,
"median_diff_a_minus_b":0.0198676884,
"W_stat":1005.0,
"p_two_sided":0.0,
"z_equiv":6.94503076,
"effect_size_r":0.6313664327,
"p_holm_global":0.0000000002,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":24,
"wins_b":96,
"ties":1,
"win_rate_a_over_b":0.2024793388,
"mean_diff_a_minus_b":-0.0401870946,
"median_diff_a_minus_b":-0.0274580419,
"W_stat":1002.0,
"p_two_sided":0.0,
"z_equiv":-6.8810785592,
"effect_size_r":0.6281536578,
"p_holm_global":0.0000000002,
"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":39,
"wins_b":82,
"ties":0,
"win_rate_a_over_b":0.3223140496,
"mean_diff_a_minus_b":-0.0806287717,
"median_diff_a_minus_b":-0.0372283832,
"W_stat":1356.0,
"p_two_sided":0.0000000016,
"z_equiv":-6.0371328841,
"effect_size_r":0.5488302622,
"p_holm_global":0.0000000518,
"p_holm_within_metric":0.000000011
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"ortho",
"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":-0.0672374252,
"median_diff_a_minus_b":-0.0189958066,
"W_stat":1855.0,
"p_two_sided":0.0000020705,
"z_equiv":-4.7464176702,
"effect_size_r":0.4314925155,
"p_holm_global":0.0000476217,
"p_holm_within_metric":0.0000124231
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt",
"alg_b":"sdp",
"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":-0.0583656001,
"median_diff_a_minus_b":-0.0074773338,
"W_stat":2119.0,
"p_two_sided":0.0000483311,
"z_equiv":-4.0635543106,
"effect_size_r":0.3694140282,
"p_holm_global":0.0009182903,
"p_holm_within_metric":0.0002416554
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":41,
"wins_b":80,
"ties":0,
"win_rate_a_over_b":0.3388429752,
"mean_diff_a_minus_b":-0.0267957481,
"median_diff_a_minus_b":-0.0168480575,
"W_stat":2142.0,
"p_two_sided":0.0000622639,
"z_equiv":-4.004062427,
"effect_size_r":0.3640056752,
"p_holm_global":0.001044842,
"p_holm_within_metric":0.0002490557
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":120,
"wins_a":81,
"wins_b":39,
"ties":1,
"win_rate_a_over_b":0.673553719,
"mean_diff_a_minus_b":0.0222631716,
"median_diff_a_minus_b":0.0118150264,
"W_stat":2104.0,
"p_two_sided":0.0000646711,
"z_equiv":3.9950848115,
"effect_size_r":0.3646996784,
"p_holm_global":0.001044842,
"p_holm_within_metric":0.0002490557
},
{
"metric":"mean_cert_acc_36",
"alg_a":"ldlt-resnet",
"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":0.0133913465,
"median_diff_a_minus_b":0.0125584751,
"W_stat":2429.0,
"p_two_sided":0.0011074347,
"z_equiv":3.2617071837,
"effect_size_r":0.2965188349,
"p_holm_global":0.0110743473,
"p_holm_within_metric":0.0022148695
},
{
"metric":"mean_cert_acc_36",
"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":0.0088718251,
"median_diff_a_minus_b":-0.0004384369,
"W_stat":3076.0,
"p_two_sided":0.1122468558,
"z_equiv":1.5881746319,
"effect_size_r":0.144379512,
"p_holm_global":0.8656977152,
"p_holm_within_metric":0.1122468558
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":7,
"wins_b":114,
"ties":0,
"win_rate_a_over_b":0.0578512397,
"mean_diff_a_minus_b":-0.2932351713,
"median_diff_a_minus_b":-0.2614153504,
"W_stat":107.0,
"p_two_sided":1.900233353e-20,
"z_equiv":-9.2678008242,
"effect_size_r":0.8425273477,
"p_holm_global":1.311161014e-18,
"p_holm_within_metric":2.85035003e-19
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ldlt-resnet",
"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":-0.2631359926,
"median_diff_a_minus_b":-0.2326939851,
"W_stat":116.0,
"p_two_sided":2.362958291e-20,
"z_equiv":-9.2445213915,
"effect_size_r":0.8404110356,
"p_holm_global":1.606811638e-18,
"p_holm_within_metric":3.308141607e-19
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"sdp",
"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":-0.2314897135,
"median_diff_a_minus_b":-0.1927602664,
"W_stat":162.0,
"p_two_sided":7.138107477e-20,
"z_equiv":-9.1255376243,
"effect_size_r":0.8295943295,
"p_holm_global":4.63976986e-18,
"p_holm_within_metric":9.27953972e-19
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":11,
"wins_b":110,
"ties":0,
"win_rate_a_over_b":0.0909090909,
"mean_diff_a_minus_b":-0.2412404211,
"median_diff_a_minus_b":-0.2063548118,
"W_stat":214.0,
"p_two_sided":2.449147439e-19,
"z_equiv":-8.9910342353,
"effect_size_r":0.8173667487,
"p_holm_global":1.518471412e-17,
"p_holm_within_metric":2.938976926e-18
},
{
"metric":"mean_cert_acc_72",
"alg_a":"aol",
"alg_b":"ldlt",
"n_common":121,
"n_nonzero":119,
"wins_a":19,
"wins_b":100,
"ties":2,
"win_rate_a_over_b":0.1652892562,
"mean_diff_a_minus_b":-0.1632772665,
"median_diff_a_minus_b":-0.125,
"W_stat":557.0,
"p_two_sided":0.0,
"z_equiv":-7.988580617,
"effect_size_r":0.7323119845,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":20,
"wins_b":101,
"ties":0,
"win_rate_a_over_b":0.1652892562,
"mean_diff_a_minus_b":-0.1299579048,
"median_diff_a_minus_b":-0.0772011476,
"W_stat":607.0,
"p_two_sided":0.0,
"z_equiv":-7.9744990067,
"effect_size_r":0.7249544552,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":19,
"wins_b":101,
"ties":1,
"win_rate_a_over_b":0.1611570248,
"mean_diff_a_minus_b":-0.0519947502,
"median_diff_a_minus_b":-0.034627907,
"W_stat":601.0,
"p_two_sided":0.0,
"z_equiv":-7.9312450682,
"effect_size_r":0.7240203055,
"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":120,
"wins_a":95,
"wins_b":25,
"ties":1,
"win_rate_a_over_b":0.7892561983,
"mean_diff_a_minus_b":0.0617454578,
"median_diff_a_minus_b":0.0392772481,
"W_stat":801.0,
"p_two_sided":0.0,
"z_equiv":7.4074712483,
"effect_size_r":0.6762065161,
"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":-0.0998587261,
"median_diff_a_minus_b":-0.062678976,
"W_stat":930.0,
"p_two_sided":0.0,
"z_equiv":-7.1390260326,
"effect_size_r":0.6490023666,
"p_holm_global":0.0,
"p_holm_within_metric":0.0
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":42,
"wins_b":79,
"ties":0,
"win_rate_a_over_b":0.347107438,
"mean_diff_a_minus_b":-0.0779631546,
"median_diff_a_minus_b":-0.020989418,
"W_stat":1696.0,
"p_two_sided":0.00000025,
"z_equiv":-5.1576876482,
"effect_size_r":0.4688806953,
"p_holm_global":0.0000075005,
"p_holm_within_metric":0.0000015001
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":42,
"wins_b":79,
"ties":0,
"win_rate_a_over_b":0.347107438,
"mean_diff_a_minus_b":-0.068212447,
"median_diff_a_minus_b":-0.0213127583,
"W_stat":1776.0,
"p_two_sided":0.0000007392,
"z_equiv":-4.9507593574,
"effect_size_r":0.4500690325,
"p_holm_global":0.0000199596,
"p_holm_within_metric":0.0000036962
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":83,
"wins_b":38,
"ties":0,
"win_rate_a_over_b":0.6859504132,
"mean_diff_a_minus_b":0.0316462791,
"median_diff_a_minus_b":0.0191756338,
"W_stat":1938.0,
"p_two_sided":0.0000058503,
"z_equiv":4.5317295685,
"effect_size_r":0.4119754153,
"p_holm_global":0.000128706,
"p_holm_within_metric":0.0000234011
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"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":-0.0300991787,
"median_diff_a_minus_b":-0.0196958482,
"W_stat":2062.0,
"p_two_sided":0.0000254253,
"z_equiv":-4.2109907178,
"effect_size_r":0.382817338,
"p_holm_global":0.0005158858,
"p_holm_within_metric":0.000076276
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ldlt-resnet",
"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":0.0218955715,
"median_diff_a_minus_b":0.0177944824,
"W_stat":2306.0,
"p_two_sided":0.0003437791,
"z_equiv":3.5798594309,
"effect_size_r":0.3254417664,
"p_holm_global":0.0044691282,
"p_holm_within_metric":0.0006875582
},
{
"metric":"mean_cert_acc_72",
"alg_a":"ortho",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":66,
"wins_b":55,
"ties":0,
"win_rate_a_over_b":0.5454545455,
"mean_diff_a_minus_b":0.0097507076,
"median_diff_a_minus_b":0.0040983558,
"W_stat":3178.0,
"p_two_sided":0.1853897994,
"z_equiv":1.3243410611,
"effect_size_r":0.1203946419,
"p_holm_global":0.8656977152,
"p_holm_within_metric":0.1853897994
},
{
"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":-0.0928145234,
"median_diff_a_minus_b":-0.0567989122,
"W_stat":412.0,
"p_two_sided":2.273589155e-17,
"z_equiv":-8.4788867156,
"effect_size_r":0.7708078832,
"p_holm_global":0.0,
"p_holm_within_metric":3.410383732e-16
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":18,
"wins_b":103,
"ties":0,
"win_rate_a_over_b":0.1487603306,
"mean_diff_a_minus_b":-0.0920325933,
"median_diff_a_minus_b":-0.0594664378,
"W_stat":518.0,
"p_two_sided":2.311544437e-16,
"z_equiv":-8.2047067302,
"effect_size_r":0.74588243,
"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":22,
"wins_b":99,
"ties":0,
"win_rate_a_over_b":0.1818181818,
"mean_diff_a_minus_b":-0.0683567538,
"median_diff_a_minus_b":-0.0351344055,
"W_stat":806.0,
"p_two_sided":0.0,
"z_equiv":-7.4597648834,
"effect_size_r":0.6781604439,
"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":27,
"wins_b":94,
"ties":0,
"win_rate_a_over_b":0.2231404959,
"mean_diff_a_minus_b":-0.0674542894,
"median_diff_a_minus_b":-0.0264984947,
"W_stat":971.0,
"p_two_sided":0.0,
"z_equiv":-7.0329752836,
"effect_size_r":0.6393613894,
"p_holm_global":0.0000000001,
"p_holm_within_metric":0.0
},
{
"metric":"mean_test_acc",
"alg_a":"aol",
"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":-0.0727332781,
"median_diff_a_minus_b":-0.0313389237,
"W_stat":1013.0,
"p_two_sided":0.0,
"z_equiv":-6.9243379309,
"effect_size_r":0.6294852664,
"p_holm_global":0.0000000002,
"p_holm_within_metric":0.0
},
{
"metric":"mean_test_acc",
"alg_a":"ortho",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":120,
"wins_a":31,
"wins_b":89,
"ties":1,
"win_rate_a_over_b":0.2603305785,
"mean_diff_a_minus_b":-0.0245783039,
"median_diff_a_minus_b":-0.0125617273,
"W_stat":1269.0,
"p_two_sided":0.0000000006,
"z_equiv":-6.1818405096,
"effect_size_r":0.564322249,
"p_holm_global":0.0000000222,
"p_holm_within_metric":0.0000000063
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":94,
"wins_b":27,
"ties":0,
"win_rate_a_over_b":0.7768595041,
"mean_diff_a_minus_b":0.025360234,
"median_diff_a_minus_b":0.013893891,
"W_stat":1309.0,
"p_two_sided":0.0000000007,
"z_equiv":6.158703255,
"effect_size_r":0.5598821141,
"p_holm_global":0.0000000249,
"p_holm_within_metric":0.0000000066
},
{
"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":0.0244577696,
"median_diff_a_minus_b":0.0110695032,
"W_stat":1503.0,
"p_two_sided":0.0000000154,
"z_equiv":5.6569021498,
"effect_size_r":0.5142638318,
"p_holm_global":0.0000004932,
"p_holm_within_metric":0.0000001233
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"ldlt-resnet",
"n_common":121,
"n_nonzero":120,
"wins_a":87,
"wins_b":33,
"ties":1,
"win_rate_a_over_b":0.7231404959,
"mean_diff_a_minus_b":0.0200812453,
"median_diff_a_minus_b":0.0108439309,
"W_stat":1689.0,
"p_two_sided":0.0000003736,
"z_equiv":5.0819154878,
"effect_size_r":0.4639132913,
"p_holm_global":0.0000108358,
"p_holm_within_metric":0.0000026155
},
{
"metric":"mean_test_acc",
"alg_a":"sandwich",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":121,
"wins_a":83,
"wins_b":38,
"ties":0,
"win_rate_a_over_b":0.6859504132,
"mean_diff_a_minus_b":0.0236758395,
"median_diff_a_minus_b":0.0089801928,
"W_stat":1789.0,
"p_two_sided":0.0000008782,
"z_equiv":4.9171335102,
"effect_size_r":0.4470121373,
"p_holm_global":0.0000228334,
"p_holm_within_metric":0.0000052692
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":43,
"wins_b":78,
"ties":0,
"win_rate_a_over_b":0.3553719008,
"mean_diff_a_minus_b":-0.0192993151,
"median_diff_a_minus_b":-0.0067946254,
"W_stat":2364.0,
"p_two_sided":0.0006039452,
"z_equiv":-3.42983642,
"effect_size_r":0.3118033109,
"p_holm_global":0.0072473427,
"p_holm_within_metric":0.0030197261
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"sdp",
"n_common":121,
"n_nonzero":120,
"wins_a":66,
"wins_b":54,
"ties":1,
"win_rate_a_over_b":0.5495867769,
"mean_diff_a_minus_b":0.0043765243,
"median_diff_a_minus_b":0.0019017287,
"W_stat":3070.0,
"p_two_sided":0.1428507422,
"z_equiv":1.4652572612,
"effect_size_r":0.1337590758,
"p_holm_global":0.8656977152,
"p_holm_within_metric":0.5714029689
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt-resnet",
"alg_b":"ortho",
"n_common":121,
"n_nonzero":121,
"wins_a":71,
"wins_b":50,
"ties":0,
"win_rate_a_over_b":0.5867768595,
"mean_diff_a_minus_b":0.0052789887,
"median_diff_a_minus_b":0.002452642,
"W_stat":3138.0,
"p_two_sided":0.1533479338,
"z_equiv":1.4278052065,
"effect_size_r":0.1298004733,
"p_holm_global":0.8656977152,
"p_holm_within_metric":0.5714029689
},
{
"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":-0.0009024644,
"median_diff_a_minus_b":-0.0003659998,
"W_stat":3623.0,
"p_two_sided":0.8624137031,
"z_equiv":-0.1733024435,
"effect_size_r":0.0157547676,
"p_holm_global":1.0,
"p_holm_within_metric":1.0
},
{
"metric":"mean_test_acc",
"alg_a":"ldlt",
"alg_b":"sandwich",
"n_common":121,
"n_nonzero":121,
"wins_a":59,
"wins_b":62,
"ties":0,
"win_rate_a_over_b":0.4876033058,
"mean_diff_a_minus_b":0.0007819301,
"median_diff_a_minus_b":-0.0001641521,
"W_stat":3624.0,
"p_two_sided":0.8644472072,
"z_equiv":0.1707158399,
"effect_size_r":0.0155196218,
"p_holm_global":1.0,
"p_holm_within_metric":1.0
}
]