
=== Start adding workers ===
=> Add worker SGDMWorker(index=0, momentum=0.9)
=> Add worker SGDMWorker(index=1, momentum=0.9)
=> Add worker SGDMWorker(index=2, momentum=0.9)
=> Add worker SGDMWorker(index=3, momentum=0.9)
=> Add worker SGDMWorker(index=4, momentum=0.9)
=> Add worker SGDMWorker(index=5, momentum=0.9)
=> Add worker SGDMWorker(index=6, momentum=0.9)
=> Add worker SGDMWorker(index=7, momentum=0.9)
=> Add worker SGDMWorker(index=8, momentum=0.9)
=> Add worker BitFlippingWorker
=> Add worker BitFlippingWorker

=== Start adding graph ===
<codes.graph_utils.TorusByzantineGraph object at 0x7f702caf0400>

Train epoch 1
[E 1B0  |    352/60000 (  1%) ] Loss: 2.3038 top1= 11.1111

=== Peeking data label distribution E1B0 ===
Worker 0 has targets: tensor([0, 0, 0, 0, 0], device='cuda:0')
Worker 1 has targets: tensor([1, 1, 1, 1, 1], device='cuda:0')
Worker 2 has targets: tensor([2, 2, 2, 2, 2], device='cuda:0')
Worker 3 has targets: tensor([4, 3, 3, 4, 3], device='cuda:0')
Worker 4 has targets: tensor([5, 4, 4, 5, 4], device='cuda:0')
Worker 5 has targets: tensor([6, 6, 6, 6, 5], device='cuda:0')
Worker 6 has targets: tensor([7, 7, 7, 7, 6], device='cuda:0')
Worker 7 has targets: tensor([8, 8, 8, 8, 7], device='cuda:0')
Worker 8 has targets: tensor([9, 9, 9, 9, 8], device='cuda:0')
Worker 9 has targets: tensor([1, 1, 2, 1, 5], device='cuda:0')
Worker 10 has targets: tensor([8, 9, 1, 0, 3], device='cuda:0')



=== Log global consensus distance @ E1B0 ===
consensus_distance=1.508


[E 1B10 |   3872/60000 (  6%) ] Loss: 1.8787 top1= 47.5694
[E 1B20 |   7392/60000 ( 12%) ] Loss: 0.9967 top1= 71.8750
[E 1B30 |  10912/60000 ( 18%) ] Loss: 0.4269 top1= 89.2361
[E 1B40 |  14432/60000 ( 24%) ] Loss: 0.3140 top1= 94.4444
[E 1B50 |  17952/60000 ( 30%) ] Loss: 0.2668 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1099 top1= 65.5950

Train epoch 2
[E 2B0  |    352/60000 (  1%) ] Loss: 0.2228 top1= 98.6111

=== Log global consensus distance @ E2B0 ===
consensus_distance=0.065


[E 2B10 |   3872/60000 (  6%) ] Loss: 0.3108 top1= 94.4444
[E 2B20 |   7392/60000 ( 12%) ] Loss: 0.2816 top1= 95.4861
[E 2B30 |  10912/60000 ( 18%) ] Loss: 0.2691 top1= 96.5278
[E 2B40 |  14432/60000 ( 24%) ] Loss: 0.2434 top1= 96.1806
[E 2B50 |  17952/60000 ( 30%) ] Loss: 0.2472 top1= 94.4444

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1483 top1= 65.5549

Train epoch 3
[E 3B0  |    352/60000 (  1%) ] Loss: 0.2020 top1= 98.2639

=== Log global consensus distance @ E3B0 ===
consensus_distance=0.052


[E 3B10 |   3872/60000 (  6%) ] Loss: 0.2845 top1= 94.7917
[E 3B20 |   7392/60000 ( 12%) ] Loss: 0.2605 top1= 96.1806
[E 3B30 |  10912/60000 ( 18%) ] Loss: 0.2501 top1= 97.5694
[E 3B40 |  14432/60000 ( 24%) ] Loss: 0.2291 top1= 96.5278
[E 3B50 |  17952/60000 ( 30%) ] Loss: 0.2451 top1= 94.7917

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1672 top1= 65.3846

Train epoch 4
[E 4B0  |    352/60000 (  1%) ] Loss: 0.1970 top1= 98.2639

=== Log global consensus distance @ E4B0 ===
consensus_distance=0.051


[E 4B10 |   3872/60000 (  6%) ] Loss: 0.2764 top1= 94.4444
[E 4B20 |   7392/60000 ( 12%) ] Loss: 0.2483 top1= 96.8750
[E 4B30 |  10912/60000 ( 18%) ] Loss: 0.2398 top1= 97.9167
[E 4B40 |  14432/60000 ( 24%) ] Loss: 0.2226 top1= 96.8750
[E 4B50 |  17952/60000 ( 30%) ] Loss: 0.2456 top1= 94.7917

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1790 top1= 65.3045

Train epoch 5
[E 5B0  |    352/60000 (  1%) ] Loss: 0.1961 top1= 98.2639

=== Log global consensus distance @ E5B0 ===
consensus_distance=0.050


[E 5B10 |   3872/60000 (  6%) ] Loss: 0.2735 top1= 94.4444
[E 5B20 |   7392/60000 ( 12%) ] Loss: 0.2428 top1= 97.5694
[E 5B30 |  10912/60000 ( 18%) ] Loss: 0.2349 top1= 97.9167
[E 5B40 |  14432/60000 ( 24%) ] Loss: 0.2185 top1= 96.8750
[E 5B50 |  17952/60000 ( 30%) ] Loss: 0.2455 top1= 94.4444

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1846 top1= 65.3245

Train epoch 6
[E 6B0  |    352/60000 (  1%) ] Loss: 0.1954 top1= 98.2639

=== Log global consensus distance @ E6B0 ===
consensus_distance=0.050


[E 6B10 |   3872/60000 (  6%) ] Loss: 0.2744 top1= 94.4444
[E 6B20 |   7392/60000 ( 12%) ] Loss: 0.2399 top1= 98.2639
[E 6B30 |  10912/60000 ( 18%) ] Loss: 0.2320 top1= 97.5694
[E 6B40 |  14432/60000 ( 24%) ] Loss: 0.2158 top1= 97.2222
[E 6B50 |  17952/60000 ( 30%) ] Loss: 0.2460 top1= 94.4444

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1877 top1= 65.6851

Train epoch 7
[E 7B0  |    352/60000 (  1%) ] Loss: 0.1958 top1= 97.9167

=== Log global consensus distance @ E7B0 ===
consensus_distance=0.050


[E 7B10 |   3872/60000 (  6%) ] Loss: 0.2730 top1= 94.4444
[E 7B20 |   7392/60000 ( 12%) ] Loss: 0.2369 top1= 98.2639
[E 7B30 |  10912/60000 ( 18%) ] Loss: 0.2307 top1= 97.2222
[E 7B40 |  14432/60000 ( 24%) ] Loss: 0.2145 top1= 97.5694
[E 7B50 |  17952/60000 ( 30%) ] Loss: 0.2438 top1= 95.1389

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1894 top1= 65.8754

Train epoch 8
[E 8B0  |    352/60000 (  1%) ] Loss: 0.1963 top1= 97.9167

=== Log global consensus distance @ E8B0 ===
consensus_distance=0.049


[E 8B10 |   3872/60000 (  6%) ] Loss: 0.2733 top1= 94.7917
[E 8B20 |   7392/60000 ( 12%) ] Loss: 0.2359 top1= 98.2639
[E 8B30 |  10912/60000 ( 18%) ] Loss: 0.2295 top1= 97.5694
[E 8B40 |  14432/60000 ( 24%) ] Loss: 0.2135 top1= 97.5694
[E 8B50 |  17952/60000 ( 30%) ] Loss: 0.2428 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1916 top1= 65.8053

Train epoch 9
[E 9B0  |    352/60000 (  1%) ] Loss: 0.1966 top1= 97.9167

=== Log global consensus distance @ E9B0 ===
consensus_distance=0.049


[E 9B10 |   3872/60000 (  6%) ] Loss: 0.2728 top1= 94.7917
[E 9B20 |   7392/60000 ( 12%) ] Loss: 0.2347 top1= 98.2639
[E 9B30 |  10912/60000 ( 18%) ] Loss: 0.2283 top1= 97.2222
[E 9B40 |  14432/60000 ( 24%) ] Loss: 0.2131 top1= 97.5694
[E 9B50 |  17952/60000 ( 30%) ] Loss: 0.2409 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1912 top1= 65.9856

Train epoch 10
[E10B0  |    352/60000 (  1%) ] Loss: 0.1966 top1= 97.9167

=== Log global consensus distance @ E10B0 ===
consensus_distance=0.049


[E10B10 |   3872/60000 (  6%) ] Loss: 0.2724 top1= 94.7917
[E10B20 |   7392/60000 ( 12%) ] Loss: 0.2344 top1= 98.2639
[E10B30 |  10912/60000 ( 18%) ] Loss: 0.2265 top1= 97.5694
[E10B40 |  14432/60000 ( 24%) ] Loss: 0.2128 top1= 97.5694
[E10B50 |  17952/60000 ( 30%) ] Loss: 0.2408 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1916 top1= 66.2059

Train epoch 11
[E11B0  |    352/60000 (  1%) ] Loss: 0.1967 top1= 97.9167

=== Log global consensus distance @ E11B0 ===
consensus_distance=0.049


[E11B10 |   3872/60000 (  6%) ] Loss: 0.2723 top1= 94.7917
[E11B20 |   7392/60000 ( 12%) ] Loss: 0.2342 top1= 98.2639
[E11B30 |  10912/60000 ( 18%) ] Loss: 0.2262 top1= 97.9167
[E11B40 |  14432/60000 ( 24%) ] Loss: 0.2130 top1= 97.5694
[E11B50 |  17952/60000 ( 30%) ] Loss: 0.2404 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1925 top1= 66.2460

Train epoch 12
[E12B0  |    352/60000 (  1%) ] Loss: 0.1967 top1= 97.9167

=== Log global consensus distance @ E12B0 ===
consensus_distance=0.049


[E12B10 |   3872/60000 (  6%) ] Loss: 0.2718 top1= 94.7917
[E12B20 |   7392/60000 ( 12%) ] Loss: 0.2342 top1= 97.9167
[E12B30 |  10912/60000 ( 18%) ] Loss: 0.2265 top1= 97.5694
[E12B40 |  14432/60000 ( 24%) ] Loss: 0.2133 top1= 97.5694
[E12B50 |  17952/60000 ( 30%) ] Loss: 0.2394 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1936 top1= 66.1859

Train epoch 13
[E13B0  |    352/60000 (  1%) ] Loss: 0.1965 top1= 97.9167

=== Log global consensus distance @ E13B0 ===
consensus_distance=0.049


[E13B10 |   3872/60000 (  6%) ] Loss: 0.2725 top1= 94.7917
[E13B20 |   7392/60000 ( 12%) ] Loss: 0.2340 top1= 97.9167
[E13B30 |  10912/60000 ( 18%) ] Loss: 0.2256 top1= 97.9167
[E13B40 |  14432/60000 ( 24%) ] Loss: 0.2129 top1= 97.5694
[E13B50 |  17952/60000 ( 30%) ] Loss: 0.2375 top1= 95.4861

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1938 top1= 66.1659

Train epoch 14
[E14B0  |    352/60000 (  1%) ] Loss: 0.1968 top1= 97.9167

=== Log global consensus distance @ E14B0 ===
consensus_distance=0.049


[E14B10 |   3872/60000 (  6%) ] Loss: 0.2721 top1= 94.7917
[E14B20 |   7392/60000 ( 12%) ] Loss: 0.2339 top1= 97.9167
[E14B30 |  10912/60000 ( 18%) ] Loss: 0.2249 top1= 97.9167
[E14B40 |  14432/60000 ( 24%) ] Loss: 0.2127 top1= 97.5694
[E14B50 |  17952/60000 ( 30%) ] Loss: 0.2367 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1943 top1= 66.1959

Train epoch 15
[E15B0  |    352/60000 (  1%) ] Loss: 0.1964 top1= 97.9167

=== Log global consensus distance @ E15B0 ===
consensus_distance=0.049


[E15B10 |   3872/60000 (  6%) ] Loss: 0.2715 top1= 94.7917
[E15B20 |   7392/60000 ( 12%) ] Loss: 0.2342 top1= 97.9167
[E15B30 |  10912/60000 ( 18%) ] Loss: 0.2243 top1= 97.9167
[E15B40 |  14432/60000 ( 24%) ] Loss: 0.2125 top1= 97.5694
[E15B50 |  17952/60000 ( 30%) ] Loss: 0.2355 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1951 top1= 66.1859

Train epoch 16
[E16B0  |    352/60000 (  1%) ] Loss: 0.1965 top1= 97.9167

=== Log global consensus distance @ E16B0 ===
consensus_distance=0.049


[E16B10 |   3872/60000 (  6%) ] Loss: 0.2715 top1= 94.0972
[E16B20 |   7392/60000 ( 12%) ] Loss: 0.2340 top1= 97.9167
[E16B30 |  10912/60000 ( 18%) ] Loss: 0.2242 top1= 97.9167
[E16B40 |  14432/60000 ( 24%) ] Loss: 0.2123 top1= 97.5694
[E16B50 |  17952/60000 ( 30%) ] Loss: 0.2347 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1964 top1= 66.0557

Train epoch 17
[E17B0  |    352/60000 (  1%) ] Loss: 0.1962 top1= 97.9167

=== Log global consensus distance @ E17B0 ===
consensus_distance=0.049


[E17B10 |   3872/60000 (  6%) ] Loss: 0.2711 top1= 94.4444
[E17B20 |   7392/60000 ( 12%) ] Loss: 0.2333 top1= 97.9167
[E17B30 |  10912/60000 ( 18%) ] Loss: 0.2233 top1= 97.9167
[E17B40 |  14432/60000 ( 24%) ] Loss: 0.2119 top1= 97.5694
[E17B50 |  17952/60000 ( 30%) ] Loss: 0.2332 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1961 top1= 66.1358

Train epoch 18
[E18B0  |    352/60000 (  1%) ] Loss: 0.1955 top1= 97.9167

=== Log global consensus distance @ E18B0 ===
consensus_distance=0.049


[E18B10 |   3872/60000 (  6%) ] Loss: 0.2714 top1= 94.7917
[E18B20 |   7392/60000 ( 12%) ] Loss: 0.2335 top1= 97.9167
[E18B30 |  10912/60000 ( 18%) ] Loss: 0.2230 top1= 97.9167
[E18B40 |  14432/60000 ( 24%) ] Loss: 0.2118 top1= 97.5694
[E18B50 |  17952/60000 ( 30%) ] Loss: 0.2319 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1970 top1= 66.0757

Train epoch 19
[E19B0  |    352/60000 (  1%) ] Loss: 0.1952 top1= 97.9167

=== Log global consensus distance @ E19B0 ===
consensus_distance=0.049


[E19B10 |   3872/60000 (  6%) ] Loss: 0.2717 top1= 94.0972
[E19B20 |   7392/60000 ( 12%) ] Loss: 0.2326 top1= 97.9167
[E19B30 |  10912/60000 ( 18%) ] Loss: 0.2225 top1= 97.9167
[E19B40 |  14432/60000 ( 24%) ] Loss: 0.2116 top1= 97.5694
[E19B50 |  17952/60000 ( 30%) ] Loss: 0.2314 top1= 95.8333

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1974 top1= 66.1358

Train epoch 20
[E20B0  |    352/60000 (  1%) ] Loss: 0.1953 top1= 97.9167

=== Log global consensus distance @ E20B0 ===
consensus_distance=0.049


[E20B10 |   3872/60000 (  6%) ] Loss: 0.2717 top1= 94.4444
[E20B20 |   7392/60000 ( 12%) ] Loss: 0.2324 top1= 97.9167
[E20B30 |  10912/60000 ( 18%) ] Loss: 0.2222 top1= 97.9167
[E20B40 |  14432/60000 ( 24%) ] Loss: 0.2117 top1= 97.5694
[E20B50 |  17952/60000 ( 30%) ] Loss: 0.2290 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1986 top1= 65.9956

Train epoch 21
[E21B0  |    352/60000 (  1%) ] Loss: 0.1947 top1= 97.9167

=== Log global consensus distance @ E21B0 ===
consensus_distance=0.049


[E21B10 |   3872/60000 (  6%) ] Loss: 0.2716 top1= 94.7917
[E21B20 |   7392/60000 ( 12%) ] Loss: 0.2318 top1= 98.2639
[E21B30 |  10912/60000 ( 18%) ] Loss: 0.2219 top1= 97.9167
[E21B40 |  14432/60000 ( 24%) ] Loss: 0.2115 top1= 97.5694
[E21B50 |  17952/60000 ( 30%) ] Loss: 0.2280 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1998 top1= 65.8754

Train epoch 22
[E22B0  |    352/60000 (  1%) ] Loss: 0.1949 top1= 97.9167

=== Log global consensus distance @ E22B0 ===
consensus_distance=0.049


[E22B10 |   3872/60000 (  6%) ] Loss: 0.2726 top1= 94.4444
[E22B20 |   7392/60000 ( 12%) ] Loss: 0.2321 top1= 97.9167
[E22B30 |  10912/60000 ( 18%) ] Loss: 0.2216 top1= 97.9167
[E22B40 |  14432/60000 ( 24%) ] Loss: 0.2115 top1= 97.5694
[E22B50 |  17952/60000 ( 30%) ] Loss: 0.2268 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2009 top1= 65.8153

Train epoch 23
[E23B0  |    352/60000 (  1%) ] Loss: 0.1946 top1= 97.9167

=== Log global consensus distance @ E23B0 ===
consensus_distance=0.049


[E23B10 |   3872/60000 (  6%) ] Loss: 0.2728 top1= 94.4444
[E23B20 |   7392/60000 ( 12%) ] Loss: 0.2320 top1= 97.9167
[E23B30 |  10912/60000 ( 18%) ] Loss: 0.2212 top1= 97.9167
[E23B40 |  14432/60000 ( 24%) ] Loss: 0.2111 top1= 97.2222
[E23B50 |  17952/60000 ( 30%) ] Loss: 0.2260 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2020 top1= 65.7051

Train epoch 24
[E24B0  |    352/60000 (  1%) ] Loss: 0.1945 top1= 97.9167

=== Log global consensus distance @ E24B0 ===
consensus_distance=0.049


[E24B10 |   3872/60000 (  6%) ] Loss: 0.2736 top1= 94.4444
[E24B20 |   7392/60000 ( 12%) ] Loss: 0.2326 top1= 98.2639
[E24B30 |  10912/60000 ( 18%) ] Loss: 0.2208 top1= 97.9167
[E24B40 |  14432/60000 ( 24%) ] Loss: 0.2112 top1= 97.2222
[E24B50 |  17952/60000 ( 30%) ] Loss: 0.2254 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2026 top1= 65.6951

Train epoch 25
[E25B0  |    352/60000 (  1%) ] Loss: 0.1945 top1= 97.9167

=== Log global consensus distance @ E25B0 ===
consensus_distance=0.049


[E25B10 |   3872/60000 (  6%) ] Loss: 0.2733 top1= 94.4444
[E25B20 |   7392/60000 ( 12%) ] Loss: 0.2331 top1= 98.2639
[E25B30 |  10912/60000 ( 18%) ] Loss: 0.2203 top1= 97.9167
[E25B40 |  14432/60000 ( 24%) ] Loss: 0.2107 top1= 97.2222
[E25B50 |  17952/60000 ( 30%) ] Loss: 0.2247 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2024 top1= 65.7352

Train epoch 26
[E26B0  |    352/60000 (  1%) ] Loss: 0.1945 top1= 97.9167

=== Log global consensus distance @ E26B0 ===
consensus_distance=0.048


[E26B10 |   3872/60000 (  6%) ] Loss: 0.2738 top1= 94.4444
[E26B20 |   7392/60000 ( 12%) ] Loss: 0.2332 top1= 98.2639
[E26B30 |  10912/60000 ( 18%) ] Loss: 0.2202 top1= 97.9167
[E26B40 |  14432/60000 ( 24%) ] Loss: 0.2108 top1= 97.2222
[E26B50 |  17952/60000 ( 30%) ] Loss: 0.2236 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2037 top1= 65.6250

Train epoch 27
[E27B0  |    352/60000 (  1%) ] Loss: 0.1945 top1= 97.9167

=== Log global consensus distance @ E27B0 ===
consensus_distance=0.048


[E27B10 |   3872/60000 (  6%) ] Loss: 0.2737 top1= 94.4444
[E27B20 |   7392/60000 ( 12%) ] Loss: 0.2337 top1= 98.2639
[E27B30 |  10912/60000 ( 18%) ] Loss: 0.2195 top1= 97.9167
[E27B40 |  14432/60000 ( 24%) ] Loss: 0.2106 top1= 97.2222
[E27B50 |  17952/60000 ( 30%) ] Loss: 0.2224 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2040 top1= 65.5148

Train epoch 28
[E28B0  |    352/60000 (  1%) ] Loss: 0.1942 top1= 97.9167

=== Log global consensus distance @ E28B0 ===
consensus_distance=0.048


[E28B10 |   3872/60000 (  6%) ] Loss: 0.2735 top1= 94.4444
[E28B20 |   7392/60000 ( 12%) ] Loss: 0.2341 top1= 98.2639
[E28B30 |  10912/60000 ( 18%) ] Loss: 0.2194 top1= 97.9167
[E28B40 |  14432/60000 ( 24%) ] Loss: 0.2105 top1= 97.2222
[E28B50 |  17952/60000 ( 30%) ] Loss: 0.2223 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2045 top1= 65.4046

Train epoch 29
[E29B0  |    352/60000 (  1%) ] Loss: 0.1943 top1= 97.9167

=== Log global consensus distance @ E29B0 ===
consensus_distance=0.048


[E29B10 |   3872/60000 (  6%) ] Loss: 0.2731 top1= 94.4444
[E29B20 |   7392/60000 ( 12%) ] Loss: 0.2340 top1= 98.2639
[E29B30 |  10912/60000 ( 18%) ] Loss: 0.2193 top1= 97.9167
[E29B40 |  14432/60000 ( 24%) ] Loss: 0.2104 top1= 97.2222
[E29B50 |  17952/60000 ( 30%) ] Loss: 0.2216 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2045 top1= 65.4647

Train epoch 30
[E30B0  |    352/60000 (  1%) ] Loss: 0.1942 top1= 97.9167

=== Log global consensus distance @ E30B0 ===
consensus_distance=0.048


[E30B10 |   3872/60000 (  6%) ] Loss: 0.2729 top1= 94.7917
[E30B20 |   7392/60000 ( 12%) ] Loss: 0.2334 top1= 98.2639
[E30B30 |  10912/60000 ( 18%) ] Loss: 0.2195 top1= 97.9167
[E30B40 |  14432/60000 ( 24%) ] Loss: 0.2108 top1= 97.2222
[E30B50 |  17952/60000 ( 30%) ] Loss: 0.2203 top1= 96.1806

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2044 top1= 65.3746

