Ablation M-LOW-M3_sgd_high_momentum
Browse files
M/LOW/M3_sgd_high_momentum/seed0/final_report.json
ADDED
|
@@ -0,0 +1,205 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"group": "M",
|
| 4 |
+
"variant": "M3_sgd_high_momentum",
|
| 5 |
+
"band": "LOW",
|
| 6 |
+
"seed": 0,
|
| 7 |
+
"phase": 1,
|
| 8 |
+
"overrides": {
|
| 9 |
+
"optimizer": "sgd",
|
| 10 |
+
"lr": 0.003,
|
| 11 |
+
"momentum": 0.99
|
| 12 |
+
},
|
| 13 |
+
"description": "M-LOW-M3_sgd_high_momentum"
|
| 14 |
+
},
|
| 15 |
+
"run_config": {
|
| 16 |
+
"matrix_v": 64,
|
| 17 |
+
"D": 16,
|
| 18 |
+
"patch_size": 16,
|
| 19 |
+
"hidden": 64,
|
| 20 |
+
"depth": 1,
|
| 21 |
+
"n_cross_layers": 1,
|
| 22 |
+
"n_heads": 4,
|
| 23 |
+
"max_alpha": 0.2,
|
| 24 |
+
"alpha_init": -2.0,
|
| 25 |
+
"img_size": 64,
|
| 26 |
+
"batch_size": 128,
|
| 27 |
+
"lr": 0.003,
|
| 28 |
+
"epochs": 1,
|
| 29 |
+
"weight_decay": 0.0,
|
| 30 |
+
"use_cv_ema": true,
|
| 31 |
+
"cv_ema_alpha": 0.01,
|
| 32 |
+
"cv_alignment_epochs": 0,
|
| 33 |
+
"cv_measure_every": 50,
|
| 34 |
+
"cv_sigma_scale": 0.3,
|
| 35 |
+
"boost": 0.5,
|
| 36 |
+
"cross_attn_clip": 0.5,
|
| 37 |
+
"allowed_types": [
|
| 38 |
+
0,
|
| 39 |
+
1,
|
| 40 |
+
2,
|
| 41 |
+
3,
|
| 42 |
+
4,
|
| 43 |
+
5,
|
| 44 |
+
6,
|
| 45 |
+
7,
|
| 46 |
+
8,
|
| 47 |
+
9,
|
| 48 |
+
10,
|
| 49 |
+
11,
|
| 50 |
+
12,
|
| 51 |
+
13,
|
| 52 |
+
14,
|
| 53 |
+
15
|
| 54 |
+
],
|
| 55 |
+
"train_size": 1000000,
|
| 56 |
+
"val_size": 10000,
|
| 57 |
+
"num_workers": 2,
|
| 58 |
+
"report_every": 100,
|
| 59 |
+
"major_report_every": 10,
|
| 60 |
+
"save_every": 5,
|
| 61 |
+
"seed": 0,
|
| 62 |
+
"hf_repo": "AbstractPhil/geolip-svae-batteries",
|
| 63 |
+
"upload": false
|
| 64 |
+
},
|
| 65 |
+
"cv_ema_final": 0.2039467897293445,
|
| 66 |
+
"cv_last": 0.2048882260410599,
|
| 67 |
+
"predicted_band": "LOW",
|
| 68 |
+
"expected_band": "LOW",
|
| 69 |
+
"band_match": true,
|
| 70 |
+
"test_mse": 0.9959005117416382,
|
| 71 |
+
"recon_ema": 1.0036648829000523,
|
| 72 |
+
"S0": 2.7821898460388184,
|
| 73 |
+
"SD": 1.1573772430419922,
|
| 74 |
+
"ratio": 2.4038746560175177,
|
| 75 |
+
"erank": 15.523387908935547,
|
| 76 |
+
"observed_sphere_cv": 0.20477168730807327,
|
| 77 |
+
"uniform_sphere_cv_prediction": 0.19902003258265127,
|
| 78 |
+
"band_deviation": 0.005751654725421995,
|
| 79 |
+
"params_count": 250211,
|
| 80 |
+
"wallclock_seconds": 14.191136360168457,
|
| 81 |
+
"batches_completed": 1000,
|
| 82 |
+
"batch_limit": 1000,
|
| 83 |
+
"cv_trajectory": [
|
| 84 |
+
{
|
| 85 |
+
"batch": 0,
|
| 86 |
+
"cv": 0.20395987159654982,
|
| 87 |
+
"cv_ema": 0.20395987159654982,
|
| 88 |
+
"recon": 1.6308250427246094
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"batch": 50,
|
| 92 |
+
"cv": 0.1937983926667847,
|
| 93 |
+
"cv_ema": 0.20385825680725217,
|
| 94 |
+
"recon": 1.6899583339691162
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"batch": 100,
|
| 98 |
+
"cv": 0.21203549714999062,
|
| 99 |
+
"cv_ema": 0.20394002921067955,
|
| 100 |
+
"recon": 1.5099332332611084
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"batch": 150,
|
| 104 |
+
"cv": 0.19226184860796794,
|
| 105 |
+
"cv_ema": 0.20382324740465244,
|
| 106 |
+
"recon": 1.7411473989486694
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"batch": 200,
|
| 110 |
+
"cv": 0.2140056891706862,
|
| 111 |
+
"cv_ema": 0.20392507182231276,
|
| 112 |
+
"recon": 1.3361775875091553
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"batch": 250,
|
| 116 |
+
"cv": 0.20131880115500828,
|
| 117 |
+
"cv_ema": 0.2038990091156397,
|
| 118 |
+
"recon": 1.2594659328460693
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"batch": 300,
|
| 122 |
+
"cv": 0.207123242704634,
|
| 123 |
+
"cv_ema": 0.20393125145152965,
|
| 124 |
+
"recon": 1.082836627960205
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"batch": 350,
|
| 128 |
+
"cv": 0.22263688421487332,
|
| 129 |
+
"cv_ema": 0.2041183077791631,
|
| 130 |
+
"recon": 0.988299548625946
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"batch": 400,
|
| 134 |
+
"cv": 0.19616606667486416,
|
| 135 |
+
"cv_ema": 0.2040387853681201,
|
| 136 |
+
"recon": 0.9831360578536987
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"batch": 450,
|
| 140 |
+
"cv": 0.18965845895022088,
|
| 141 |
+
"cv_ema": 0.20389498210394108,
|
| 142 |
+
"recon": 1.1402337551116943
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"batch": 500,
|
| 146 |
+
"cv": 0.19633968460988524,
|
| 147 |
+
"cv_ema": 0.20381942912900053,
|
| 148 |
+
"recon": 1.0152266025543213
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"batch": 550,
|
| 152 |
+
"cv": 0.1988263929477364,
|
| 153 |
+
"cv_ema": 0.2037694987671879,
|
| 154 |
+
"recon": 1.0515997409820557
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"batch": 600,
|
| 158 |
+
"cv": 0.19222247160716088,
|
| 159 |
+
"cv_ema": 0.20365402849558764,
|
| 160 |
+
"recon": 1.0844422578811646
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"batch": 650,
|
| 164 |
+
"cv": 0.21738385279469116,
|
| 165 |
+
"cv_ema": 0.2037913267385787,
|
| 166 |
+
"recon": 1.0286909341812134
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"batch": 700,
|
| 170 |
+
"cv": 0.2033766001803901,
|
| 171 |
+
"cv_ema": 0.20378717947299682,
|
| 172 |
+
"recon": 0.9552921652793884
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"batch": 750,
|
| 176 |
+
"cv": 0.20174503906703276,
|
| 177 |
+
"cv_ema": 0.20376675806893718,
|
| 178 |
+
"recon": 1.0608594417572021
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"batch": 800,
|
| 182 |
+
"cv": 0.2182513945573627,
|
| 183 |
+
"cv_ema": 0.20391160443382142,
|
| 184 |
+
"recon": 0.9039809107780457
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"batch": 850,
|
| 188 |
+
"cv": 0.20077052796306824,
|
| 189 |
+
"cv_ema": 0.2038801936691139,
|
| 190 |
+
"recon": 0.8119789361953735
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"batch": 900,
|
| 194 |
+
"cv": 0.20958885392276652,
|
| 195 |
+
"cv_ema": 0.2039372802716504,
|
| 196 |
+
"recon": 0.868818461894989
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"batch": 950,
|
| 200 |
+
"cv": 0.2048882260410599,
|
| 201 |
+
"cv_ema": 0.2039467897293445,
|
| 202 |
+
"recon": 0.9619293212890625
|
| 203 |
+
}
|
| 204 |
+
]
|
| 205 |
+
}
|