| Epoch 0 | Batch 199/57412 | Loss: 0.445 | time 28.46233606338501 |
| Epoch 0 | Batch 399/57412 | Loss: 0.297 | time 53.30682563781738 |
| Epoch 0 | Batch 599/57412 | Loss: 0.443 | time 77.94350504875183 |
| Epoch 0 | Batch 799/57412 | Loss: 0.426 | time 101.86682081222534 |
| Epoch 0 | Batch 999/57412 | Loss: 0.512 | time 125.74787616729736 |
| Epoch 0 | Batch 1199/57412 | Loss: 0.486 | time 149.30434608459473 |
| Epoch 0 | Batch 1399/57412 | Loss: 0.688 | time 174.73947978019714 |
| Epoch 0 | Batch 1599/57412 | Loss: 0.543 | time 199.95268726348877 |
| Epoch 0 | Batch 1799/57412 | Loss: 0.467 | time 222.2605311870575 |
| Epoch 0 | Batch 1999/57412 | Loss: 0.192 | time 244.90328335762024 |
| Epoch 0 | Batch 2199/57412 | Loss: 0.256 | time 267.0881850719452 |
| Epoch 0 | Batch 2399/57412 | Loss: 0.738 | time 288.6037828922272 |
| Epoch 0 | Batch 2599/57412 | Loss: 0.396 | time 313.1428134441376 |
| Epoch 0 | Batch 2799/57412 | Loss: 0.363 | time 336.0757656097412 |
| Epoch 0 | Batch 2999/57412 | Loss: 0.531 | time 359.67420649528503 |
| Epoch 0 | Batch 3199/57412 | Loss: 0.574 | time 382.70953011512756 |
| Epoch 0 | Batch 3399/57412 | Loss: 0.355 | time 406.1006875038147 |
| Epoch 0 | Batch 3599/57412 | Loss: 0.438 | time 429.04789757728577 |
| Epoch 0 | Batch 3799/57412 | Loss: 0.488 | time 452.87257862091064 |
| Epoch 0 | Batch 3999/57412 | Loss: 0.232 | time 475.88650941848755 |
| Epoch 0 | Batch 4199/57412 | Loss: 0.574 | time 498.9548225402832 |
| Epoch 0 | Batch 4399/57412 | Loss: 0.680 | time 523.3407320976257 |
| Epoch 0 | Batch 4599/57412 | Loss: 0.301 | time 546.7935557365417 |
| Epoch 0 | Batch 4799/57412 | Loss: 0.395 | time 569.7467799186707 |
| Epoch 0 | Batch 4999/57412 | Loss: 0.539 | time 592.0153501033783 |
| Epoch 0 | Batch 5199/57412 | Loss: 0.516 | time 615.1304590702057 |
| Epoch 0 | Batch 5399/57412 | Loss: 0.363 | time 637.8864369392395 |
| Epoch 0 | Batch 199/57412 | Loss: 0.447 | time 26.585806131362915 |
| Epoch 0 | Batch 399/57412 | Loss: 0.297 | time 51.407662868499756 |
| Epoch 0 | Batch 599/57412 | Loss: 0.443 | time 76.04169297218323 |
| Epoch 0 | Batch 799/57412 | Loss: 0.428 | time 100.11484742164612 |
| Epoch 0 | Batch 999/57412 | Loss: 0.508 | time 123.53260397911072 |
| Epoch 0 | Batch 1199/57412 | Loss: 0.486 | time 146.86401271820068 |
| Epoch 0 | Batch 1399/57412 | Loss: 0.688 | time 172.57908701896667 |
| Epoch 0 | Batch 1599/57412 | Loss: 0.547 | time 197.96987533569336 |
| Epoch 0 | Batch 1799/57412 | Loss: 0.469 | time 221.6588876247406 |
| Epoch 0 | Batch 1999/57412 | Loss: 0.190 | time 245.90447878837585 |
| Epoch 0 | Batch 2199/57412 | Loss: 0.258 | time 269.2711977958679 |
| Epoch 0 | Batch 2399/57412 | Loss: 0.738 | time 292.4727909564972 |
| Epoch 0 | Batch 2599/57412 | Loss: 0.396 | time 318.03349351882935 |
| Epoch 0 | Batch 2799/57412 | Loss: 0.365 | time 342.4670832157135 |
| Epoch 0 | Batch 2999/57412 | Loss: 0.531 | time 367.49278497695923 |
| Epoch 0 | Batch 3199/57412 | Loss: 0.570 | time 392.0638704299927 |
| Epoch 0 | Batch 3399/57412 | Loss: 0.355 | time 417.3946614265442 |
| Epoch 0 | Batch 3599/57412 | Loss: 0.438 | time 442.00172567367554 |
| Epoch 0 | Batch 3799/57412 | Loss: 0.488 | time 467.10806035995483 |
| Epoch 0 | Batch 3999/57412 | Loss: 0.233 | time 491.4598455429077 |
| Epoch 0 | Batch 4199/57412 | Loss: 0.574 | time 515.6410851478577 |
| Epoch 0 | Batch 4399/57412 | Loss: 0.680 | time 540.409966468811 |
| Epoch 0 | Batch 4599/57412 | Loss: 0.299 | time 565.2786910533905 |
| Epoch 0 | Batch 4799/57412 | Loss: 0.395 | time 589.3530840873718 |
| Epoch 0 | Batch 4999/57412 | Loss: 0.539 | time 613.0416724681854 |
| Epoch 0 | Batch 5199/57412 | Loss: 0.516 | time 638.119179725647 |
| Epoch 0 | Batch 5399/57412 | Loss: 0.363 | time 662.4403145313263 |
| Epoch 0 | Batch 5599/57412 | Loss: 0.453 | time 686.884923696518 |
| Epoch 0 | Batch 5799/57412 | Loss: 0.404 | time 710.8961181640625 |
| Epoch 0 | Batch 5999/57412 | Loss: 0.586 | time 733.4252643585205 |
| Epoch 0 | Batch 6199/57412 | Loss: 0.467 | time 756.682168006897 |
| Epoch 0 | Batch 6399/57412 | Loss: 0.291 | time 780.567312002182 |
| Epoch 0 | Batch 6599/57412 | Loss: 0.334 | time 803.4959967136383 |
| Epoch 0 | Batch 6799/57412 | Loss: 0.490 | time 827.7218525409698 |
| Epoch 0 | Batch 6999/57412 | Loss: 0.498 | time 850.365451335907 |
| Epoch 0 | Batch 7199/57412 | Loss: 0.354 | time 872.7854058742523 |
| Epoch 0 | Batch 7399/57412 | Loss: 0.410 | time 896.1070301532745 |
| Epoch 0 | Batch 7599/57412 | Loss: 0.629 | time 920.9819028377533 |
| Epoch 0 | Batch 7799/57412 | Loss: 0.727 | time 945.698518037796 |
| Epoch 0 | Batch 7999/57412 | Loss: 0.688 | time 969.5436074733734 |
| Epoch 0 | Batch 8199/57412 | Loss: 0.238 | time 994.3802013397217 |
| Epoch 0 | Batch 8399/57412 | Loss: 0.516 | time 1019.0018842220306 |
| Epoch 0 | Batch 8599/57412 | Loss: 0.309 | time 1043.567131280899 |
| Epoch 0 | Batch 8799/57412 | Loss: 0.625 | time 1069.965624332428 |
| Epoch 0 | Batch 8999/57412 | Loss: 0.527 | time 1094.9515826702118 |
| Epoch 0 | Batch 9199/57412 | Loss: 0.715 | time 1120.2950911521912 |
| Epoch 0 | Batch 9399/57412 | Loss: 0.516 | time 1145.6069440841675 |
| Epoch 0 | Batch 9599/57412 | Loss: 0.398 | time 1170.8065223693848 |
| Epoch 0 | Batch 9799/57412 | Loss: 0.097 | time 1195.0658061504364 |
| Epoch 0 | Batch 9999/57412 | Loss: 0.527 | time 1219.5924994945526 |
| Epoch 0 | Batch 10199/57412 | Loss: 0.256 | time 1244.1359188556671 |
| Epoch 0 | Batch 10399/57412 | Loss: 0.562 | time 1268.7032721042633 |
| Epoch 0 | Batch 10599/57412 | Loss: 0.385 | time 1292.4727103710175 |
| Epoch 0 | Batch 10799/57412 | Loss: 0.371 | time 1317.3908100128174 |
| Epoch 0 | Batch 10999/57412 | Loss: 0.244 | time 1340.6929655075073 |
| Epoch 0 | Batch 11199/57412 | Loss: 0.299 | time 1365.312620639801 |
| Epoch 0 | Batch 11399/57412 | Loss: 0.629 | time 1389.398408651352 |
| Epoch 0 | Batch 11599/57412 | Loss: 0.381 | time 1415.850613117218 |
| Epoch 0 | Batch 11799/57412 | Loss: 0.379 | time 1439.0871560573578 |
| Epoch 0 | Batch 11999/57412 | Loss: 0.494 | time 1463.5666027069092 |
| Epoch 0 | Batch 12199/57412 | Loss: 0.531 | time 1487.9095635414124 |
| Epoch 0 | Batch 12399/57412 | Loss: 0.598 | time 1512.0723524093628 |
| Epoch 0 | Batch 12599/57412 | Loss: 0.645 | time 1536.7723898887634 |
| Epoch 0 | Batch 12799/57412 | Loss: 0.234 | time 1560.6325562000275 |
| Epoch 0 | Batch 12999/57412 | Loss: 0.586 | time 1584.757630109787 |
| Epoch 0 | Batch 13199/57412 | Loss: 0.239 | time 1607.9595396518707 |
| Epoch 0 | Batch 13399/57412 | Loss: 0.504 | time 1631.935332775116 |
| Epoch 0 | Batch 13599/57412 | Loss: 0.301 | time 1654.8075489997864 |
| Epoch 0 | Batch 13799/57412 | Loss: 0.438 | time 1677.8050532341003 |
| Epoch 0 | Batch 13999/57412 | Loss: 0.271 | time 1700.1230466365814 |
| Epoch 0 | Batch 14199/57412 | Loss: 0.480 | time 1724.1268429756165 |
| Epoch 0 | Batch 14399/57412 | Loss: 0.219 | time 1748.9224302768707 |
| Epoch 0 | Batch 14599/57412 | Loss: 0.461 | time 1774.7814781665802 |
| Epoch 0 | Batch 14799/57412 | Loss: 0.357 | time 1799.6444628238678 |
| Epoch 0 | Batch 14999/57412 | Loss: 0.201 | time 1823.8527500629425 |
| Epoch 0 | Batch 15199/57412 | Loss: 0.196 | time 1848.6208901405334 |
| Epoch 0 | Batch 15399/57412 | Loss: 0.463 | time 1874.5012714862823 |
| Epoch 0 | Batch 15599/57412 | Loss: 0.398 | time 1900.8431751728058 |
| Epoch 0 | Batch 15799/57412 | Loss: 0.398 | time 1925.6662752628326 |
| Epoch 0 | Batch 15999/57412 | Loss: 0.322 | time 1949.3210337162018 |
| Epoch 0 | Batch 16199/57412 | Loss: 0.266 | time 1974.2808561325073 |
| Epoch 0 | Batch 16399/57412 | Loss: 0.516 | time 1998.869719028473 |
| Epoch 0 | Batch 16599/57412 | Loss: 0.400 | time 2024.1205747127533 |
| Epoch 0 | Batch 16799/57412 | Loss: 0.188 | time 2047.8166337013245 |
| Epoch 0 | Batch 16999/57412 | Loss: 0.527 | time 2072.562437057495 |
| Epoch 0 | Batch 17199/57412 | Loss: 0.389 | time 2096.376016139984 |
| Epoch 0 | Batch 17399/57412 | Loss: 0.352 | time 2120.4148445129395 |
| Epoch 0 | Batch 17599/57412 | Loss: 0.373 | time 2144.1434512138367 |
| Epoch 0 | Batch 17799/57412 | Loss: 0.093 | time 2167.1037895679474 |
| Epoch 0 | Batch 17999/57412 | Loss: 0.455 | time 2189.5077688694 |
| Epoch 0 | Batch 18199/57412 | Loss: 0.340 | time 2211.9870665073395 |
| Epoch 0 | Batch 18399/57412 | Loss: 0.120 | time 2237.6436297893524 |
| Epoch 0 | Batch 18599/57412 | Loss: 0.420 | time 2261.3650047779083 |
| Epoch 0 | Batch 18799/57412 | Loss: 0.445 | time 2285.582866668701 |
| Epoch 0 | Batch 18999/57412 | Loss: 0.617 | time 2309.438586950302 |
| Epoch 0 | Batch 19199/57412 | Loss: 0.494 | time 2332.075316429138 |
| Epoch 0 | Batch 19399/57412 | Loss: 0.225 | time 2355.727332353592 |
| Epoch 0 | Batch 19599/57412 | Loss: 0.367 | time 2380.163617372513 |
| Epoch 0 | Batch 19799/57412 | Loss: 0.069 | time 2404.466043949127 |
| Epoch 0 | Batch 19999/57412 | Loss: 0.547 | time 2427.0732679367065 |
| Saving at Step 20000 |
| Epoch 0 | Batch 20199/57412 | Loss: 0.268 | time 2451.154757976532 |
| Epoch 0 | Batch 20399/57412 | Loss: 0.279 | time 2474.665589570999 |
| Epoch 0 | Batch 20599/57412 | Loss: 0.183 | time 2497.5318336486816 |
| Epoch 0 | Batch 20799/57412 | Loss: 0.344 | time 2521.10409617424 |
| Epoch 0 | Batch 20999/57412 | Loss: 0.465 | time 2546.0237367153168 |
| Epoch 0 | Batch 21199/57412 | Loss: 0.664 | time 2570.0698957443237 |
| Epoch 0 | Batch 21399/57412 | Loss: 0.375 | time 2592.2841658592224 |
| Epoch 0 | Batch 21599/57412 | Loss: 0.570 | time 2615.341502428055 |
| Epoch 0 | Batch 21799/57412 | Loss: 0.191 | time 2638.018644094467 |
| Epoch 0 | Batch 21999/57412 | Loss: 0.389 | time 2662.3117306232452 |
| Epoch 0 | Batch 22199/57412 | Loss: 0.418 | time 2685.158451795578 |
| Epoch 0 | Batch 22399/57412 | Loss: 0.449 | time 2708.591796398163 |
| Epoch 0 | Batch 22599/57412 | Loss: 0.067 | time 2733.012747526169 |
| Epoch 0 | Batch 22799/57412 | Loss: 0.418 | time 2756.1395013332367 |
| Epoch 0 | Batch 22999/57412 | Loss: 0.574 | time 2780.2582771778107 |
| Epoch 0 | Batch 23199/57412 | Loss: 0.318 | time 2804.705315589905 |
| Epoch 0 | Batch 23399/57412 | Loss: 0.322 | time 2829.169781446457 |
| Epoch 0 | Batch 23599/57412 | Loss: 0.336 | time 2853.3061356544495 |
| Epoch 0 | Batch 23799/57412 | Loss: 0.204 | time 2877.527689933777 |
| Epoch 0 | Batch 23999/57412 | Loss: 0.531 | time 2900.8488504886627 |
| Epoch 0 | Batch 24199/57412 | Loss: 0.338 | time 2925.6459901332855 |
| Epoch 0 | Batch 24399/57412 | Loss: 0.566 | time 2950.412560224533 |
| Epoch 0 | Batch 24599/57412 | Loss: 0.387 | time 2975.329214811325 |
| Epoch 0 | Batch 24799/57412 | Loss: 0.453 | time 2998.73552775383 |
| Epoch 0 | Batch 24999/57412 | Loss: 0.248 | time 3022.5303337574005 |
| Epoch 0 | Batch 25199/57412 | Loss: 0.220 | time 3046.867803812027 |
| Epoch 0 | Batch 25399/57412 | Loss: 0.447 | time 3070.3551485538483 |
| Epoch 0 | Batch 25599/57412 | Loss: 0.354 | time 3093.7413532733917 |
| Epoch 0 | Batch 25799/57412 | Loss: 0.617 | time 3118.7690596580505 |
| Epoch 0 | Batch 25999/57412 | Loss: 0.126 | time 3142.931298017502 |
| Epoch 0 | Batch 26199/57412 | Loss: 0.406 | time 3166.9052934646606 |
| Epoch 0 | Batch 26399/57412 | Loss: 0.348 | time 3190.130454301834 |
| Epoch 0 | Batch 26599/57412 | Loss: 0.469 | time 3213.6062121391296 |
| Epoch 0 | Batch 26799/57412 | Loss: 0.486 | time 3237.373036623001 |
| Epoch 0 | Batch 26999/57412 | Loss: 0.520 | time 3261.3503799438477 |
| Epoch 0 | Batch 27199/57412 | Loss: 0.443 | time 3286.1028592586517 |
| Epoch 0 | Batch 27399/57412 | Loss: 0.512 | time 3309.3578732013702 |
| Epoch 0 | Batch 27599/57412 | Loss: 0.520 | time 3332.533192873001 |
| Epoch 0 | Batch 27799/57412 | Loss: 0.332 | time 3355.1274511814117 |
| Epoch 0 | Batch 27999/57412 | Loss: 0.680 | time 3378.7428274154663 |
| Epoch 0 | Batch 28199/57412 | Loss: 0.586 | time 3402.6961975097656 |
| Epoch 0 | Batch 28399/57412 | Loss: 0.449 | time 3426.8572759628296 |
| Epoch 0 | Batch 28599/57412 | Loss: 0.389 | time 3450.477519273758 |
| Epoch 0 | Batch 28799/57412 | Loss: 0.543 | time 3472.0067903995514 |
| Epoch 0 | Batch 28999/57412 | Loss: 0.355 | time 3495.249744653702 |
| Epoch 0 | Batch 29199/57412 | Loss: 0.527 | time 3520.162987947464 |
| Epoch 0 | Batch 29399/57412 | Loss: 0.305 | time 3544.0286009311676 |
| Epoch 0 | Batch 29599/57412 | Loss: 0.402 | time 3568.1937918663025 |
| Epoch 0 | Batch 29799/57412 | Loss: 0.711 | time 3591.7185487747192 |
| Epoch 0 | Batch 29999/57412 | Loss: 0.621 | time 3615.0912561416626 |
| Epoch 0 | Batch 30199/57412 | Loss: 0.453 | time 3638.2557113170624 |
| Epoch 0 | Batch 30399/57412 | Loss: 0.318 | time 3663.7334401607513 |
| Epoch 0 | Batch 30599/57412 | Loss: 0.271 | time 3688.8847608566284 |
| Epoch 0 | Batch 30799/57412 | Loss: 0.496 | time 3712.4490633010864 |
| Epoch 0 | Batch 30999/57412 | Loss: 0.354 | time 3736.9057552814484 |
| Epoch 0 | Batch 31199/57412 | Loss: 0.770 | time 3761.335116624832 |
| Epoch 0 | Batch 31399/57412 | Loss: 0.065 | time 3785.959890604019 |
| Epoch 0 | Batch 31599/57412 | Loss: 0.645 | time 3810.4877939224243 |
| Epoch 0 | Batch 31799/57412 | Loss: 0.334 | time 3834.790798664093 |
| Epoch 0 | Batch 31999/57412 | Loss: 0.578 | time 3860.0233290195465 |
| Epoch 0 | Batch 32199/57412 | Loss: 0.723 | time 3884.1123950481415 |
| Epoch 0 | Batch 32399/57412 | Loss: 0.266 | time 3908.1102936267853 |
| Epoch 0 | Batch 32599/57412 | Loss: 0.490 | time 3931.4471797943115 |
| Epoch 0 | Batch 32799/57412 | Loss: 0.621 | time 3956.3029255867004 |
| Epoch 0 | Batch 32999/57412 | Loss: 0.691 | time 3979.3164319992065 |
| Epoch 0 | Batch 33199/57412 | Loss: 0.531 | time 4003.886766433716 |
| Epoch 0 | Batch 33399/57412 | Loss: 0.238 | time 4027.869696855545 |
| Epoch 0 | Batch 33599/57412 | Loss: 0.381 | time 4051.880104780197 |
| Epoch 0 | Batch 33799/57412 | Loss: 0.273 | time 4076.1642718315125 |
| Epoch 0 | Batch 33999/57412 | Loss: 0.438 | time 4100.932791471481 |
| Epoch 0 | Batch 34199/57412 | Loss: 0.434 | time 4125.147915840149 |
| Epoch 0 | Batch 34399/57412 | Loss: 0.367 | time 4150.2091681957245 |
| Epoch 0 | Batch 34599/57412 | Loss: 0.535 | time 4176.217367649078 |
| Epoch 0 | Batch 34799/57412 | Loss: 0.486 | time 4200.821031570435 |
| Epoch 0 | Batch 34999/57412 | Loss: 0.750 | time 4226.511318922043 |
| Epoch 0 | Batch 35199/57412 | Loss: 0.385 | time 4250.640113830566 |
| Epoch 0 | Batch 35399/57412 | Loss: 0.275 | time 4275.789942741394 |
| Epoch 0 | Batch 35599/57412 | Loss: 0.361 | time 4299.5063762664795 |
| Epoch 0 | Batch 35799/57412 | Loss: 0.582 | time 4323.474545717239 |
| Epoch 0 | Batch 35999/57412 | Loss: 0.504 | time 4347.578270435333 |
| Epoch 0 | Batch 36199/57412 | Loss: 0.113 | time 4372.06854724884 |
| Epoch 0 | Batch 36399/57412 | Loss: 0.375 | time 4395.904716491699 |
| Epoch 0 | Batch 36599/57412 | Loss: 0.264 | time 4418.497104644775 |
| Epoch 0 | Batch 36799/57412 | Loss: 0.447 | time 4441.302644252777 |
| Epoch 0 | Batch 36999/57412 | Loss: 0.155 | time 4464.565608263016 |
| Epoch 0 | Batch 37199/57412 | Loss: 0.504 | time 4487.521849155426 |
| Epoch 0 | Batch 37399/57412 | Loss: 0.289 | time 4510.895763158798 |
| Epoch 0 | Batch 37599/57412 | Loss: 0.439 | time 4533.07492351532 |
| Epoch 0 | Batch 37799/57412 | Loss: 0.379 | time 4555.424180984497 |
| Epoch 0 | Batch 37999/57412 | Loss: 0.365 | time 4579.596609830856 |
| Epoch 0 | Batch 38199/57412 | Loss: 0.157 | time 4605.51198887825 |
| Epoch 0 | Batch 38399/57412 | Loss: 0.287 | time 4628.968412399292 |
| Epoch 0 | Batch 38599/57412 | Loss: 0.228 | time 4652.826132297516 |
| Epoch 0 | Batch 38799/57412 | Loss: 0.738 | time 4677.130941867828 |
| Epoch 0 | Batch 38999/57412 | Loss: 0.543 | time 4702.575179100037 |
| Epoch 0 | Batch 39199/57412 | Loss: 0.633 | time 4726.572310447693 |
| Epoch 0 | Batch 39399/57412 | Loss: 0.395 | time 4751.495271682739 |
| Epoch 0 | Batch 39599/57412 | Loss: 0.742 | time 4776.32727265358 |
| Epoch 0 | Batch 39799/57412 | Loss: 0.633 | time 4801.217889308929 |
| Epoch 0 | Batch 39999/57412 | Loss: 0.605 | time 4826.108942270279 |
| Saving at Step 40000 |
| Epoch 0 | Batch 40199/57412 | Loss: 0.058 | time 4852.700262784958 |
| Epoch 0 | Batch 40399/57412 | Loss: 0.715 | time 4874.720931529999 |
| Epoch 0 | Batch 40599/57412 | Loss: 0.379 | time 4898.251784086227 |
| Epoch 0 | Batch 40799/57412 | Loss: 0.307 | time 4921.193106889725 |
| Epoch 0 | Batch 40999/57412 | Loss: 0.291 | time 4946.638162851334 |
| Epoch 0 | Batch 41199/57412 | Loss: 0.303 | time 4970.674071550369 |
| Epoch 0 | Batch 41399/57412 | Loss: 0.369 | time 4995.7080755233765 |
| Epoch 0 | Batch 41599/57412 | Loss: 0.422 | time 5019.358082771301 |
| Epoch 0 | Batch 41799/57412 | Loss: 0.184 | time 5044.459336519241 |
| Epoch 0 | Batch 41999/57412 | Loss: 0.420 | time 5068.2931599617 |
| Epoch 0 | Batch 42199/57412 | Loss: 0.283 | time 5094.243029594421 |
| Epoch 0 | Batch 42399/57412 | Loss: 0.311 | time 5119.103428602219 |
| Epoch 0 | Batch 42599/57412 | Loss: 0.555 | time 5143.702688694 |
| Epoch 0 | Batch 42799/57412 | Loss: 0.424 | time 5168.41917014122 |
| Epoch 0 | Batch 42999/57412 | Loss: 0.578 | time 5191.903990268707 |
| Epoch 0 | Batch 43199/57412 | Loss: 0.354 | time 5217.4181134700775 |
| Epoch 0 | Batch 43399/57412 | Loss: 0.438 | time 5241.681266069412 |
| Epoch 0 | Batch 43599/57412 | Loss: 0.367 | time 5267.236545324326 |
| Epoch 0 | Batch 43799/57412 | Loss: 0.535 | time 5291.452278137207 |
| Epoch 0 | Batch 43999/57412 | Loss: 0.605 | time 5315.792713165283 |
| Epoch 0 | Batch 44199/57412 | Loss: 0.605 | time 5339.558612823486 |
| Epoch 0 | Batch 44399/57412 | Loss: 0.428 | time 5364.789148807526 |
| Epoch 0 | Batch 44599/57412 | Loss: 0.457 | time 5389.847823143005 |
| Epoch 0 | Batch 44799/57412 | Loss: 0.379 | time 5413.355014801025 |
| Epoch 0 | Batch 44999/57412 | Loss: 0.420 | time 5436.635505914688 |
| Epoch 0 | Batch 45199/57412 | Loss: 0.512 | time 5461.399576187134 |
| Epoch 0 | Batch 45399/57412 | Loss: 0.445 | time 5485.589755058289 |
| Epoch 0 | Batch 45599/57412 | Loss: 0.482 | time 5510.534025192261 |
| Epoch 0 | Batch 45799/57412 | Loss: 0.307 | time 5533.845446109772 |
| Epoch 0 | Batch 45999/57412 | Loss: 0.379 | time 5557.813657045364 |
| Epoch 0 | Batch 46199/57412 | Loss: 0.523 | time 5581.582619428635 |
| Epoch 0 | Batch 46399/57412 | Loss: 0.222 | time 5604.651105165482 |
| Epoch 0 | Batch 46599/57412 | Loss: 0.322 | time 5630.277412414551 |
| Epoch 0 | Batch 46799/57412 | Loss: 0.695 | time 5654.585182905197 |
| Epoch 0 | Batch 46999/57412 | Loss: 0.387 | time 5679.123800754547 |
| Epoch 0 | Batch 47199/57412 | Loss: 0.322 | time 5702.963451862335 |
| Epoch 0 | Batch 47399/57412 | Loss: 0.430 | time 5724.807294845581 |
| Epoch 0 | Batch 47599/57412 | Loss: 0.393 | time 5748.897443771362 |
| Epoch 0 | Batch 47799/57412 | Loss: 0.355 | time 5771.973505496979 |
| Epoch 0 | Batch 47999/57412 | Loss: 0.201 | time 5796.525881290436 |
| Epoch 0 | Batch 48199/57412 | Loss: 0.652 | time 5821.215528964996 |
| Epoch 0 | Batch 48399/57412 | Loss: 0.328 | time 5843.753592967987 |
| Epoch 0 | Batch 48599/57412 | Loss: 0.488 | time 5867.465299367905 |
| Epoch 0 | Batch 48799/57412 | Loss: 0.547 | time 5889.371001958847 |
| Epoch 0 | Batch 48999/57412 | Loss: 0.523 | time 5912.661686420441 |
| Epoch 0 | Batch 49199/57412 | Loss: 0.367 | time 5935.9240844249725 |
| Epoch 0 | Batch 49399/57412 | Loss: 0.385 | time 5959.409880399704 |
| Epoch 0 | Batch 49599/57412 | Loss: 0.449 | time 5983.658529520035 |
| Epoch 0 | Batch 49799/57412 | Loss: 0.355 | time 6007.124490737915 |
| Epoch 0 | Batch 49999/57412 | Loss: 0.443 | time 6030.677422761917 |
| Epoch 0 | Batch 50199/57412 | Loss: 0.354 | time 6054.396829843521 |
| Epoch 0 | Batch 50399/57412 | Loss: 0.438 | time 6078.219150066376 |
| Epoch 0 | Batch 50599/57412 | Loss: 0.617 | time 6101.245578050613 |
| Epoch 0 | Batch 50799/57412 | Loss: 0.346 | time 6124.954098701477 |
| Epoch 0 | Batch 50999/57412 | Loss: 0.578 | time 6150.026329040527 |
| Epoch 0 | Batch 51199/57412 | Loss: 0.570 | time 6173.494077920914 |
| Epoch 0 | Batch 51399/57412 | Loss: 0.338 | time 6198.053581953049 |
| Epoch 0 | Batch 51599/57412 | Loss: 0.412 | time 6221.437692165375 |
| Epoch 0 | Batch 51799/57412 | Loss: 0.480 | time 6244.654960155487 |
| Epoch 0 | Batch 51999/57412 | Loss: 0.598 | time 6268.151013612747 |
| Epoch 0 | Batch 52199/57412 | Loss: 0.416 | time 6293.279064655304 |
| Epoch 0 | Batch 52399/57412 | Loss: 0.406 | time 6317.517779350281 |
| Epoch 0 | Batch 52599/57412 | Loss: 0.641 | time 6342.51355099678 |
| Epoch 0 | Batch 52799/57412 | Loss: 0.175 | time 6367.97954916954 |
| Epoch 0 | Batch 52999/57412 | Loss: 0.914 | time 6391.8965129852295 |
| Epoch 0 | Batch 53199/57412 | Loss: 0.457 | time 6415.360770225525 |
| Epoch 0 | Batch 53399/57412 | Loss: 0.130 | time 6439.640261650085 |
| Epoch 0 | Batch 53599/57412 | Loss: 0.266 | time 6463.693968057632 |
| Epoch 0 | Batch 53799/57412 | Loss: 0.389 | time 6488.471304178238 |
| Epoch 0 | Batch 53999/57412 | Loss: 0.229 | time 6514.434978723526 |
| Epoch 0 | Batch 54199/57412 | Loss: 0.594 | time 6540.400573015213 |
| Epoch 0 | Batch 54399/57412 | Loss: 0.500 | time 6565.654086112976 |
| Epoch 0 | Batch 54599/57412 | Loss: 0.461 | time 6590.569655895233 |
| Epoch 0 | Batch 54799/57412 | Loss: 0.424 | time 6614.36125254631 |
| Epoch 0 | Batch 54999/57412 | Loss: 0.664 | time 6639.330735206604 |
| Epoch 0 | Batch 55199/57412 | Loss: 0.357 | time 6664.421330213547 |
| Epoch 0 | Batch 55399/57412 | Loss: 0.449 | time 6688.44490146637 |
| Epoch 0 | Batch 55599/57412 | Loss: 0.424 | time 6713.240394115448 |
| Epoch 0 | Batch 55799/57412 | Loss: 0.555 | time 6738.33109998703 |
| Epoch 0 | Batch 55999/57412 | Loss: 0.082 | time 6762.313829660416 |
| Epoch 0 | Batch 56199/57412 | Loss: 0.289 | time 6786.542687654495 |
| Epoch 0 | Batch 56399/57412 | Loss: 0.123 | time 6809.9832882881165 |
| Epoch 0 | Batch 56599/57412 | Loss: 0.330 | time 6836.372585296631 |
| Epoch 0 | Batch 56799/57412 | Loss: 0.467 | time 6859.459310531616 |
| Epoch 0 | Batch 56999/57412 | Loss: 0.179 | time 6884.382868051529 |
| Epoch 0 | Batch 57199/57412 | Loss: 0.389 | time 6908.809622049332 |
| Epoch 0 | Batch 57399/57412 | Loss: 0.279 | time 6932.565470218658 |
| ========================================================================================= |
| ========================================================================================= |
| ========================================================================================= |
| Epoch 0 | Validation Samples 5082 | Validation Acc: 0.3976780794962613 | time 340.7869293689728 |
| ========================================================================================= |
| Saving at Epoch 0 | Learning rate: 2e-05 |
| Epoch 1 | Batch 199/57412 | Loss: 0.428 | time 25.152500867843628 |
| Epoch 1 | Batch 399/57412 | Loss: 0.277 | time 49.564815044403076 |
| Epoch 1 | Batch 599/57412 | Loss: 0.422 | time 73.68636846542358 |
| Epoch 1 | Batch 799/57412 | Loss: 0.395 | time 97.34948992729187 |
| Epoch 1 | Batch 999/57412 | Loss: 0.498 | time 120.66676688194275 |
| Epoch 1 | Batch 1199/57412 | Loss: 0.477 | time 144.18094658851624 |
| Epoch 1 | Batch 1399/57412 | Loss: 0.668 | time 169.6847264766693 |
| Epoch 1 | Batch 1599/57412 | Loss: 0.520 | time 195.13828539848328 |
| Epoch 1 | Batch 1799/57412 | Loss: 0.461 | time 218.20551872253418 |
| Epoch 1 | Batch 1999/57412 | Loss: 0.173 | time 241.74184107780457 |
| Epoch 1 | Batch 2199/57412 | Loss: 0.229 | time 264.57536721229553 |
| Epoch 1 | Batch 2399/57412 | Loss: 0.734 | time 287.40018677711487 |
| Epoch 1 | Batch 2599/57412 | Loss: 0.393 | time 312.5087516307831 |
| Epoch 1 | Batch 2799/57412 | Loss: 0.359 | time 336.1284291744232 |
| Epoch 1 | Batch 2999/57412 | Loss: 0.512 | time 359.8480806350708 |
| Epoch 1 | Batch 3199/57412 | Loss: 0.566 | time 382.99631524086 |
| Epoch 1 | Batch 3399/57412 | Loss: 0.354 | time 406.73514771461487 |
| Epoch 1 | Batch 3599/57412 | Loss: 0.434 | time 429.79076504707336 |
| Epoch 1 | Batch 3799/57412 | Loss: 0.479 | time 453.423410654068 |
| Epoch 1 | Batch 3999/57412 | Loss: 0.228 | time 476.54126358032227 |
| Epoch 1 | Batch 4199/57412 | Loss: 0.547 | time 498.7276725769043 |
| Epoch 1 | Batch 4399/57412 | Loss: 0.664 | time 522.2046229839325 |
| Epoch 1 | Batch 4599/57412 | Loss: 0.283 | time 545.6331901550293 |
| Epoch 1 | Batch 4799/57412 | Loss: 0.385 | time 568.2640671730042 |
| Epoch 1 | Batch 4999/57412 | Loss: 0.527 | time 590.5029389858246 |
| Epoch 1 | Batch 5199/57412 | Loss: 0.494 | time 613.8660070896149 |
| Epoch 1 | Batch 5399/57412 | Loss: 0.361 | time 636.8751304149628 |
| Epoch 1 | Batch 5599/57412 | Loss: 0.441 | time 660.0274250507355 |
| Epoch 1 | Batch 5799/57412 | Loss: 0.389 | time 683.7851514816284 |
| Epoch 1 | Batch 5999/57412 | Loss: 0.555 | time 706.8026134967804 |
| Epoch 1 | Batch 6199/57412 | Loss: 0.455 | time 729.6186363697052 |
| Epoch 1 | Batch 6399/57412 | Loss: 0.281 | time 753.4973366260529 |
| Epoch 1 | Batch 6599/57412 | Loss: 0.324 | time 776.3994233608246 |
| Epoch 1 | Batch 6799/57412 | Loss: 0.473 | time 800.6006350517273 |
| Epoch 1 | Batch 6999/57412 | Loss: 0.494 | time 823.4411687850952 |
| Epoch 1 | Batch 7199/57412 | Loss: 0.340 | time 845.7594308853149 |
| Epoch 1 | Batch 7399/57412 | Loss: 0.396 | time 868.8487038612366 |
| Epoch 1 | Batch 7599/57412 | Loss: 0.605 | time 892.22705078125 |
| Epoch 1 | Batch 7799/57412 | Loss: 0.934 | time 915.1435506343842 |
| Epoch 1 | Batch 7999/57412 | Loss: 0.664 | time 937.6753394603729 |
| Epoch 1 | Batch 8199/57412 | Loss: 0.229 | time 961.0526208877563 |
| Epoch 1 | Batch 8399/57412 | Loss: 0.494 | time 984.3599088191986 |
| Epoch 1 | Batch 8599/57412 | Loss: 0.291 | time 1007.3717477321625 |
| Epoch 1 | Batch 8799/57412 | Loss: 0.609 | time 1032.531198501587 |
| Epoch 1 | Batch 8999/57412 | Loss: 0.516 | time 1056.4528493881226 |
| Epoch 1 | Batch 9199/57412 | Loss: 0.676 | time 1080.419026851654 |
| Epoch 1 | Batch 9399/57412 | Loss: 0.508 | time 1104.7033867835999 |
| Epoch 1 | Batch 9599/57412 | Loss: 0.383 | time 1128.7875192165375 |
| Epoch 1 | Batch 9799/57412 | Loss: 0.093 | time 1152.2985162734985 |
| Epoch 1 | Batch 9999/57412 | Loss: 0.520 | time 1175.8067212104797 |
| Epoch 1 | Batch 10199/57412 | Loss: 0.229 | time 1198.747262954712 |
| Epoch 1 | Batch 10399/57412 | Loss: 0.562 | time 1222.4107995033264 |
| Epoch 1 | Batch 10599/57412 | Loss: 0.377 | time 1244.5697796344757 |
| Epoch 1 | Batch 10799/57412 | Loss: 0.303 | time 1267.9775586128235 |
| Epoch 1 | Batch 10999/57412 | Loss: 0.240 | time 1290.2110166549683 |
| Epoch 1 | Batch 11199/57412 | Loss: 0.283 | time 1313.712236881256 |
| Epoch 1 | Batch 11399/57412 | Loss: 0.605 | time 1336.6876122951508 |
| Epoch 1 | Batch 11599/57412 | Loss: 0.365 | time 1361.7324969768524 |
| Epoch 1 | Batch 11799/57412 | Loss: 0.363 | time 1383.971843957901 |
| Epoch 1 | Batch 11999/57412 | Loss: 0.482 | time 1407.5552432537079 |
| Epoch 1 | Batch 12199/57412 | Loss: 0.508 | time 1430.982731103897 |
| Epoch 1 | Batch 12399/57412 | Loss: 0.535 | time 1454.2700922489166 |
| Epoch 1 | Batch 12599/57412 | Loss: 0.598 | time 1478.2867958545685 |
| Epoch 1 | Batch 12799/57412 | Loss: 0.230 | time 1501.4793057441711 |
| Epoch 1 | Batch 12999/57412 | Loss: 0.582 | time 1524.7268381118774 |
| Epoch 1 | Batch 13199/57412 | Loss: 0.184 | time 1548.1157376766205 |
| Epoch 1 | Batch 13399/57412 | Loss: 0.488 | time 1572.3292875289917 |
| Epoch 1 | Batch 13599/57412 | Loss: 0.289 | time 1595.2204329967499 |
| Epoch 1 | Batch 13799/57412 | Loss: 0.428 | time 1618.7670669555664 |
| Epoch 1 | Batch 13999/57412 | Loss: 0.262 | time 1641.4741876125336 |
| Epoch 1 | Batch 14199/57412 | Loss: 0.469 | time 1665.3712487220764 |
| Epoch 1 | Batch 14399/57412 | Loss: 0.214 | time 1689.126660823822 |
| Epoch 1 | Batch 14599/57412 | Loss: 0.453 | time 1713.3327107429504 |
| Epoch 1 | Batch 14799/57412 | Loss: 0.348 | time 1736.7060208320618 |
| Epoch 1 | Batch 14999/57412 | Loss: 0.195 | time 1759.723312139511 |
| Epoch 1 | Batch 15199/57412 | Loss: 0.185 | time 1783.1578316688538 |
| Epoch 1 | Batch 15399/57412 | Loss: 0.455 | time 1807.8174934387207 |
| Epoch 1 | Batch 15599/57412 | Loss: 0.393 | time 1832.6342480182648 |
| Epoch 1 | Batch 15799/57412 | Loss: 0.387 | time 1856.322815656662 |
| Epoch 1 | Batch 15999/57412 | Loss: 0.316 | time 1878.8119449615479 |
| Epoch 1 | Batch 16199/57412 | Loss: 0.202 | time 1902.407318353653 |
| Epoch 1 | Batch 16399/57412 | Loss: 0.512 | time 1925.6256864070892 |
| Epoch 1 | Batch 16599/57412 | Loss: 0.385 | time 1949.510419845581 |
| Epoch 1 | Batch 16799/57412 | Loss: 0.182 | time 1972.1610343456268 |
| Epoch 1 | Batch 16999/57412 | Loss: 0.504 | time 1995.8035311698914 |
| Epoch 1 | Batch 17199/57412 | Loss: 0.377 | time 2018.5414729118347 |
| Epoch 1 | Batch 17399/57412 | Loss: 0.256 | time 2041.855714082718 |
| Epoch 1 | Batch 17599/57412 | Loss: 0.379 | time 2065.8055164813995 |
| Epoch 1 | Batch 17799/57412 | Loss: 0.092 | time 2088.962349176407 |
| Epoch 1 | Batch 17999/57412 | Loss: 0.426 | time 2110.973618030548 |
| Epoch 1 | Batch 18199/57412 | Loss: 0.334 | time 2133.358871936798 |
| Epoch 1 | Batch 18399/57412 | Loss: 0.112 | time 2158.96036529541 |
| Epoch 1 | Batch 18599/57412 | Loss: 0.408 | time 2182.4746770858765 |
| Epoch 1 | Batch 18799/57412 | Loss: 0.438 | time 2206.287404537201 |
| Epoch 1 | Batch 18999/57412 | Loss: 0.582 | time 2229.1306812763214 |
| Epoch 1 | Batch 19199/57412 | Loss: 0.479 | time 2251.9099671840668 |
| Epoch 1 | Batch 19399/57412 | Loss: 0.207 | time 2275.59796833992 |
| Epoch 1 | Batch 19599/57412 | Loss: 0.359 | time 2299.762620449066 |
| Epoch 1 | Batch 19799/57412 | Loss: 0.081 | time 2323.8409507274628 |
| Epoch 1 | Batch 19999/57412 | Loss: 0.539 | time 2346.0704956054688 |
| Saving at Step 20000 |
| Epoch 1 | Batch 20199/57412 | Loss: 0.258 | time 2372.0281767845154 |
| Epoch 1 | Batch 20399/57412 | Loss: 0.279 | time 2395.89617729187 |
| Epoch 1 | Batch 20599/57412 | Loss: 0.129 | time 2418.276226758957 |
| Epoch 1 | Batch 20799/57412 | Loss: 0.338 | time 2442.074633836746 |
| Epoch 1 | Batch 20999/57412 | Loss: 0.434 | time 2466.56094789505 |
| Epoch 1 | Batch 21199/57412 | Loss: 0.641 | time 2490.335264444351 |
| Epoch 1 | Batch 21399/57412 | Loss: 0.330 | time 2512.353021621704 |
| Epoch 1 | Batch 21599/57412 | Loss: 0.570 | time 2534.728078842163 |
| Epoch 1 | Batch 21799/57412 | Loss: 0.184 | time 2557.000411748886 |
| Epoch 1 | Batch 21999/57412 | Loss: 0.381 | time 2580.322984457016 |
| Epoch 1 | Batch 22199/57412 | Loss: 0.414 | time 2602.487189769745 |
| Epoch 1 | Batch 22399/57412 | Loss: 0.438 | time 2625.698721408844 |
| Epoch 1 | Batch 22599/57412 | Loss: 0.065 | time 2649.2276928424835 |
| Epoch 1 | Batch 22799/57412 | Loss: 0.383 | time 2671.919160604477 |
| Epoch 1 | Batch 22999/57412 | Loss: 0.559 | time 2695.3723196983337 |
| Epoch 1 | Batch 23199/57412 | Loss: 0.295 | time 2718.9435110092163 |
| Epoch 1 | Batch 23399/57412 | Loss: 0.316 | time 2742.5084125995636 |
| Epoch 1 | Batch 23599/57412 | Loss: 0.322 | time 2765.1948988437653 |
| Epoch 1 | Batch 23799/57412 | Loss: 0.196 | time 2788.7875661849976 |
| Epoch 1 | Batch 23999/57412 | Loss: 0.494 | time 2811.5863292217255 |
| Epoch 1 | Batch 24199/57412 | Loss: 0.328 | time 2835.380228996277 |
| Epoch 1 | Batch 24399/57412 | Loss: 0.547 | time 2859.787971019745 |
| Epoch 1 | Batch 24599/57412 | Loss: 0.379 | time 2884.0211255550385 |
| Epoch 1 | Batch 24799/57412 | Loss: 0.443 | time 2906.817175388336 |
| Epoch 1 | Batch 24999/57412 | Loss: 0.248 | time 2930.0824539661407 |
| Epoch 1 | Batch 25199/57412 | Loss: 0.199 | time 2953.633894443512 |
| Epoch 1 | Batch 25399/57412 | Loss: 0.432 | time 2976.705135822296 |
| Epoch 1 | Batch 25599/57412 | Loss: 0.342 | time 2999.3975088596344 |
| Epoch 1 | Batch 25799/57412 | Loss: 0.602 | time 3023.9223289489746 |
| Epoch 1 | Batch 25999/57412 | Loss: 0.122 | time 3047.4264357089996 |
| Epoch 1 | Batch 26199/57412 | Loss: 0.379 | time 3070.51353764534 |
| Epoch 1 | Batch 26399/57412 | Loss: 0.342 | time 3093.214451789856 |
| Epoch 1 | Batch 26599/57412 | Loss: 0.451 | time 3115.8971202373505 |
| Epoch 1 | Batch 26799/57412 | Loss: 0.447 | time 3139.0638749599457 |
| Epoch 1 | Batch 26999/57412 | Loss: 0.492 | time 3162.2074117660522 |
| Epoch 1 | Batch 27199/57412 | Loss: 0.438 | time 3186.7575070858 |
| Epoch 1 | Batch 27399/57412 | Loss: 0.508 | time 3210.2178559303284 |
| Epoch 1 | Batch 27599/57412 | Loss: 0.504 | time 3233.1716253757477 |
| Epoch 1 | Batch 27799/57412 | Loss: 0.314 | time 3255.9679923057556 |
| Epoch 1 | Batch 27999/57412 | Loss: 0.676 | time 3279.3701469898224 |
| Epoch 1 | Batch 28199/57412 | Loss: 0.551 | time 3302.851050376892 |
| Epoch 1 | Batch 28399/57412 | Loss: 0.428 | time 3327.081267595291 |
| Epoch 1 | Batch 28599/57412 | Loss: 0.363 | time 3350.6625277996063 |
| Epoch 1 | Batch 28799/57412 | Loss: 0.539 | time 3373.0329973697662 |
| Epoch 1 | Batch 28999/57412 | Loss: 0.352 | time 3395.8402297496796 |
| Epoch 1 | Batch 29199/57412 | Loss: 0.463 | time 3419.6369297504425 |
| Epoch 1 | Batch 29399/57412 | Loss: 0.295 | time 3442.4504585266113 |
| Epoch 1 | Batch 29599/57412 | Loss: 0.393 | time 3465.6357858181 |
| Epoch 1 | Batch 29799/57412 | Loss: 0.578 | time 3488.061747789383 |
| Epoch 1 | Batch 29999/57412 | Loss: 0.602 | time 3510.0549166202545 |
| Epoch 1 | Batch 30199/57412 | Loss: 0.451 | time 3531.7149834632874 |
| Epoch 1 | Batch 30399/57412 | Loss: 0.309 | time 3556.2649099826813 |
| Epoch 1 | Batch 30599/57412 | Loss: 0.262 | time 3580.1146936416626 |
| Epoch 1 | Batch 30799/57412 | Loss: 0.492 | time 3602.40949511528 |
| Epoch 1 | Batch 30999/57412 | Loss: 0.344 | time 3625.674136161804 |
| Epoch 1 | Batch 31199/57412 | Loss: 0.535 | time 3648.822820186615 |
| Epoch 1 | Batch 31399/57412 | Loss: 0.061 | time 3672.259842157364 |
| Epoch 1 | Batch 31599/57412 | Loss: 0.625 | time 3695.1593532562256 |
| Epoch 1 | Batch 31799/57412 | Loss: 0.330 | time 3718.5235373973846 |
| Epoch 1 | Batch 31999/57412 | Loss: 0.559 | time 3742.4218351840973 |
| Epoch 1 | Batch 32199/57412 | Loss: 0.699 | time 3765.1825664043427 |
| Epoch 1 | Batch 32399/57412 | Loss: 0.246 | time 3787.8695328235626 |
| Epoch 1 | Batch 32599/57412 | Loss: 0.482 | time 3809.749710083008 |
| Epoch 1 | Batch 32799/57412 | Loss: 0.609 | time 3833.3986933231354 |
| Epoch 1 | Batch 32999/57412 | Loss: 0.672 | time 3855.1701023578644 |
| Epoch 1 | Batch 33199/57412 | Loss: 0.516 | time 3877.9804286956787 |
| Epoch 1 | Batch 33399/57412 | Loss: 0.211 | time 3900.6474726200104 |
| Epoch 1 | Batch 33599/57412 | Loss: 0.316 | time 3923.086834669113 |
| Epoch 1 | Batch 33799/57412 | Loss: 0.252 | time 3946.0216529369354 |
| Epoch 1 | Batch 33999/57412 | Loss: 0.432 | time 3969.240823984146 |
| Epoch 1 | Batch 34199/57412 | Loss: 0.422 | time 3991.985395669937 |
| Epoch 1 | Batch 34399/57412 | Loss: 0.357 | time 4015.777585029602 |
| Epoch 1 | Batch 34599/57412 | Loss: 0.520 | time 4040.348848581314 |
| Epoch 1 | Batch 34799/57412 | Loss: 0.465 | time 4064.5692613124847 |
| Epoch 1 | Batch 34999/57412 | Loss: 0.730 | time 4090.1602244377136 |
| Epoch 1 | Batch 35199/57412 | Loss: 0.379 | time 4114.253205299377 |
| Epoch 1 | Batch 35399/57412 | Loss: 0.270 | time 4139.293807983398 |
| Epoch 1 | Batch 35599/57412 | Loss: 0.354 | time 4162.483316898346 |
| Epoch 1 | Batch 35799/57412 | Loss: 0.555 | time 4185.9932951927185 |
| Epoch 1 | Batch 35999/57412 | Loss: 0.500 | time 4209.581789493561 |
| Epoch 1 | Batch 36199/57412 | Loss: 0.103 | time 4233.752952098846 |
| Epoch 1 | Batch 36399/57412 | Loss: 0.352 | time 4258.287425994873 |
| Epoch 1 | Batch 36599/57412 | Loss: 0.254 | time 4281.933899402618 |
| Epoch 1 | Batch 36799/57412 | Loss: 0.432 | time 4306.168152809143 |
| Epoch 1 | Batch 36999/57412 | Loss: 0.140 | time 4330.777748346329 |
| Epoch 1 | Batch 37199/57412 | Loss: 0.486 | time 4355.102669715881 |
| Epoch 1 | Batch 37399/57412 | Loss: 0.271 | time 4380.077262639999 |
| Epoch 1 | Batch 37599/57412 | Loss: 0.432 | time 4403.5635550022125 |
| Epoch 1 | Batch 37799/57412 | Loss: 0.375 | time 4427.331566095352 |
| Epoch 1 | Batch 37999/57412 | Loss: 0.357 | time 4451.232823133469 |
| Epoch 1 | Batch 38199/57412 | Loss: 0.001 | time 4476.849972486496 |
| Epoch 1 | Batch 38399/57412 | Loss: 0.279 | time 4500.337634086609 |
| Epoch 1 | Batch 38599/57412 | Loss: 0.223 | time 4524.257683753967 |
| Epoch 1 | Batch 38799/57412 | Loss: 0.688 | time 4548.587116718292 |
| Epoch 1 | Batch 38999/57412 | Loss: 0.523 | time 4573.6334726810455 |
| Epoch 1 | Batch 39199/57412 | Loss: 0.590 | time 4597.352431774139 |
| Epoch 1 | Batch 39399/57412 | Loss: 0.396 | time 4622.310613393784 |
| Epoch 1 | Batch 39599/57412 | Loss: 0.723 | time 4646.819803237915 |
| Epoch 1 | Batch 39799/57412 | Loss: 0.582 | time 4671.651681661606 |
| Epoch 1 | Batch 39999/57412 | Loss: 0.602 | time 4696.158849477768 |
| Saving at Step 40000 |
| Epoch 1 | Batch 40199/57412 | Loss: 0.006 | time 4724.224468946457 |
| Epoch 1 | Batch 40399/57412 | Loss: 0.688 | time 4747.648717880249 |
| Epoch 1 | Batch 40599/57412 | Loss: 0.354 | time 4772.150272369385 |
| Epoch 1 | Batch 40799/57412 | Loss: 0.303 | time 4795.247138738632 |
| Epoch 1 | Batch 40999/57412 | Loss: 0.285 | time 4820.5355706214905 |
| Epoch 1 | Batch 41199/57412 | Loss: 0.303 | time 4844.179864883423 |
| Epoch 1 | Batch 41399/57412 | Loss: 0.359 | time 4869.24219417572 |
| Epoch 1 | Batch 41599/57412 | Loss: 0.410 | time 4892.707112073898 |
| Epoch 1 | Batch 41799/57412 | Loss: 0.182 | time 4917.45814037323 |
| Epoch 1 | Batch 41999/57412 | Loss: 0.408 | time 4939.935014486313 |
| Epoch 1 | Batch 42199/57412 | Loss: 0.275 | time 4964.8333349227905 |
| Epoch 1 | Batch 42399/57412 | Loss: 0.301 | time 4988.383727073669 |
| Epoch 1 | Batch 42599/57412 | Loss: 0.547 | time 5011.639510631561 |
| Epoch 1 | Batch 42799/57412 | Loss: 0.387 | time 5034.937587022781 |
| Epoch 1 | Batch 42999/57412 | Loss: 0.531 | time 5057.203242063522 |
| Epoch 1 | Batch 43199/57412 | Loss: 0.338 | time 5081.412537574768 |
| Epoch 1 | Batch 43399/57412 | Loss: 0.424 | time 5104.605773448944 |
| Epoch 1 | Batch 43599/57412 | Loss: 0.359 | time 5128.648938417435 |
| Epoch 1 | Batch 43799/57412 | Loss: 0.508 | time 5151.679401636124 |
| Epoch 1 | Batch 43999/57412 | Loss: 0.582 | time 5174.5223660469055 |
| Epoch 1 | Batch 44199/57412 | Loss: 0.566 | time 5197.078404188156 |
| Epoch 1 | Batch 44399/57412 | Loss: 0.416 | time 5220.862770795822 |
| Epoch 1 | Batch 44599/57412 | Loss: 0.441 | time 5244.67792224884 |
| Epoch 1 | Batch 44799/57412 | Loss: 0.375 | time 5267.066955566406 |
| Epoch 1 | Batch 44999/57412 | Loss: 0.346 | time 5289.382257938385 |
| Epoch 1 | Batch 45199/57412 | Loss: 0.484 | time 5312.637675523758 |
| Epoch 1 | Batch 45399/57412 | Loss: 0.416 | time 5335.727750301361 |
| Epoch 1 | Batch 45599/57412 | Loss: 0.473 | time 5359.6941039562225 |
| Epoch 1 | Batch 45799/57412 | Loss: 0.297 | time 5381.962297677994 |
| Epoch 1 | Batch 45999/57412 | Loss: 0.230 | time 5404.686697483063 |
| Epoch 1 | Batch 46199/57412 | Loss: 0.508 | time 5427.352993488312 |
| Epoch 1 | Batch 46399/57412 | Loss: 0.207 | time 5449.468719959259 |
| Epoch 1 | Batch 46599/57412 | Loss: 0.316 | time 5475.133266210556 |
| Epoch 1 | Batch 46799/57412 | Loss: 0.684 | time 5499.214484214783 |
| Epoch 1 | Batch 46999/57412 | Loss: 0.373 | time 5523.843861341476 |
| Epoch 1 | Batch 47199/57412 | Loss: 0.303 | time 5547.734761238098 |
| Epoch 1 | Batch 47399/57412 | Loss: 0.398 | time 5570.189453840256 |
| Epoch 1 | Batch 47599/57412 | Loss: 0.375 | time 5593.616540670395 |
| Epoch 1 | Batch 47799/57412 | Loss: 0.342 | time 5616.553333044052 |
| Epoch 1 | Batch 47999/57412 | Loss: 0.193 | time 5640.959502220154 |
| Epoch 1 | Batch 48199/57412 | Loss: 0.645 | time 5665.642489910126 |
| Epoch 1 | Batch 48399/57412 | Loss: 0.307 | time 5687.8356647491455 |
| Epoch 1 | Batch 48599/57412 | Loss: 0.373 | time 5711.604918003082 |
| Epoch 1 | Batch 48799/57412 | Loss: 0.520 | time 5733.310106992722 |
| Epoch 1 | Batch 48999/57412 | Loss: 0.504 | time 5756.34495973587 |
| Epoch 1 | Batch 49199/57412 | Loss: 0.357 | time 5779.4891402721405 |
| Epoch 1 | Batch 49399/57412 | Loss: 0.381 | time 5802.382716655731 |
| Epoch 1 | Batch 49599/57412 | Loss: 0.414 | time 5826.269940137863 |
| Epoch 1 | Batch 49799/57412 | Loss: 0.316 | time 5849.472201824188 |
| Epoch 1 | Batch 49999/57412 | Loss: 0.410 | time 5872.50378036499 |
| Epoch 1 | Batch 50199/57412 | Loss: 0.312 | time 5895.9286522865295 |
| Epoch 1 | Batch 50399/57412 | Loss: 0.410 | time 5919.67635679245 |
| Epoch 1 | Batch 50599/57412 | Loss: 0.555 | time 5942.518639802933 |
| Epoch 1 | Batch 50799/57412 | Loss: 0.336 | time 5966.007194519043 |
| Epoch 1 | Batch 50999/57412 | Loss: 0.570 | time 5990.191049575806 |
| Epoch 1 | Batch 51199/57412 | Loss: 0.555 | time 6012.996586799622 |
| Epoch 1 | Batch 51399/57412 | Loss: 0.324 | time 6037.448735237122 |
| Epoch 1 | Batch 51599/57412 | Loss: 0.393 | time 6060.475448608398 |
| Epoch 1 | Batch 51799/57412 | Loss: 0.449 | time 6083.577203512192 |
| Epoch 1 | Batch 51999/57412 | Loss: 0.586 | time 6106.79761838913 |
| Epoch 1 | Batch 52199/57412 | Loss: 0.408 | time 6132.520553350449 |
| Epoch 1 | Batch 52399/57412 | Loss: 0.383 | time 6157.13490819931 |
| Epoch 1 | Batch 52599/57412 | Loss: 0.617 | time 6182.054984092712 |
| Epoch 1 | Batch 52799/57412 | Loss: 0.168 | time 6207.045657634735 |
| Epoch 1 | Batch 52999/57412 | Loss: 0.805 | time 6230.86003112793 |
| Epoch 1 | Batch 53199/57412 | Loss: 0.449 | time 6254.339380979538 |
| Epoch 1 | Batch 53399/57412 | Loss: 0.127 | time 6278.489122390747 |
| Epoch 1 | Batch 53599/57412 | Loss: 0.258 | time 6302.4866461753845 |
| Epoch 1 | Batch 53799/57412 | Loss: 0.369 | time 6327.1323437690735 |
| Epoch 1 | Batch 53999/57412 | Loss: 0.221 | time 6352.959258794785 |
| Epoch 1 | Batch 54199/57412 | Loss: 0.512 | time 6378.603352069855 |
| Epoch 1 | Batch 54399/57412 | Loss: 0.453 | time 6403.514925479889 |
| Epoch 1 | Batch 54599/57412 | Loss: 0.438 | time 6428.30436372757 |
| Epoch 1 | Batch 54799/57412 | Loss: 0.412 | time 6451.362206697464 |
| Epoch 1 | Batch 54999/57412 | Loss: 0.629 | time 6474.880658864975 |
| Epoch 1 | Batch 55199/57412 | Loss: 0.352 | time 6498.482791423798 |
| Epoch 1 | Batch 55399/57412 | Loss: 0.432 | time 6521.272460699081 |
| Epoch 1 | Batch 55599/57412 | Loss: 0.414 | time 6544.7295434474945 |
| Epoch 1 | Batch 55799/57412 | Loss: 0.543 | time 6568.30468416214 |
| Epoch 1 | Batch 55999/57412 | Loss: 0.081 | time 6590.748111009598 |
| Epoch 1 | Batch 56199/57412 | Loss: 0.283 | time 6614.173028469086 |
| Epoch 1 | Batch 56399/57412 | Loss: 0.118 | time 6636.271471500397 |
| Epoch 1 | Batch 56599/57412 | Loss: 0.326 | time 6661.270175218582 |
| Epoch 1 | Batch 56799/57412 | Loss: 0.455 | time 6682.712189674377 |
| Epoch 1 | Batch 56999/57412 | Loss: 0.166 | time 6706.200773000717 |
| Epoch 1 | Batch 57199/57412 | Loss: 0.385 | time 6728.867811203003 |
| Epoch 1 | Batch 57399/57412 | Loss: 0.260 | time 6750.7280950546265 |
| ========================================================================================= |
| ========================================================================================= |
| ========================================================================================= |
| Epoch 1 | Validation Samples 5082 | Validation Acc: 0.4777646595828414 | time 314.1923449039459 |
| ========================================================================================= |
| Saving at Epoch 1 | Learning rate: 2e-05 |
| Epoch 2 | Batch 199/57412 | Loss: 0.410 | time 25.10978603363037 |
| Epoch 2 | Batch 399/57412 | Loss: 0.264 | time 49.9205482006073 |
| Epoch 2 | Batch 599/57412 | Loss: 0.408 | time 74.27282571792603 |
| Epoch 2 | Batch 799/57412 | Loss: 0.354 | time 98.18675255775452 |
| Epoch 2 | Batch 999/57412 | Loss: 0.480 | time 121.70753908157349 |
| Epoch 2 | Batch 1199/57412 | Loss: 0.473 | time 145.3441412448883 |
| Epoch 2 | Batch 1399/57412 | Loss: 0.625 | time 170.97456431388855 |
| Epoch 2 | Batch 1599/57412 | Loss: 0.496 | time 196.33842635154724 |
| Epoch 2 | Batch 1799/57412 | Loss: 0.445 | time 219.9434518814087 |
| Epoch 2 | Batch 1999/57412 | Loss: 0.165 | time 242.38225722312927 |
| Epoch 2 | Batch 2199/57412 | Loss: 0.218 | time 263.9837899208069 |
| Epoch 2 | Batch 2399/57412 | Loss: 0.727 | time 285.31724762916565 |
| Epoch 2 | Batch 2599/57412 | Loss: 0.385 | time 309.30340218544006 |
| Epoch 2 | Batch 2799/57412 | Loss: 0.359 | time 332.21145248413086 |
| Epoch 2 | Batch 2999/57412 | Loss: 0.463 | time 355.57630920410156 |
| Epoch 2 | Batch 3199/57412 | Loss: 0.555 | time 378.2579321861267 |
| Epoch 2 | Batch 3399/57412 | Loss: 0.348 | time 401.59544920921326 |
| Epoch 2 | Batch 3599/57412 | Loss: 0.434 | time 424.5635118484497 |
| Epoch 2 | Batch 3799/57412 | Loss: 0.480 | time 448.03302097320557 |
| Epoch 2 | Batch 3999/57412 | Loss: 0.224 | time 470.93274903297424 |
| Epoch 2 | Batch 4199/57412 | Loss: 0.512 | time 492.37237644195557 |
| Epoch 2 | Batch 4399/57412 | Loss: 0.652 | time 515.174569606781 |
| Epoch 2 | Batch 4599/57412 | Loss: 0.268 | time 537.9686734676361 |
| Epoch 2 | Batch 4799/57412 | Loss: 0.379 | time 559.7228028774261 |
| Epoch 2 | Batch 4999/57412 | Loss: 0.516 | time 581.2522308826447 |
| Epoch 2 | Batch 5199/57412 | Loss: 0.469 | time 604.0912749767303 |
| Epoch 2 | Batch 5399/57412 | Loss: 0.352 | time 626.6006507873535 |
| Epoch 2 | Batch 5599/57412 | Loss: 0.436 | time 649.0032954216003 |
| Epoch 2 | Batch 5799/57412 | Loss: 0.379 | time 671.2869606018066 |
| Epoch 2 | Batch 5999/57412 | Loss: 0.531 | time 693.7598359584808 |
| Epoch 2 | Batch 6199/57412 | Loss: 0.436 | time 716.1393201351166 |
| Epoch 2 | Batch 6399/57412 | Loss: 0.277 | time 739.0895597934723 |
| Epoch 2 | Batch 6599/57412 | Loss: 0.316 | time 761.1982419490814 |
| Epoch 2 | Batch 6799/57412 | Loss: 0.453 | time 784.9556097984314 |
| Epoch 2 | Batch 6999/57412 | Loss: 0.488 | time 807.2372434139252 |
| Epoch 2 | Batch 7199/57412 | Loss: 0.324 | time 828.9791073799133 |
| Epoch 2 | Batch 7399/57412 | Loss: 0.373 | time 851.2383899688721 |
| Epoch 2 | Batch 7599/57412 | Loss: 0.570 | time 874.0372378826141 |
| Epoch 2 | Batch 7799/57412 | Loss: 0.004 | time 896.3143742084503 |
| Epoch 2 | Batch 7999/57412 | Loss: 0.613 | time 918.1195707321167 |
| Epoch 2 | Batch 8199/57412 | Loss: 0.219 | time 940.8352828025818 |
| Epoch 2 | Batch 8399/57412 | Loss: 0.492 | time 963.4791674613953 |
| Epoch 2 | Batch 8599/57412 | Loss: 0.277 | time 986.1106441020966 |
| Epoch 2 | Batch 8799/57412 | Loss: 0.602 | time 1010.6764185428619 |
| Epoch 2 | Batch 8999/57412 | Loss: 0.496 | time 1033.8592252731323 |
| Epoch 2 | Batch 9199/57412 | Loss: 0.645 | time 1056.9696829319 |
| Epoch 2 | Batch 9399/57412 | Loss: 0.494 | time 1080.3805720806122 |
| Epoch 2 | Batch 9599/57412 | Loss: 0.357 | time 1103.317348241806 |
| Epoch 2 | Batch 9799/57412 | Loss: 0.092 | time 1125.967008113861 |
| Epoch 2 | Batch 9999/57412 | Loss: 0.512 | time 1149.2071924209595 |
| Epoch 2 | Batch 10199/57412 | Loss: 0.215 | time 1171.3369743824005 |
| Epoch 2 | Batch 10399/57412 | Loss: 0.559 | time 1194.202832698822 |
| Epoch 2 | Batch 10599/57412 | Loss: 0.375 | time 1215.7031075954437 |
| Epoch 2 | Batch 10799/57412 | Loss: 0.238 | time 1238.4381861686707 |
| Epoch 2 | Batch 10999/57412 | Loss: 0.238 | time 1259.8318586349487 |
| Epoch 2 | Batch 11199/57412 | Loss: 0.271 | time 1282.4603779315948 |
| Epoch 2 | Batch 11399/57412 | Loss: 0.574 | time 1304.6375658512115 |
| Epoch 2 | Batch 11599/57412 | Loss: 0.352 | time 1329.3990495204926 |
| Epoch 2 | Batch 11799/57412 | Loss: 0.361 | time 1350.8695483207703 |
| Epoch 2 | Batch 11999/57412 | Loss: 0.465 | time 1373.7973990440369 |
| Epoch 2 | Batch 12199/57412 | Loss: 0.492 | time 1396.4786920547485 |
| Epoch 2 | Batch 12399/57412 | Loss: 0.430 | time 1418.9511632919312 |
| Epoch 2 | Batch 12599/57412 | Loss: 0.523 | time 1442.2929673194885 |
| Epoch 2 | Batch 12799/57412 | Loss: 0.225 | time 1464.5365393161774 |
| Epoch 2 | Batch 12999/57412 | Loss: 0.570 | time 1487.3031883239746 |
| Epoch 2 | Batch 13199/57412 | Loss: 0.137 | time 1510.3828065395355 |
| Epoch 2 | Batch 13399/57412 | Loss: 0.475 | time 1534.0947768688202 |
| Epoch 2 | Batch 13599/57412 | Loss: 0.270 | time 1556.4195022583008 |
| Epoch 2 | Batch 13799/57412 | Loss: 0.424 | time 1579.06209897995 |
| Epoch 2 | Batch 13999/57412 | Loss: 0.256 | time 1600.9756112098694 |
| Epoch 2 | Batch 14199/57412 | Loss: 0.465 | time 1624.169101715088 |
| Epoch 2 | Batch 14399/57412 | Loss: 0.211 | time 1647.0683271884918 |
| Epoch 2 | Batch 14599/57412 | Loss: 0.441 | time 1670.6229491233826 |
| Epoch 2 | Batch 14799/57412 | Loss: 0.344 | time 1693.4909317493439 |
| Epoch 2 | Batch 14999/57412 | Loss: 0.193 | time 1716.2110285758972 |
| Epoch 2 | Batch 15199/57412 | Loss: 0.156 | time 1739.0344512462616 |
| Epoch 2 | Batch 15399/57412 | Loss: 0.449 | time 1763.2728562355042 |
| Epoch 2 | Batch 15599/57412 | Loss: 0.391 | time 1787.4648170471191 |
| Epoch 2 | Batch 15799/57412 | Loss: 0.371 | time 1810.6648466587067 |
| Epoch 2 | Batch 15999/57412 | Loss: 0.314 | time 1832.2875218391418 |
| Epoch 2 | Batch 16199/57412 | Loss: 0.148 | time 1855.4082159996033 |
| Epoch 2 | Batch 16399/57412 | Loss: 0.498 | time 1878.1308426856995 |
| Epoch 2 | Batch 16599/57412 | Loss: 0.379 | time 1901.6811690330505 |
| Epoch 2 | Batch 16799/57412 | Loss: 0.182 | time 1923.678453207016 |
| Epoch 2 | Batch 16999/57412 | Loss: 0.473 | time 1946.4839036464691 |
| Epoch 2 | Batch 17199/57412 | Loss: 0.355 | time 1968.3315880298615 |
| Epoch 2 | Batch 17399/57412 | Loss: 0.168 | time 1990.9245550632477 |
| Epoch 2 | Batch 17599/57412 | Loss: 0.375 | time 2013.7534413337708 |
| Epoch 2 | Batch 17799/57412 | Loss: 0.091 | time 2036.284755706787 |
| Epoch 2 | Batch 17999/57412 | Loss: 0.379 | time 2057.6548161506653 |
| Epoch 2 | Batch 18199/57412 | Loss: 0.324 | time 2080.0111486911774 |
| Epoch 2 | Batch 18399/57412 | Loss: 0.107 | time 2105.399399280548 |
| Epoch 2 | Batch 18599/57412 | Loss: 0.395 | time 2128.4646697044373 |
| Epoch 2 | Batch 18799/57412 | Loss: 0.434 | time 2151.819466114044 |
| Epoch 2 | Batch 18999/57412 | Loss: 0.520 | time 2174.3301548957825 |
| Epoch 2 | Batch 19199/57412 | Loss: 0.455 | time 2196.2520496845245 |
| Epoch 2 | Batch 19399/57412 | Loss: 0.196 | time 2219.553879737854 |
| Epoch 2 | Batch 19599/57412 | Loss: 0.354 | time 2243.2194409370422 |
| Epoch 2 | Batch 19799/57412 | Loss: 0.040 | time 2266.81244969368 |
| Epoch 2 | Batch 19999/57412 | Loss: 0.520 | time 2288.5517115592957 |
| Saving at Step 20000 |
| Epoch 2 | Batch 20199/57412 | Loss: 0.249 | time 2313.6617205142975 |
| Epoch 2 | Batch 20399/57412 | Loss: 0.279 | time 2336.139302253723 |
| Epoch 2 | Batch 20599/57412 | Loss: 0.107 | time 2357.8166830539703 |
| Epoch 2 | Batch 20799/57412 | Loss: 0.332 | time 2380.5488991737366 |
| Epoch 2 | Batch 20999/57412 | Loss: 0.414 | time 2404.347787618637 |
| Epoch 2 | Batch 21199/57412 | Loss: 0.613 | time 2427.648470878601 |
| Epoch 2 | Batch 21399/57412 | Loss: 0.252 | time 2449.219306707382 |
| Epoch 2 | Batch 21599/57412 | Loss: 0.566 | time 2470.9717948436737 |
| Epoch 2 | Batch 21799/57412 | Loss: 0.181 | time 2492.8275434970856 |
| Epoch 2 | Batch 21999/57412 | Loss: 0.367 | time 2515.765342235565 |
| Epoch 2 | Batch 22199/57412 | Loss: 0.410 | time 2537.3223514556885 |
| Epoch 2 | Batch 22399/57412 | Loss: 0.432 | time 2559.768495798111 |
| Epoch 2 | Batch 22599/57412 | Loss: 0.064 | time 2582.700060606003 |
| Epoch 2 | Batch 22799/57412 | Loss: 0.328 | time 2604.8618960380554 |
| Epoch 2 | Batch 22999/57412 | Loss: 0.527 | time 2628.0654768943787 |
| Epoch 2 | Batch 23199/57412 | Loss: 0.252 | time 2651.044773578644 |
| Epoch 2 | Batch 23399/57412 | Loss: 0.305 | time 2673.881287574768 |
| Epoch 2 | Batch 23599/57412 | Loss: 0.309 | time 2695.948484659195 |
| Epoch 2 | Batch 23799/57412 | Loss: 0.191 | time 2718.6102883815765 |
| Epoch 2 | Batch 23999/57412 | Loss: 0.396 | time 2740.3380348682404 |
| Epoch 2 | Batch 24199/57412 | Loss: 0.318 | time 2763.1808202266693 |
| Epoch 2 | Batch 24399/57412 | Loss: 0.535 | time 2786.5847551822662 |
| Epoch 2 | Batch 24599/57412 | Loss: 0.365 | time 2810.509800195694 |
| Epoch 2 | Batch 24799/57412 | Loss: 0.436 | time 2833.150787115097 |
| Epoch 2 | Batch 24999/57412 | Loss: 0.245 | time 2856.082792043686 |
| Epoch 2 | Batch 25199/57412 | Loss: 0.181 | time 2879.025512933731 |
| Epoch 2 | Batch 25399/57412 | Loss: 0.422 | time 2901.3419029712677 |
| Epoch 2 | Batch 25599/57412 | Loss: 0.332 | time 2924.6528985500336 |
| Epoch 2 | Batch 25799/57412 | Loss: 0.594 | time 2950.171401500702 |
| Epoch 2 | Batch 25999/57412 | Loss: 0.124 | time 2974.61799287796 |
| Epoch 2 | Batch 26199/57412 | Loss: 0.328 | time 2998.707906961441 |
| Epoch 2 | Batch 26399/57412 | Loss: 0.328 | time 3021.016126394272 |
| Epoch 2 | Batch 26599/57412 | Loss: 0.414 | time 3043.572729110718 |
| Epoch 2 | Batch 26799/57412 | Loss: 0.400 | time 3067.6549203395844 |
| Epoch 2 | Batch 26999/57412 | Loss: 0.459 | time 3092.0263273715973 |
| Epoch 2 | Batch 27199/57412 | Loss: 0.428 | time 3117.0151550769806 |
| Epoch 2 | Batch 27399/57412 | Loss: 0.500 | time 3141.061578273773 |
| Epoch 2 | Batch 27599/57412 | Loss: 0.475 | time 3164.730370283127 |
| Epoch 2 | Batch 27799/57412 | Loss: 0.303 | time 3188.4495017528534 |
| Epoch 2 | Batch 27999/57412 | Loss: 0.668 | time 3212.7473578453064 |
| Epoch 2 | Batch 28199/57412 | Loss: 0.494 | time 3237.1793529987335 |
| Epoch 2 | Batch 28399/57412 | Loss: 0.400 | time 3261.8031692504883 |
| Epoch 2 | Batch 28599/57412 | Loss: 0.328 | time 3285.8855328559875 |
| Epoch 2 | Batch 28799/57412 | Loss: 0.535 | time 3308.4772715568542 |
| Epoch 2 | Batch 28999/57412 | Loss: 0.338 | time 3332.6162254810333 |
| Epoch 2 | Batch 29199/57412 | Loss: 0.379 | time 3357.57351398468 |
| Epoch 2 | Batch 29399/57412 | Loss: 0.285 | time 3381.381025314331 |
| Epoch 2 | Batch 29599/57412 | Loss: 0.393 | time 3405.625486135483 |
| Epoch 2 | Batch 29799/57412 | Loss: 0.447 | time 3428.3546895980835 |
| Epoch 2 | Batch 29999/57412 | Loss: 0.574 | time 3450.638384580612 |
| Epoch 2 | Batch 30199/57412 | Loss: 0.445 | time 3472.709751844406 |
| Epoch 2 | Batch 30399/57412 | Loss: 0.297 | time 3497.1173243522644 |
| Epoch 2 | Batch 30599/57412 | Loss: 0.250 | time 3521.2063229084015 |
| Epoch 2 | Batch 30799/57412 | Loss: 0.484 | time 3543.8339908123016 |
| Epoch 2 | Batch 30999/57412 | Loss: 0.328 | time 3567.426477432251 |
| Epoch 2 | Batch 31199/57412 | Loss: 0.041 | time 3590.712532043457 |
| Epoch 2 | Batch 31399/57412 | Loss: 0.061 | time 3614.3002965450287 |
| Epoch 2 | Batch 31599/57412 | Loss: 0.609 | time 3637.3593530654907 |
| Epoch 2 | Batch 31799/57412 | Loss: 0.305 | time 3660.7599546909332 |
| Epoch 2 | Batch 31999/57412 | Loss: 0.512 | time 3684.895528078079 |
| Epoch 2 | Batch 32199/57412 | Loss: 0.652 | time 3707.8200163841248 |
| Epoch 2 | Batch 32399/57412 | Loss: 0.222 | time 3730.7500591278076 |
| Epoch 2 | Batch 32599/57412 | Loss: 0.477 | time 3752.917376279831 |
| Epoch 2 | Batch 32799/57412 | Loss: 0.574 | time 3776.6839497089386 |
| Epoch 2 | Batch 32999/57412 | Loss: 0.637 | time 3798.495386123657 |
| Epoch 2 | Batch 33199/57412 | Loss: 0.500 | time 3821.437020778656 |
| Epoch 2 | Batch 33399/57412 | Loss: 0.166 | time 3844.141665458679 |
| Epoch 2 | Batch 33599/57412 | Loss: 0.233 | time 3866.5866253376007 |
| Epoch 2 | Batch 33799/57412 | Loss: 0.238 | time 3889.288573026657 |
| Epoch 2 | Batch 33999/57412 | Loss: 0.424 | time 3912.606156349182 |
| Epoch 2 | Batch 34199/57412 | Loss: 0.406 | time 3935.590599298477 |
| Epoch 2 | Batch 34399/57412 | Loss: 0.340 | time 3959.33886885643 |
| Epoch 2 | Batch 34599/57412 | Loss: 0.504 | time 3983.955731153488 |
| Epoch 2 | Batch 34799/57412 | Loss: 0.426 | time 4006.5982813835144 |
| Epoch 2 | Batch 34999/57412 | Loss: 0.707 | time 4030.7396845817566 |
| Epoch 2 | Batch 35199/57412 | Loss: 0.369 | time 4053.5580010414124 |
| Epoch 2 | Batch 35399/57412 | Loss: 0.254 | time 4077.1060609817505 |
| Epoch 2 | Batch 35599/57412 | Loss: 0.338 | time 4099.297807216644 |
| Epoch 2 | Batch 35799/57412 | Loss: 0.516 | time 4122.085989236832 |
| Epoch 2 | Batch 35999/57412 | Loss: 0.494 | time 4144.537331581116 |
| Epoch 2 | Batch 36199/57412 | Loss: 0.087 | time 4167.559956789017 |
| Epoch 2 | Batch 36399/57412 | Loss: 0.330 | time 4190.902949810028 |
| Epoch 2 | Batch 36599/57412 | Loss: 0.249 | time 4213.433697938919 |
| Epoch 2 | Batch 36799/57412 | Loss: 0.424 | time 4236.734105825424 |
| Epoch 2 | Batch 36999/57412 | Loss: 0.128 | time 4260.380426168442 |
| Epoch 2 | Batch 37199/57412 | Loss: 0.467 | time 4283.446613788605 |
| Epoch 2 | Batch 37399/57412 | Loss: 0.256 | time 4307.466418027878 |
| Epoch 2 | Batch 37599/57412 | Loss: 0.418 | time 4329.868490457535 |
| Epoch 2 | Batch 37799/57412 | Loss: 0.371 | time 4352.520091056824 |
| Epoch 2 | Batch 37999/57412 | Loss: 0.357 | time 4375.242646932602 |
| Epoch 2 | Batch 38199/57412 | Loss: 0.002 | time 4399.861700534821 |
| Epoch 2 | Batch 38399/57412 | Loss: 0.275 | time 4422.341259479523 |
| Epoch 2 | Batch 38599/57412 | Loss: 0.225 | time 4445.076668739319 |
| Epoch 2 | Batch 38799/57412 | Loss: 0.621 | time 4468.417022705078 |
| Epoch 2 | Batch 38999/57412 | Loss: 0.500 | time 4492.831372976303 |
| Epoch 2 | Batch 39199/57412 | Loss: 0.543 | time 4515.621602535248 |
| Epoch 2 | Batch 39399/57412 | Loss: 0.391 | time 4539.5258066654205 |
| Epoch 2 | Batch 39599/57412 | Loss: 0.711 | time 4563.298105478287 |
| Epoch 2 | Batch 39799/57412 | Loss: 0.494 | time 4587.32234287262 |
| Epoch 2 | Batch 39999/57412 | Loss: 0.594 | time 4610.971751689911 |
| Saving at Step 40000 |
| Epoch 2 | Batch 40199/57412 | Loss: 0.001 | time 4638.081170558929 |
| Epoch 2 | Batch 40399/57412 | Loss: 0.664 | time 4660.8843195438385 |
| Epoch 2 | Batch 40599/57412 | Loss: 0.336 | time 4684.940701723099 |
| Epoch 2 | Batch 40799/57412 | Loss: 0.297 | time 4707.369230747223 |
| Epoch 2 | Batch 40999/57412 | Loss: 0.270 | time 4731.560967683792 |
| Epoch 2 | Batch 41199/57412 | Loss: 0.299 | time 4754.550015449524 |
| Epoch 2 | Batch 41399/57412 | Loss: 0.354 | time 4779.061139822006 |
| Epoch 2 | Batch 41599/57412 | Loss: 0.395 | time 4801.294118881226 |
| Epoch 2 | Batch 41799/57412 | Loss: 0.183 | time 4825.319691896439 |
| Epoch 2 | Batch 41999/57412 | Loss: 0.373 | time 4847.817550897598 |
| Epoch 2 | Batch 42199/57412 | Loss: 0.270 | time 4873.001101016998 |
| Epoch 2 | Batch 42399/57412 | Loss: 0.279 | time 4896.492738485336 |
| Epoch 2 | Batch 42599/57412 | Loss: 0.531 | time 4919.595214605331 |
| Epoch 2 | Batch 42799/57412 | Loss: 0.346 | time 4942.816931724548 |
| Epoch 2 | Batch 42999/57412 | Loss: 0.438 | time 4964.944697141647 |
| Epoch 2 | Batch 43199/57412 | Loss: 0.318 | time 4989.274143695831 |
| Epoch 2 | Batch 43399/57412 | Loss: 0.416 | time 5012.336147785187 |
| Epoch 2 | Batch 43599/57412 | Loss: 0.363 | time 5036.4196128845215 |
| Epoch 2 | Batch 43799/57412 | Loss: 0.469 | time 5059.609100580215 |
| Epoch 2 | Batch 43999/57412 | Loss: 0.566 | time 5082.354423761368 |
| Epoch 2 | Batch 44199/57412 | Loss: 0.523 | time 5104.871793270111 |
| Epoch 2 | Batch 44399/57412 | Loss: 0.398 | time 5128.652804374695 |
| Epoch 2 | Batch 44599/57412 | Loss: 0.432 | time 5152.467030286789 |
| Epoch 2 | Batch 44799/57412 | Loss: 0.377 | time 5174.946868896484 |
| Epoch 2 | Batch 44999/57412 | Loss: 0.254 | time 5196.995698928833 |
| Epoch 2 | Batch 45199/57412 | Loss: 0.471 | time 5220.291134357452 |
| Epoch 2 | Batch 45399/57412 | Loss: 0.387 | time 5243.428669691086 |
| Epoch 2 | Batch 45599/57412 | Loss: 0.463 | time 5267.171909093857 |
| Epoch 2 | Batch 45799/57412 | Loss: 0.295 | time 5289.304484844208 |
| Epoch 2 | Batch 45999/57412 | Loss: 0.105 | time 5311.815444469452 |
| Epoch 2 | Batch 46199/57412 | Loss: 0.486 | time 5334.325279474258 |
| Epoch 2 | Batch 46399/57412 | Loss: 0.194 | time 5356.0518844127655 |
| Epoch 2 | Batch 46599/57412 | Loss: 0.312 | time 5380.230415582657 |
| Epoch 2 | Batch 46799/57412 | Loss: 0.652 | time 5403.154785394669 |
| Epoch 2 | Batch 46999/57412 | Loss: 0.367 | time 5426.8636338710785 |
| Epoch 2 | Batch 47199/57412 | Loss: 0.293 | time 5449.597641706467 |
| Epoch 2 | Batch 47399/57412 | Loss: 0.359 | time 5471.102988243103 |
| Epoch 2 | Batch 47599/57412 | Loss: 0.352 | time 5494.292503595352 |
| Epoch 2 | Batch 47799/57412 | Loss: 0.320 | time 5516.849585771561 |
| Epoch 2 | Batch 47999/57412 | Loss: 0.184 | time 5540.875703573227 |
| Epoch 2 | Batch 48199/57412 | Loss: 0.625 | time 5565.141613006592 |
| Epoch 2 | Batch 48399/57412 | Loss: 0.285 | time 5587.155515193939 |
| Epoch 2 | Batch 48599/57412 | Loss: 0.291 | time 5610.765896320343 |
| Epoch 2 | Batch 48799/57412 | Loss: 0.484 | time 5632.349676847458 |
| Epoch 2 | Batch 48999/57412 | Loss: 0.479 | time 5655.129397153854 |
| Epoch 2 | Batch 49199/57412 | Loss: 0.344 | time 5678.054527044296 |
| Epoch 2 | Batch 49399/57412 | Loss: 0.375 | time 5700.85922908783 |
| Epoch 2 | Batch 49599/57412 | Loss: 0.383 | time 5724.587399721146 |
| Epoch 2 | Batch 49799/57412 | Loss: 0.277 | time 5747.488655328751 |
| Epoch 2 | Batch 49999/57412 | Loss: 0.371 | time 5770.610054969788 |
| Epoch 2 | Batch 50199/57412 | Loss: 0.216 | time 5794.123499631882 |
| Epoch 2 | Batch 50399/57412 | Loss: 0.369 | time 5817.933428287506 |
| Epoch 2 | Batch 50599/57412 | Loss: 0.477 | time 5840.712598562241 |
| Epoch 2 | Batch 50799/57412 | Loss: 0.326 | time 5864.209442853928 |
| Epoch 2 | Batch 50999/57412 | Loss: 0.551 | time 5888.439728021622 |
| Epoch 2 | Batch 51199/57412 | Loss: 0.531 | time 5911.218539476395 |
| Epoch 2 | Batch 51399/57412 | Loss: 0.316 | time 5935.488235235214 |
| Epoch 2 | Batch 51599/57412 | Loss: 0.365 | time 5958.713482141495 |
| Epoch 2 | Batch 51799/57412 | Loss: 0.422 | time 5981.81529545784 |
| Epoch 2 | Batch 51999/57412 | Loss: 0.566 | time 6005.232220172882 |
| Epoch 2 | Batch 52199/57412 | Loss: 0.393 | time 6030.000046014786 |
| Epoch 2 | Batch 52399/57412 | Loss: 0.355 | time 6053.533740758896 |
| Epoch 2 | Batch 52599/57412 | Loss: 0.586 | time 6077.373037099838 |
| Epoch 2 | Batch 52799/57412 | Loss: 0.162 | time 6101.011016368866 |
| Epoch 2 | Batch 52999/57412 | Loss: 0.668 | time 6123.409858942032 |
| Epoch 2 | Batch 53199/57412 | Loss: 0.441 | time 6145.688636541367 |
| Epoch 2 | Batch 53399/57412 | Loss: 0.115 | time 6168.565146923065 |
| Epoch 2 | Batch 53599/57412 | Loss: 0.258 | time 6191.6989533901215 |
| Epoch 2 | Batch 53799/57412 | Loss: 0.346 | time 6215.399564743042 |
| Epoch 2 | Batch 53999/57412 | Loss: 0.217 | time 6240.238403558731 |
| Epoch 2 | Batch 54199/57412 | Loss: 0.420 | time 6264.911828756332 |
| Epoch 2 | Batch 54399/57412 | Loss: 0.396 | time 6288.83554649353 |
| Epoch 2 | Batch 54599/57412 | Loss: 0.404 | time 6312.363488674164 |
| Epoch 2 | Batch 54799/57412 | Loss: 0.404 | time 6334.690718889236 |
| Epoch 2 | Batch 54999/57412 | Loss: 0.605 | time 6358.399973630905 |
| Epoch 2 | Batch 55199/57412 | Loss: 0.346 | time 6382.545773983002 |
| Epoch 2 | Batch 55399/57412 | Loss: 0.418 | time 6405.605789422989 |
| Epoch 2 | Batch 55599/57412 | Loss: 0.412 | time 6429.3501398563385 |
| Epoch 2 | Batch 55799/57412 | Loss: 0.531 | time 6453.125873327255 |
| Epoch 2 | Batch 55999/57412 | Loss: 0.081 | time 6475.908611059189 |
| Epoch 2 | Batch 56199/57412 | Loss: 0.271 | time 6498.940671682358 |
| Epoch 2 | Batch 56399/57412 | Loss: 0.117 | time 6521.474977731705 |
| Epoch 2 | Batch 56599/57412 | Loss: 0.322 | time 6546.66610121727 |
| Epoch 2 | Batch 56799/57412 | Loss: 0.447 | time 6568.972702264786 |
| Epoch 2 | Batch 56999/57412 | Loss: 0.163 | time 6592.895660638809 |
| Epoch 2 | Batch 57199/57412 | Loss: 0.379 | time 6616.392576932907 |
| Epoch 2 | Batch 57399/57412 | Loss: 0.245 | time 6638.749967336655 |
| ========================================================================================= |
| ========================================================================================= |
| ========================================================================================= |
| Epoch 2 | Validation Samples 5082 | Validation Acc: 0.5419126328217237 | time 333.93119978904724 |
| ========================================================================================= |
| Saving at Epoch 2 | Learning rate: 2e-05 |
| Epoch 3 | Batch 199/57412 | Loss: 0.383 | time 25.277272939682007 |
| Epoch 3 | Batch 399/57412 | Loss: 0.252 | time 49.86625528335571 |
| Epoch 3 | Batch 599/57412 | Loss: 0.395 | time 74.14237475395203 |
| Epoch 3 | Batch 799/57412 | Loss: 0.336 | time 97.88849973678589 |
| Epoch 3 | Batch 999/57412 | Loss: 0.461 | time 121.62149500846863 |
| Epoch 3 | Batch 1199/57412 | Loss: 0.461 | time 145.37583589553833 |
| Epoch 3 | Batch 1399/57412 | Loss: 0.594 | time 171.3168385028839 |
| Epoch 3 | Batch 1599/57412 | Loss: 0.461 | time 196.68726778030396 |
| Epoch 3 | Batch 1799/57412 | Loss: 0.410 | time 220.20061326026917 |
| Epoch 3 | Batch 1999/57412 | Loss: 0.117 | time 244.0197470188141 |
| Epoch 3 | Batch 2199/57412 | Loss: 0.204 | time 267.1690504550934 |
| Epoch 3 | Batch 2399/57412 | Loss: 0.715 | time 290.0810091495514 |
| Epoch 3 | Batch 2599/57412 | Loss: 0.375 | time 315.5562915802002 |
| Epoch 3 | Batch 2799/57412 | Loss: 0.361 | time 339.35110545158386 |
| Epoch 3 | Batch 2999/57412 | Loss: 0.412 | time 363.8720271587372 |
| Epoch 3 | Batch 3199/57412 | Loss: 0.543 | time 387.8257441520691 |
| Epoch 3 | Batch 3399/57412 | Loss: 0.344 | time 412.5532021522522 |
| Epoch 3 | Batch 3599/57412 | Loss: 0.426 | time 436.7203378677368 |
| Epoch 3 | Batch 3799/57412 | Loss: 0.469 | time 461.53082036972046 |
| Epoch 3 | Batch 3999/57412 | Loss: 0.218 | time 485.7213635444641 |
| Epoch 3 | Batch 4199/57412 | Loss: 0.467 | time 507.8597102165222 |
| Epoch 3 | Batch 4399/57412 | Loss: 0.637 | time 531.2017953395844 |
| Epoch 3 | Batch 4599/57412 | Loss: 0.260 | time 554.3820123672485 |
| Epoch 3 | Batch 4799/57412 | Loss: 0.375 | time 576.6143131256104 |
| Epoch 3 | Batch 4999/57412 | Loss: 0.494 | time 598.6927344799042 |
| Epoch 3 | Batch 5199/57412 | Loss: 0.436 | time 621.634045124054 |
| Epoch 3 | Batch 5399/57412 | Loss: 0.344 | time 644.5868489742279 |
| Epoch 3 | Batch 5599/57412 | Loss: 0.432 | time 667.2422680854797 |
| Epoch 3 | Batch 5799/57412 | Loss: 0.373 | time 689.9459910392761 |
| Epoch 3 | Batch 5999/57412 | Loss: 0.516 | time 712.419367313385 |
| Epoch 3 | Batch 6199/57412 | Loss: 0.428 | time 734.7626419067383 |
| Epoch 3 | Batch 6399/57412 | Loss: 0.273 | time 757.8252711296082 |
| Epoch 3 | Batch 6599/57412 | Loss: 0.311 | time 780.2981004714966 |
| Epoch 3 | Batch 6799/57412 | Loss: 0.443 | time 804.1989946365356 |
| Epoch 3 | Batch 6999/57412 | Loss: 0.480 | time 826.8722033500671 |
| Epoch 3 | Batch 7199/57412 | Loss: 0.318 | time 848.7093367576599 |
| Epoch 3 | Batch 7399/57412 | Loss: 0.348 | time 871.3009538650513 |
| Epoch 3 | Batch 7599/57412 | Loss: 0.539 | time 894.2842743396759 |
| Epoch 3 | Batch 7799/57412 | Loss: 0.004 | time 916.9190440177917 |
| Epoch 3 | Batch 7999/57412 | Loss: 0.555 | time 938.9947102069855 |
| Epoch 3 | Batch 8199/57412 | Loss: 0.213 | time 962.0206677913666 |
| Epoch 3 | Batch 8399/57412 | Loss: 0.482 | time 984.9602456092834 |
| Epoch 3 | Batch 8599/57412 | Loss: 0.270 | time 1007.802003622055 |
| Epoch 3 | Batch 8799/57412 | Loss: 0.566 | time 1032.7655622959137 |
| Epoch 3 | Batch 8999/57412 | Loss: 0.465 | time 1056.5884191989899 |
| Epoch 3 | Batch 9199/57412 | Loss: 0.578 | time 1080.0328948497772 |
| Epoch 3 | Batch 9399/57412 | Loss: 0.488 | time 1103.8939979076385 |
| Epoch 3 | Batch 9599/57412 | Loss: 0.328 | time 1127.9906497001648 |
| Epoch 3 | Batch 9799/57412 | Loss: 0.092 | time 1152.1745674610138 |
| Epoch 3 | Batch 9999/57412 | Loss: 0.500 | time 1176.516916513443 |
| Epoch 3 | Batch 10199/57412 | Loss: 0.191 | time 1200.0734972953796 |
| Epoch 3 | Batch 10399/57412 | Loss: 0.555 | time 1224.2761061191559 |
| Epoch 3 | Batch 10599/57412 | Loss: 0.371 | time 1247.4301619529724 |
| Epoch 3 | Batch 10799/57412 | Loss: 0.143 | time 1271.528439283371 |
| Epoch 3 | Batch 10999/57412 | Loss: 0.235 | time 1294.2422986030579 |
| Epoch 3 | Batch 11199/57412 | Loss: 0.258 | time 1318.124127149582 |
| Epoch 3 | Batch 11399/57412 | Loss: 0.531 | time 1341.79252576828 |
| Epoch 3 | Batch 11599/57412 | Loss: 0.324 | time 1367.5343809127808 |
| Epoch 3 | Batch 11799/57412 | Loss: 0.340 | time 1390.5758616924286 |
| Epoch 3 | Batch 11999/57412 | Loss: 0.443 | time 1414.693240404129 |
| Epoch 3 | Batch 12199/57412 | Loss: 0.465 | time 1439.0083005428314 |
| Epoch 3 | Batch 12399/57412 | Loss: 0.312 | time 1463.0160355567932 |
| Epoch 3 | Batch 12599/57412 | Loss: 0.441 | time 1487.5764169692993 |
| Epoch 3 | Batch 12799/57412 | Loss: 0.224 | time 1511.2728443145752 |
| Epoch 3 | Batch 12999/57412 | Loss: 0.559 | time 1535.3029022216797 |
| Epoch 3 | Batch 13199/57412 | Loss: 0.081 | time 1559.3411936759949 |
| Epoch 3 | Batch 13399/57412 | Loss: 0.441 | time 1584.240803718567 |
| Epoch 3 | Batch 13599/57412 | Loss: 0.249 | time 1607.8973603248596 |
| Epoch 3 | Batch 13799/57412 | Loss: 0.422 | time 1632.2225122451782 |
| Epoch 3 | Batch 13999/57412 | Loss: 0.233 | time 1655.4203374385834 |
| Epoch 3 | Batch 14199/57412 | Loss: 0.447 | time 1680.1129055023193 |
| Epoch 3 | Batch 14399/57412 | Loss: 0.199 | time 1704.2171947956085 |
| Epoch 3 | Batch 14599/57412 | Loss: 0.426 | time 1728.9900794029236 |
| Epoch 3 | Batch 14799/57412 | Loss: 0.334 | time 1753.0778036117554 |
| Epoch 3 | Batch 14999/57412 | Loss: 0.193 | time 1776.5754594802856 |
| Epoch 3 | Batch 15199/57412 | Loss: 0.123 | time 1800.6564214229584 |
| Epoch 3 | Batch 15399/57412 | Loss: 0.430 | time 1826.3803355693817 |
| Epoch 3 | Batch 15599/57412 | Loss: 0.391 | time 1851.930449962616 |
| Epoch 3 | Batch 15799/57412 | Loss: 0.355 | time 1875.4536263942719 |
| Epoch 3 | Batch 15999/57412 | Loss: 0.307 | time 1897.4079689979553 |
| Epoch 3 | Batch 16199/57412 | Loss: 0.067 | time 1920.9279572963715 |
| Epoch 3 | Batch 16399/57412 | Loss: 0.475 | time 1944.150166273117 |
| Epoch 3 | Batch 16599/57412 | Loss: 0.359 | time 1967.7210488319397 |
| Epoch 3 | Batch 16799/57412 | Loss: 0.178 | time 1990.091307401657 |
| Epoch 3 | Batch 16999/57412 | Loss: 0.426 | time 2013.5648930072784 |
| Epoch 3 | Batch 17199/57412 | Loss: 0.326 | time 2036.1069629192352 |
| Epoch 3 | Batch 17399/57412 | Loss: 0.104 | time 2058.884939432144 |
| Epoch 3 | Batch 17599/57412 | Loss: 0.373 | time 2082.075231075287 |
| Epoch 3 | Batch 17799/57412 | Loss: 0.093 | time 2104.928823709488 |
| Epoch 3 | Batch 17999/57412 | Loss: 0.312 | time 2126.766769170761 |
| Epoch 3 | Batch 18199/57412 | Loss: 0.322 | time 2149.1274156570435 |
| Epoch 3 | Batch 18399/57412 | Loss: 0.097 | time 2174.8185696601868 |
| Epoch 3 | Batch 18599/57412 | Loss: 0.377 | time 2198.4072675704956 |
| Epoch 3 | Batch 18799/57412 | Loss: 0.424 | time 2221.927515268326 |
| Epoch 3 | Batch 18999/57412 | Loss: 0.445 | time 2244.8301482200623 |
| Epoch 3 | Batch 19199/57412 | Loss: 0.420 | time 2267.478867292404 |
| Epoch 3 | Batch 19399/57412 | Loss: 0.188 | time 2291.3176589012146 |
| Epoch 3 | Batch 19599/57412 | Loss: 0.352 | time 2315.28608250618 |
| Epoch 3 | Batch 19799/57412 | Loss: 0.019 | time 2339.2640614509583 |
| Epoch 3 | Batch 19999/57412 | Loss: 0.508 | time 2361.722441673279 |
| Saving at Step 20000 |
| Epoch 3 | Batch 20199/57412 | Loss: 0.232 | time 2387.6544086933136 |
| Epoch 3 | Batch 20399/57412 | Loss: 0.283 | time 2410.8213489055634 |
| Epoch 3 | Batch 20599/57412 | Loss: 0.071 | time 2433.1349880695343 |
| Epoch 3 | Batch 20799/57412 | Loss: 0.318 | time 2456.7159292697906 |
| Epoch 3 | Batch 20999/57412 | Loss: 0.383 | time 2480.80903172493 |
| Epoch 3 | Batch 21199/57412 | Loss: 0.574 | time 2504.0749270915985 |
| Epoch 3 | Batch 21399/57412 | Loss: 0.176 | time 2525.6576249599457 |
| Epoch 3 | Batch 21599/57412 | Loss: 0.551 | time 2547.7162144184113 |
| Epoch 3 | Batch 21799/57412 | Loss: 0.176 | time 2570.0862984657288 |
| Epoch 3 | Batch 21999/57412 | Loss: 0.346 | time 2593.2387568950653 |
| Epoch 3 | Batch 22199/57412 | Loss: 0.404 | time 2615.1904871463776 |
| Epoch 3 | Batch 22399/57412 | Loss: 0.418 | time 2638.01700091362 |
| Epoch 3 | Batch 22599/57412 | Loss: 0.065 | time 2661.4601576328278 |
| Epoch 3 | Batch 22799/57412 | Loss: 0.262 | time 2683.943788766861 |
| Epoch 3 | Batch 22999/57412 | Loss: 0.492 | time 2707.3082010746 |
| Epoch 3 | Batch 23199/57412 | Loss: 0.203 | time 2730.7327411174774 |
| Epoch 3 | Batch 23399/57412 | Loss: 0.287 | time 2754.4149417877197 |
| Epoch 3 | Batch 23599/57412 | Loss: 0.289 | time 2777.026580095291 |
| Epoch 3 | Batch 23799/57412 | Loss: 0.186 | time 2800.3435661792755 |
| Epoch 3 | Batch 23999/57412 | Loss: 0.328 | time 2822.6085591316223 |
| Epoch 3 | Batch 24199/57412 | Loss: 0.307 | time 2846.013195514679 |
| Epoch 3 | Batch 24399/57412 | Loss: 0.512 | time 2869.989272594452 |
| Epoch 3 | Batch 24599/57412 | Loss: 0.363 | time 2894.1936371326447 |
| Epoch 3 | Batch 24799/57412 | Loss: 0.422 | time 2916.9790909290314 |
| Epoch 3 | Batch 24999/57412 | Loss: 0.246 | time 2940.278367996216 |
| Epoch 3 | Batch 25199/57412 | Loss: 0.175 | time 2963.447712659836 |
| Epoch 3 | Batch 25399/57412 | Loss: 0.410 | time 2985.9524919986725 |
| Epoch 3 | Batch 25599/57412 | Loss: 0.326 | time 3008.2245423793793 |
| Epoch 3 | Batch 25799/57412 | Loss: 0.586 | time 3032.3743720054626 |
| Epoch 3 | Batch 25999/57412 | Loss: 0.116 | time 3055.672827720642 |
| Epoch 3 | Batch 26199/57412 | Loss: 0.281 | time 3078.6556005477905 |
| Epoch 3 | Batch 26399/57412 | Loss: 0.320 | time 3101.2326440811157 |
| Epoch 3 | Batch 26599/57412 | Loss: 0.369 | time 3124.1519243717194 |
| Epoch 3 | Batch 26799/57412 | Loss: 0.352 | time 3147.004638671875 |
| Epoch 3 | Batch 26999/57412 | Loss: 0.414 | time 3170.1297340393066 |
| Epoch 3 | Batch 27199/57412 | Loss: 0.418 | time 3194.123152732849 |
| Epoch 3 | Batch 27399/57412 | Loss: 0.496 | time 3217.0843164920807 |
| Epoch 3 | Batch 27599/57412 | Loss: 0.453 | time 3239.4660131931305 |
| Epoch 3 | Batch 27799/57412 | Loss: 0.281 | time 3261.648086309433 |
| Epoch 3 | Batch 27999/57412 | Loss: 0.660 | time 3284.825266122818 |
| Epoch 3 | Batch 28199/57412 | Loss: 0.447 | time 3308.389226913452 |
| Epoch 3 | Batch 28399/57412 | Loss: 0.367 | time 3332.0319731235504 |
| Epoch 3 | Batch 28599/57412 | Loss: 0.285 | time 3355.1665682792664 |
| Epoch 3 | Batch 28799/57412 | Loss: 0.531 | time 3376.514209508896 |
| Epoch 3 | Batch 28999/57412 | Loss: 0.326 | time 3399.5467870235443 |
| Epoch 3 | Batch 29199/57412 | Loss: 0.252 | time 3423.5144033432007 |
| Epoch 3 | Batch 29399/57412 | Loss: 0.268 | time 3446.4941334724426 |
| Epoch 3 | Batch 29599/57412 | Loss: 0.375 | time 3469.7240076065063 |
| Epoch 3 | Batch 29799/57412 | Loss: 0.301 | time 3492.7764270305634 |
| Epoch 3 | Batch 29999/57412 | Loss: 0.531 | time 3515.082939147949 |
| Epoch 3 | Batch 30199/57412 | Loss: 0.436 | time 3537.014974594116 |
| Epoch 3 | Batch 30399/57412 | Loss: 0.281 | time 3561.205808877945 |
| Epoch 3 | Batch 30599/57412 | Loss: 0.235 | time 3585.1383357048035 |
| Epoch 3 | Batch 30799/57412 | Loss: 0.475 | time 3607.579728603363 |
| Epoch 3 | Batch 30999/57412 | Loss: 0.324 | time 3630.968156814575 |
| Epoch 3 | Batch 31199/57412 | Loss: 0.004 | time 3654.537914991379 |
| Epoch 3 | Batch 31399/57412 | Loss: 0.059 | time 3678.59618973732 |
| Epoch 3 | Batch 31599/57412 | Loss: 0.586 | time 3701.5789675712585 |
| Epoch 3 | Batch 31799/57412 | Loss: 0.277 | time 3725.117594242096 |
| Epoch 3 | Batch 31999/57412 | Loss: 0.439 | time 3749.1817350387573 |
| Epoch 3 | Batch 32199/57412 | Loss: 0.590 | time 3772.0010759830475 |
| Epoch 3 | Batch 32399/57412 | Loss: 0.193 | time 3795.05877161026 |
| Epoch 3 | Batch 32599/57412 | Loss: 0.479 | time 3817.3109591007233 |
| Epoch 3 | Batch 32799/57412 | Loss: 0.535 | time 3841.398181915283 |
| Epoch 3 | Batch 32999/57412 | Loss: 0.598 | time 3863.6101653575897 |
| Epoch 3 | Batch 33199/57412 | Loss: 0.473 | time 3887.163375854492 |
| Epoch 3 | Batch 33399/57412 | Loss: 0.126 | time 3909.9980487823486 |
| Epoch 3 | Batch 33599/57412 | Loss: 0.135 | time 3932.3990290164948 |
| Epoch 3 | Batch 33799/57412 | Loss: 0.214 | time 3955.1816573143005 |
| Epoch 3 | Batch 33999/57412 | Loss: 0.410 | time 3978.5499546527863 |
| Epoch 3 | Batch 34199/57412 | Loss: 0.387 | time 4001.372322320938 |
| Epoch 3 | Batch 34399/57412 | Loss: 0.328 | time 4025.20555973053 |
| Epoch 3 | Batch 34599/57412 | Loss: 0.486 | time 4049.4531366825104 |
| Epoch 3 | Batch 34799/57412 | Loss: 0.395 | time 4072.5129849910736 |
| Epoch 3 | Batch 34999/57412 | Loss: 0.688 | time 4096.999430179596 |
| Epoch 3 | Batch 35199/57412 | Loss: 0.359 | time 4119.947369098663 |
| Epoch 3 | Batch 35399/57412 | Loss: 0.236 | time 4143.676243543625 |
| Epoch 3 | Batch 35599/57412 | Loss: 0.324 | time 4166.168452739716 |
| Epoch 3 | Batch 35799/57412 | Loss: 0.477 | time 4188.578604698181 |
| Epoch 3 | Batch 35999/57412 | Loss: 0.490 | time 4211.138625383377 |
| Epoch 3 | Batch 36199/57412 | Loss: 0.083 | time 4234.155810832977 |
| Epoch 3 | Batch 36399/57412 | Loss: 0.289 | time 4257.857188463211 |
| Epoch 3 | Batch 36599/57412 | Loss: 0.244 | time 4280.635377407074 |
| Epoch 3 | Batch 36799/57412 | Loss: 0.396 | time 4303.7600836753845 |
| Epoch 3 | Batch 36999/57412 | Loss: 0.106 | time 4327.302592754364 |
| Epoch 3 | Batch 37199/57412 | Loss: 0.439 | time 4350.756130456924 |
| Epoch 3 | Batch 37399/57412 | Loss: 0.234 | time 4374.7218108177185 |
| Epoch 3 | Batch 37599/57412 | Loss: 0.389 | time 4396.977281332016 |
| Epoch 3 | Batch 37799/57412 | Loss: 0.363 | time 4419.5213351249695 |
| Epoch 3 | Batch 37999/57412 | Loss: 0.348 | time 4442.457938909531 |
| Epoch 3 | Batch 38199/57412 | Loss: 0.001 | time 4467.114045143127 |
| Epoch 3 | Batch 38399/57412 | Loss: 0.264 | time 4489.653801679611 |
| Epoch 3 | Batch 38599/57412 | Loss: 0.217 | time 4512.477380514145 |
| Epoch 3 | Batch 38799/57412 | Loss: 0.578 | time 4535.848119497299 |
| Epoch 3 | Batch 38999/57412 | Loss: 0.475 | time 4560.175380706787 |
| Epoch 3 | Batch 39199/57412 | Loss: 0.494 | time 4583.11111164093 |
| Epoch 3 | Batch 39399/57412 | Loss: 0.387 | time 4607.068244457245 |
| Epoch 3 | Batch 39599/57412 | Loss: 0.695 | time 4630.919080734253 |
| Epoch 3 | Batch 39799/57412 | Loss: 0.395 | time 4654.906218528748 |
| Epoch 3 | Batch 39999/57412 | Loss: 0.582 | time 4678.4728474617 |
| Saving at Step 40000 |
| Epoch 3 | Batch 40199/57412 | Loss: 0.000 | time 4705.367149353027 |
| Epoch 3 | Batch 40399/57412 | Loss: 0.625 | time 4727.797984361649 |
| Epoch 3 | Batch 40599/57412 | Loss: 0.307 | time 4751.3316214084625 |
| Epoch 3 | Batch 40799/57412 | Loss: 0.289 | time 4773.484359264374 |
| Epoch 3 | Batch 40999/57412 | Loss: 0.256 | time 4797.688658714294 |
| Epoch 3 | Batch 41199/57412 | Loss: 0.291 | time 4820.651803016663 |
| Epoch 3 | Batch 41399/57412 | Loss: 0.352 | time 4844.737418174744 |
| Epoch 3 | Batch 41599/57412 | Loss: 0.375 | time 4867.236777305603 |
| Epoch 3 | Batch 41799/57412 | Loss: 0.187 | time 4891.328143835068 |
| Epoch 3 | Batch 41999/57412 | Loss: 0.301 | time 4913.87936258316 |
| Epoch 3 | Batch 42199/57412 | Loss: 0.260 | time 4939.013718128204 |
| Epoch 3 | Batch 42399/57412 | Loss: 0.260 | time 4963.03479218483 |
| Epoch 3 | Batch 42599/57412 | Loss: 0.520 | time 4986.6232368946075 |
| Epoch 3 | Batch 42799/57412 | Loss: 0.283 | time 5010.496536254883 |
| Epoch 3 | Batch 42999/57412 | Loss: 0.316 | time 5032.7325348854065 |
| Epoch 3 | Batch 43199/57412 | Loss: 0.289 | time 5057.362567663193 |
| Epoch 3 | Batch 43399/57412 | Loss: 0.398 | time 5080.911366701126 |
| Epoch 3 | Batch 43599/57412 | Loss: 0.357 | time 5105.352005243301 |
| Epoch 3 | Batch 43799/57412 | Loss: 0.426 | time 5128.566524028778 |
| Epoch 3 | Batch 43999/57412 | Loss: 0.559 | time 5151.5418400764465 |
| Epoch 3 | Batch 44199/57412 | Loss: 0.457 | time 5174.371671676636 |
| Epoch 3 | Batch 44399/57412 | Loss: 0.385 | time 5198.381245851517 |
| Epoch 3 | Batch 44599/57412 | Loss: 0.424 | time 5222.415331363678 |
| Epoch 3 | Batch 44799/57412 | Loss: 0.379 | time 5245.043105125427 |
| Epoch 3 | Batch 44999/57412 | Loss: 0.188 | time 5267.359858512878 |
| Epoch 3 | Batch 45199/57412 | Loss: 0.438 | time 5290.810136795044 |
| Epoch 3 | Batch 45399/57412 | Loss: 0.336 | time 5313.939893484116 |
| Epoch 3 | Batch 45599/57412 | Loss: 0.447 | time 5338.02358174324 |
| Epoch 3 | Batch 45799/57412 | Loss: 0.287 | time 5360.396451711655 |
| Epoch 3 | Batch 45999/57412 | Loss: 0.034 | time 5383.638774871826 |
| Epoch 3 | Batch 46199/57412 | Loss: 0.471 | time 5406.4434287548065 |
| Epoch 3 | Batch 46399/57412 | Loss: 0.187 | time 5428.44180893898 |
| Epoch 3 | Batch 46599/57412 | Loss: 0.307 | time 5453.035503149033 |
| Epoch 3 | Batch 46799/57412 | Loss: 0.613 | time 5476.416801691055 |
| Epoch 3 | Batch 46999/57412 | Loss: 0.348 | time 5500.2231459617615 |
| Epoch 3 | Batch 47199/57412 | Loss: 0.271 | time 5523.278842449188 |
| Epoch 3 | Batch 47399/57412 | Loss: 0.326 | time 5545.145682096481 |
| Epoch 3 | Batch 47599/57412 | Loss: 0.328 | time 5568.913462400436 |
| Epoch 3 | Batch 47799/57412 | Loss: 0.311 | time 5591.852295637131 |
| Epoch 3 | Batch 47999/57412 | Loss: 0.170 | time 5616.073298931122 |
| Epoch 3 | Batch 48199/57412 | Loss: 0.613 | time 5640.4591019153595 |
| Epoch 3 | Batch 48399/57412 | Loss: 0.271 | time 5662.576171398163 |
| Epoch 3 | Batch 48599/57412 | Loss: 0.229 | time 5686.076673269272 |
| Epoch 3 | Batch 48799/57412 | Loss: 0.434 | time 5707.987272024155 |
| Epoch 3 | Batch 48999/57412 | Loss: 0.449 | time 5730.814256906509 |
| Epoch 3 | Batch 49199/57412 | Loss: 0.314 | time 5754.124645233154 |
| Epoch 3 | Batch 49399/57412 | Loss: 0.365 | time 5777.053218841553 |
| Epoch 3 | Batch 49599/57412 | Loss: 0.350 | time 5800.987100601196 |
| Epoch 3 | Batch 49799/57412 | Loss: 0.243 | time 5823.86700296402 |
| Epoch 3 | Batch 49999/57412 | Loss: 0.326 | time 5847.0807683467865 |
| Epoch 3 | Batch 50199/57412 | Loss: 0.158 | time 5870.738385438919 |
| Epoch 3 | Batch 50399/57412 | Loss: 0.346 | time 5894.578683376312 |
| Epoch 3 | Batch 50599/57412 | Loss: 0.391 | time 5917.4002339839935 |
| Epoch 3 | Batch 50799/57412 | Loss: 0.311 | time 5941.251713991165 |
| Epoch 3 | Batch 50999/57412 | Loss: 0.527 | time 5965.64574098587 |
| Epoch 3 | Batch 51199/57412 | Loss: 0.504 | time 5988.556133747101 |
| Epoch 3 | Batch 51399/57412 | Loss: 0.309 | time 6013.077621936798 |
| Epoch 3 | Batch 51599/57412 | Loss: 0.344 | time 6036.408913850784 |
| Epoch 3 | Batch 51799/57412 | Loss: 0.406 | time 6059.832098007202 |
| Epoch 3 | Batch 51999/57412 | Loss: 0.547 | time 6083.2948088645935 |
| Epoch 3 | Batch 52199/57412 | Loss: 0.383 | time 6108.2534692287445 |
| Epoch 3 | Batch 52399/57412 | Loss: 0.332 | time 6132.319977521896 |
| Epoch 3 | Batch 52599/57412 | Loss: 0.559 | time 6156.333340883255 |
| Epoch 3 | Batch 52799/57412 | Loss: 0.158 | time 6180.090841293335 |
| Epoch 3 | Batch 52999/57412 | Loss: 0.512 | time 6202.630301475525 |
| Epoch 3 | Batch 53199/57412 | Loss: 0.432 | time 6225.086124658585 |
| Epoch 3 | Batch 53399/57412 | Loss: 0.102 | time 6248.214440822601 |
| Epoch 3 | Batch 53599/57412 | Loss: 0.252 | time 6271.024737358093 |
| Epoch 3 | Batch 53799/57412 | Loss: 0.320 | time 6294.702907800674 |
| Epoch 3 | Batch 53999/57412 | Loss: 0.207 | time 6319.758134365082 |
| Epoch 3 | Batch 54199/57412 | Loss: 0.367 | time 6344.300902366638 |
| Epoch 3 | Batch 54399/57412 | Loss: 0.334 | time 6368.156083106995 |
| Epoch 3 | Batch 54599/57412 | Loss: 0.355 | time 6391.59591627121 |
| Epoch 3 | Batch 54799/57412 | Loss: 0.389 | time 6414.272758245468 |
| Epoch 3 | Batch 54999/57412 | Loss: 0.562 | time 6438.155944108963 |
| Epoch 3 | Batch 55199/57412 | Loss: 0.328 | time 6462.041528940201 |
| Epoch 3 | Batch 55399/57412 | Loss: 0.389 | time 6485.022021532059 |
| Epoch 3 | Batch 55599/57412 | Loss: 0.398 | time 6508.931003570557 |
| Epoch 3 | Batch 55799/57412 | Loss: 0.516 | time 6532.689184427261 |
| Epoch 3 | Batch 55999/57412 | Loss: 0.075 | time 6555.6579830646515 |
| Epoch 3 | Batch 56199/57412 | Loss: 0.258 | time 6578.895622730255 |
| Epoch 3 | Batch 56399/57412 | Loss: 0.111 | time 6601.354923963547 |
| Epoch 3 | Batch 56599/57412 | Loss: 0.312 | time 6628.01166510582 |
| Epoch 3 | Batch 56799/57412 | Loss: 0.445 | time 6651.135204315186 |
| Epoch 3 | Batch 56999/57412 | Loss: 0.160 | time 6676.094683647156 |
| Epoch 3 | Batch 57199/57412 | Loss: 0.367 | time 6700.700124502182 |
| Epoch 3 | Batch 57399/57412 | Loss: 0.238 | time 6724.263410329819 |
| ========================================================================================= |
| ========================================================================================= |
| ========================================================================================= |
| Epoch 3 | Validation Samples 5082 | Validation Acc: 0.6442345533254624 | time 335.08467149734497 |
| ========================================================================================= |
| Saving at Epoch 3 | Learning rate: 2e-05 |
| Epoch 4 | Batch 199/57412 | Loss: 0.354 | time 24.36689829826355 |
| Epoch 4 | Batch 399/57412 | Loss: 0.241 | time 48.1044704914093 |
| Epoch 4 | Batch 599/57412 | Loss: 0.377 | time 71.6808557510376 |
| Epoch 4 | Batch 799/57412 | Loss: 0.281 | time 94.92673444747925 |
| Epoch 4 | Batch 999/57412 | Loss: 0.441 | time 117.4968729019165 |
| Epoch 4 | Batch 1199/57412 | Loss: 0.449 | time 140.26612877845764 |
| Epoch 4 | Batch 1399/57412 | Loss: 0.539 | time 165.6131021976471 |
| Epoch 4 | Batch 1599/57412 | Loss: 0.424 | time 190.34646487236023 |
| Epoch 4 | Batch 1799/57412 | Loss: 0.363 | time 212.68677520751953 |
| Epoch 4 | Batch 1999/57412 | Loss: 0.098 | time 235.6713948249817 |
| Epoch 4 | Batch 2199/57412 | Loss: 0.192 | time 257.7385983467102 |
| Epoch 4 | Batch 2399/57412 | Loss: 0.699 | time 279.7700788974762 |
| Epoch 4 | Batch 2599/57412 | Loss: 0.371 | time 304.1338791847229 |
| Epoch 4 | Batch 2799/57412 | Loss: 0.357 | time 327.1127965450287 |
| Epoch 4 | Batch 2999/57412 | Loss: 0.361 | time 351.2816889286041 |
| Epoch 4 | Batch 3199/57412 | Loss: 0.520 | time 374.5480704307556 |
| Epoch 4 | Batch 3399/57412 | Loss: 0.340 | time 398.54401874542236 |
| Epoch 4 | Batch 3599/57412 | Loss: 0.424 | time 421.75643968582153 |
| Epoch 4 | Batch 3799/57412 | Loss: 0.461 | time 446.13209652900696 |
| Epoch 4 | Batch 3999/57412 | Loss: 0.213 | time 469.5954418182373 |
| Epoch 4 | Batch 4199/57412 | Loss: 0.404 | time 491.98157238960266 |
| Epoch 4 | Batch 4399/57412 | Loss: 0.625 | time 515.6342830657959 |
| Epoch 4 | Batch 4599/57412 | Loss: 0.242 | time 539.7441890239716 |
| Epoch 4 | Batch 4799/57412 | Loss: 0.369 | time 562.0571594238281 |
| Epoch 4 | Batch 4999/57412 | Loss: 0.482 | time 584.2583739757538 |
| Epoch 4 | Batch 5199/57412 | Loss: 0.404 | time 607.7478001117706 |
| Epoch 4 | Batch 5399/57412 | Loss: 0.334 | time 630.990119934082 |
| Epoch 4 | Batch 5599/57412 | Loss: 0.418 | time 654.1483161449432 |
| Epoch 4 | Batch 5799/57412 | Loss: 0.350 | time 677.4689164161682 |
| Epoch 4 | Batch 5999/57412 | Loss: 0.500 | time 700.6690652370453 |
| Epoch 4 | Batch 6199/57412 | Loss: 0.414 | time 723.8998355865479 |
| Epoch 4 | Batch 6399/57412 | Loss: 0.256 | time 747.7298617362976 |
| Epoch 4 | Batch 6599/57412 | Loss: 0.295 | time 771.5192062854767 |
| Epoch 4 | Batch 6799/57412 | Loss: 0.422 | time 796.6871383190155 |
| Epoch 4 | Batch 6999/57412 | Loss: 0.479 | time 820.7280824184418 |
| Epoch 4 | Batch 7199/57412 | Loss: 0.311 | time 844.1770751476288 |
| Epoch 4 | Batch 7399/57412 | Loss: 0.324 | time 867.5507764816284 |
| Epoch 4 | Batch 7599/57412 | Loss: 0.508 | time 891.5266101360321 |
| Epoch 4 | Batch 7799/57412 | Loss: 0.004 | time 915.9359257221222 |
| Epoch 4 | Batch 7999/57412 | Loss: 0.516 | time 938.3805129528046 |
| Epoch 4 | Batch 8199/57412 | Loss: 0.192 | time 961.8675553798676 |
| Epoch 4 | Batch 8399/57412 | Loss: 0.463 | time 985.3883843421936 |
| Epoch 4 | Batch 8599/57412 | Loss: 0.250 | time 1009.4400238990784 |
| Epoch 4 | Batch 8799/57412 | Loss: 0.527 | time 1035.4226496219635 |
| Epoch 4 | Batch 8999/57412 | Loss: 0.434 | time 1060.4466450214386 |
| Epoch 4 | Batch 9199/57412 | Loss: 0.527 | time 1085.2911865711212 |
| Epoch 4 | Batch 9399/57412 | Loss: 0.467 | time 1110.9656817913055 |
| Epoch 4 | Batch 9599/57412 | Loss: 0.287 | time 1135.9828443527222 |
| Epoch 4 | Batch 9799/57412 | Loss: 0.092 | time 1160.642065525055 |
| Epoch 4 | Batch 9999/57412 | Loss: 0.492 | time 1185.0910873413086 |
| Epoch 4 | Batch 10199/57412 | Loss: 0.162 | time 1209.052606344223 |
| Epoch 4 | Batch 10399/57412 | Loss: 0.551 | time 1233.6134226322174 |
| Epoch 4 | Batch 10599/57412 | Loss: 0.365 | time 1256.7099039554596 |
| Epoch 4 | Batch 10799/57412 | Loss: 0.068 | time 1281.0756137371063 |
| Epoch 4 | Batch 10999/57412 | Loss: 0.236 | time 1304.2188804149628 |
| Epoch 4 | Batch 11199/57412 | Loss: 0.236 | time 1328.2793533802032 |
| Epoch 4 | Batch 11399/57412 | Loss: 0.475 | time 1351.9445230960846 |
| Epoch 4 | Batch 11599/57412 | Loss: 0.309 | time 1377.8863677978516 |
| Epoch 4 | Batch 11799/57412 | Loss: 0.322 | time 1401.0995514392853 |
| Epoch 4 | Batch 11999/57412 | Loss: 0.439 | time 1425.6843705177307 |
| Epoch 4 | Batch 12199/57412 | Loss: 0.451 | time 1450.2585818767548 |
| Epoch 4 | Batch 12399/57412 | Loss: 0.227 | time 1474.5420925617218 |
| Epoch 4 | Batch 12599/57412 | Loss: 0.361 | time 1499.5125741958618 |
| Epoch 4 | Batch 12799/57412 | Loss: 0.226 | time 1523.3865714073181 |
| Epoch 4 | Batch 12999/57412 | Loss: 0.562 | time 1547.7950751781464 |
| Epoch 4 | Batch 13199/57412 | Loss: 0.056 | time 1571.8889963626862 |
| Epoch 4 | Batch 13399/57412 | Loss: 0.426 | time 1597.0879261493683 |
| Epoch 4 | Batch 13599/57412 | Loss: 0.248 | time 1621.0460579395294 |
| Epoch 4 | Batch 13799/57412 | Loss: 0.420 | time 1645.0873939990997 |
| Epoch 4 | Batch 13999/57412 | Loss: 0.231 | time 1668.5003101825714 |
| Epoch 4 | Batch 14199/57412 | Loss: 0.436 | time 1693.7063710689545 |
| Epoch 4 | Batch 14399/57412 | Loss: 0.188 | time 1718.1466681957245 |
| Epoch 4 | Batch 14599/57412 | Loss: 0.420 | time 1743.152248620987 |
| Epoch 4 | Batch 14799/57412 | Loss: 0.320 | time 1767.376046180725 |
| Epoch 4 | Batch 14999/57412 | Loss: 0.193 | time 1790.8851969242096 |
| Epoch 4 | Batch 15199/57412 | Loss: 0.103 | time 1814.938633441925 |
| Epoch 4 | Batch 15399/57412 | Loss: 0.416 | time 1840.5210325717926 |
| Epoch 4 | Batch 15599/57412 | Loss: 0.391 | time 1866.2879490852356 |
| Epoch 4 | Batch 15799/57412 | Loss: 0.332 | time 1890.7367014884949 |
| Epoch 4 | Batch 15999/57412 | Loss: 0.295 | time 1913.9210841655731 |
| Epoch 4 | Batch 16199/57412 | Loss: 0.017 | time 1937.1711566448212 |
| Epoch 4 | Batch 16399/57412 | Loss: 0.447 | time 1960.2118692398071 |
| Epoch 4 | Batch 16599/57412 | Loss: 0.330 | time 1983.6238422393799 |
| Epoch 4 | Batch 16799/57412 | Loss: 0.182 | time 2005.915447473526 |
| Epoch 4 | Batch 16999/57412 | Loss: 0.371 | time 2029.1312718391418 |
| Epoch 4 | Batch 17199/57412 | Loss: 0.312 | time 2051.5964148044586 |
| Epoch 4 | Batch 17399/57412 | Loss: 0.057 | time 2074.3933730125427 |
| Epoch 4 | Batch 17599/57412 | Loss: 0.365 | time 2097.842917919159 |
| Epoch 4 | Batch 17799/57412 | Loss: 0.093 | time 2120.494915008545 |
| Epoch 4 | Batch 17999/57412 | Loss: 0.240 | time 2142.409027338028 |
| Epoch 4 | Batch 18199/57412 | Loss: 0.316 | time 2164.993678331375 |
| Epoch 4 | Batch 18399/57412 | Loss: 0.084 | time 2190.9753189086914 |
| Epoch 4 | Batch 18599/57412 | Loss: 0.361 | time 2214.3861541748047 |
| Epoch 4 | Batch 18799/57412 | Loss: 0.426 | time 2238.2787079811096 |
| Epoch 4 | Batch 18999/57412 | Loss: 0.359 | time 2260.9510226249695 |
| Epoch 4 | Batch 19199/57412 | Loss: 0.393 | time 2283.4741060733795 |
| Epoch 4 | Batch 19399/57412 | Loss: 0.175 | time 2307.0103294849396 |
| Epoch 4 | Batch 19599/57412 | Loss: 0.340 | time 2331.0298733711243 |
| Epoch 4 | Batch 19799/57412 | Loss: 0.015 | time 2354.988091945648 |
| Epoch 4 | Batch 19999/57412 | Loss: 0.482 | time 2377.1519737243652 |
| Saving at Step 20000 |
| Epoch 4 | Batch 20199/57412 | Loss: 0.222 | time 2402.7941341400146 |
| Epoch 4 | Batch 20399/57412 | Loss: 0.273 | time 2425.784821033478 |
| Epoch 4 | Batch 20599/57412 | Loss: 0.047 | time 2448.019722700119 |
| Epoch 4 | Batch 20799/57412 | Loss: 0.320 | time 2471.727391242981 |
| Epoch 4 | Batch 20999/57412 | Loss: 0.348 | time 2496.127469062805 |
| Epoch 4 | Batch 21199/57412 | Loss: 0.543 | time 2519.385324716568 |
| Epoch 4 | Batch 21399/57412 | Loss: 0.115 | time 2541.0773096084595 |
| Epoch 4 | Batch 21599/57412 | Loss: 0.523 | time 2563.097997188568 |
| Epoch 4 | Batch 21799/57412 | Loss: 0.176 | time 2585.2969179153442 |
| Epoch 4 | Batch 21999/57412 | Loss: 0.336 | time 2608.609926223755 |
| Epoch 4 | Batch 22199/57412 | Loss: 0.404 | time 2630.6576375961304 |
| Epoch 4 | Batch 22399/57412 | Loss: 0.402 | time 2653.7464582920074 |
| Epoch 4 | Batch 22599/57412 | Loss: 0.065 | time 2677.2225165367126 |
| Epoch 4 | Batch 22799/57412 | Loss: 0.176 | time 2699.7369661331177 |
| Epoch 4 | Batch 22999/57412 | Loss: 0.449 | time 2723.0884618759155 |
| Epoch 4 | Batch 23199/57412 | Loss: 0.167 | time 2746.3266627788544 |
| Epoch 4 | Batch 23399/57412 | Loss: 0.271 | time 2769.7457432746887 |
| Epoch 4 | Batch 23599/57412 | Loss: 0.266 | time 2792.2621154785156 |
| Epoch 4 | Batch 23799/57412 | Loss: 0.176 | time 2815.3663341999054 |
| Epoch 4 | Batch 23999/57412 | Loss: 0.254 | time 2837.5089251995087 |
| Epoch 4 | Batch 24199/57412 | Loss: 0.297 | time 2860.796506881714 |
| Epoch 4 | Batch 24399/57412 | Loss: 0.504 | time 2884.6839866638184 |
| Epoch 4 | Batch 24599/57412 | Loss: 0.359 | time 2908.556876897812 |
| Epoch 4 | Batch 24799/57412 | Loss: 0.412 | time 2931.249741077423 |
| Epoch 4 | Batch 24999/57412 | Loss: 0.241 | time 2954.2007677555084 |
| Epoch 4 | Batch 25199/57412 | Loss: 0.144 | time 2977.1075904369354 |
| Epoch 4 | Batch 25399/57412 | Loss: 0.391 | time 2999.530401945114 |
| Epoch 4 | Batch 25599/57412 | Loss: 0.320 | time 3021.9782807826996 |
| Epoch 4 | Batch 25799/57412 | Loss: 0.555 | time 3046.383702516556 |
| Epoch 4 | Batch 25999/57412 | Loss: 0.115 | time 3069.6856410503387 |
| Epoch 4 | Batch 26199/57412 | Loss: 0.223 | time 3092.532830953598 |
| Epoch 4 | Batch 26399/57412 | Loss: 0.301 | time 3114.833112001419 |
| Epoch 4 | Batch 26599/57412 | Loss: 0.338 | time 3137.270812034607 |
| Epoch 4 | Batch 26799/57412 | Loss: 0.297 | time 3159.848204135895 |
| Epoch 4 | Batch 26999/57412 | Loss: 0.373 | time 3182.718150615692 |
| Epoch 4 | Batch 27199/57412 | Loss: 0.404 | time 3206.659379720688 |
| Epoch 4 | Batch 27399/57412 | Loss: 0.482 | time 3229.394419670105 |
| Epoch 4 | Batch 27599/57412 | Loss: 0.424 | time 3251.669728040695 |
| Epoch 4 | Batch 27799/57412 | Loss: 0.266 | time 3273.6558787822723 |
| Epoch 4 | Batch 27999/57412 | Loss: 0.652 | time 3296.7549500465393 |
| Epoch 4 | Batch 28199/57412 | Loss: 0.398 | time 3320.0113518238068 |
| Epoch 4 | Batch 28399/57412 | Loss: 0.334 | time 3343.478996515274 |
| Epoch 4 | Batch 28599/57412 | Loss: 0.239 | time 3366.305715084076 |
| Epoch 4 | Batch 28799/57412 | Loss: 0.527 | time 3387.699791908264 |
| Epoch 4 | Batch 28999/57412 | Loss: 0.312 | time 3410.5539495944977 |
| Epoch 4 | Batch 29199/57412 | Loss: 0.179 | time 3434.120017051697 |
| Epoch 4 | Batch 29399/57412 | Loss: 0.256 | time 3456.6513454914093 |
| Epoch 4 | Batch 29599/57412 | Loss: 0.367 | time 3479.551523208618 |
| Epoch 4 | Batch 29799/57412 | Loss: 0.258 | time 3501.732957839966 |
| Epoch 4 | Batch 29999/57412 | Loss: 0.490 | time 3523.669947385788 |
| Epoch 4 | Batch 30199/57412 | Loss: 0.434 | time 3545.2941434383392 |
| Epoch 4 | Batch 30399/57412 | Loss: 0.266 | time 3569.615243911743 |
| Epoch 4 | Batch 30599/57412 | Loss: 0.215 | time 3593.4513144493103 |
| Epoch 4 | Batch 30799/57412 | Loss: 0.469 | time 3615.6303515434265 |
| Epoch 4 | Batch 30999/57412 | Loss: 0.314 | time 3639.4726178646088 |
| Epoch 4 | Batch 31199/57412 | Loss: 0.000 | time 3663.4733612537384 |
| Epoch 4 | Batch 31399/57412 | Loss: 0.062 | time 3687.879812479019 |
| Epoch 4 | Batch 31599/57412 | Loss: 0.543 | time 3711.7268583774567 |
| Epoch 4 | Batch 31799/57412 | Loss: 0.256 | time 3735.8132469654083 |
| Epoch 4 | Batch 31999/57412 | Loss: 0.371 | time 3760.9614849090576 |
| Epoch 4 | Batch 32199/57412 | Loss: 0.516 | time 3784.7334196567535 |
| Epoch 4 | Batch 32399/57412 | Loss: 0.163 | time 3808.5275070667267 |
| Epoch 4 | Batch 32599/57412 | Loss: 0.449 | time 3831.4370155334473 |
| Epoch 4 | Batch 32799/57412 | Loss: 0.447 | time 3856.002972126007 |
| Epoch 4 | Batch 32999/57412 | Loss: 0.547 | time 3877.9396851062775 |
| Epoch 4 | Batch 33199/57412 | Loss: 0.438 | time 3900.8977615833282 |
| Epoch 4 | Batch 33399/57412 | Loss: 0.089 | time 3923.6203191280365 |
| Epoch 4 | Batch 33599/57412 | Loss: 0.075 | time 3946.3739132881165 |
| Epoch 4 | Batch 33799/57412 | Loss: 0.194 | time 3969.111894130707 |
| Epoch 4 | Batch 33999/57412 | Loss: 0.398 | time 3992.417114496231 |
| Epoch 4 | Batch 34199/57412 | Loss: 0.359 | time 4015.347536802292 |
| Epoch 4 | Batch 34399/57412 | Loss: 0.316 | time 4039.3347520828247 |
| Epoch 4 | Batch 34599/57412 | Loss: 0.480 | time 4063.934530735016 |
| Epoch 4 | Batch 34799/57412 | Loss: 0.359 | time 4086.8443534374237 |
| Epoch 4 | Batch 34999/57412 | Loss: 0.664 | time 4111.043667554855 |
| Epoch 4 | Batch 35199/57412 | Loss: 0.338 | time 4134.325147390366 |
| Epoch 4 | Batch 35399/57412 | Loss: 0.217 | time 4158.160282611847 |
| Epoch 4 | Batch 35599/57412 | Loss: 0.299 | time 4180.479394674301 |
| Epoch 4 | Batch 35799/57412 | Loss: 0.416 | time 4203.857700109482 |
| Epoch 4 | Batch 35999/57412 | Loss: 0.484 | time 4227.661268234253 |
| Epoch 4 | Batch 36199/57412 | Loss: 0.066 | time 4251.890486240387 |
| Epoch 4 | Batch 36399/57412 | Loss: 0.244 | time 4276.427851438522 |
| Epoch 4 | Batch 36599/57412 | Loss: 0.230 | time 4299.977943181992 |
| Epoch 4 | Batch 36799/57412 | Loss: 0.365 | time 4324.636483669281 |
| Epoch 4 | Batch 36999/57412 | Loss: 0.097 | time 4349.312540531158 |
| Epoch 4 | Batch 37199/57412 | Loss: 0.418 | time 4372.65091586113 |
| Epoch 4 | Batch 37399/57412 | Loss: 0.214 | time 4397.20125079155 |
| Epoch 4 | Batch 37599/57412 | Loss: 0.357 | time 4420.960392951965 |
| Epoch 4 | Batch 37799/57412 | Loss: 0.350 | time 4445.084036827087 |
| Epoch 4 | Batch 37999/57412 | Loss: 0.338 | time 4469.165984630585 |
| Epoch 4 | Batch 38199/57412 | Loss: 0.000 | time 4494.020138502121 |
| Epoch 4 | Batch 38399/57412 | Loss: 0.247 | time 4516.544036388397 |
| Epoch 4 | Batch 38599/57412 | Loss: 0.205 | time 4538.912894487381 |
| Epoch 4 | Batch 38799/57412 | Loss: 0.482 | time 4561.875642299652 |
| Epoch 4 | Batch 38999/57412 | Loss: 0.438 | time 4585.805818557739 |
| Epoch 4 | Batch 39199/57412 | Loss: 0.414 | time 4608.1632533073425 |
| Epoch 4 | Batch 39399/57412 | Loss: 0.373 | time 4631.715314149857 |
| Epoch 4 | Batch 39599/57412 | Loss: 0.664 | time 4655.087517976761 |
| Epoch 4 | Batch 39799/57412 | Loss: 0.275 | time 4678.952908039093 |
| Epoch 4 | Batch 39999/57412 | Loss: 0.562 | time 4702.953208684921 |
| Saving at Step 40000 |
| Epoch 4 | Batch 40199/57412 | Loss: 0.000 | time 4730.9554126262665 |
| Epoch 4 | Batch 40399/57412 | Loss: 0.582 | time 4754.501868247986 |
| Epoch 4 | Batch 40599/57412 | Loss: 0.256 | time 4778.756006479263 |
| Epoch 4 | Batch 40799/57412 | Loss: 0.281 | time 4801.282264471054 |
| Epoch 4 | Batch 40999/57412 | Loss: 0.234 | time 4826.218091011047 |
| Epoch 4 | Batch 41199/57412 | Loss: 0.285 | time 4849.43669128418 |
| Epoch 4 | Batch 41399/57412 | Loss: 0.348 | time 4873.856832027435 |
| Epoch 4 | Batch 41599/57412 | Loss: 0.354 | time 4896.587930679321 |
| Epoch 4 | Batch 41799/57412 | Loss: 0.188 | time 4920.824689626694 |
| Epoch 4 | Batch 41999/57412 | Loss: 0.186 | time 4943.662119150162 |
| Epoch 4 | Batch 42199/57412 | Loss: 0.247 | time 4968.939979791641 |
| Epoch 4 | Batch 42399/57412 | Loss: 0.254 | time 4992.672446489334 |
| Epoch 4 | Batch 42599/57412 | Loss: 0.498 | time 5015.9377064704895 |
| Epoch 4 | Batch 42799/57412 | Loss: 0.223 | time 5039.175618171692 |
| Epoch 4 | Batch 42999/57412 | Loss: 0.187 | time 5061.419154882431 |
| Epoch 4 | Batch 43199/57412 | Loss: 0.273 | time 5086.18358540535 |
| Epoch 4 | Batch 43399/57412 | Loss: 0.375 | time 5109.528914690018 |
| Epoch 4 | Batch 43599/57412 | Loss: 0.352 | time 5133.857528686523 |
| Epoch 4 | Batch 43799/57412 | Loss: 0.371 | time 5157.000137090683 |
| Epoch 4 | Batch 43999/57412 | Loss: 0.527 | time 5179.872038602829 |
| Epoch 4 | Batch 44199/57412 | Loss: 0.377 | time 5202.4445905685425 |
| Epoch 4 | Batch 44399/57412 | Loss: 0.373 | time 5226.489552974701 |
| Epoch 4 | Batch 44599/57412 | Loss: 0.404 | time 5250.453411817551 |
| Epoch 4 | Batch 44799/57412 | Loss: 0.375 | time 5273.023619174957 |
| Epoch 4 | Batch 44999/57412 | Loss: 0.099 | time 5294.995301485062 |
| Epoch 4 | Batch 45199/57412 | Loss: 0.418 | time 5318.402184963226 |
| Epoch 4 | Batch 45399/57412 | Loss: 0.291 | time 5341.334710121155 |
| Epoch 4 | Batch 45599/57412 | Loss: 0.426 | time 5365.078992843628 |
| Epoch 4 | Batch 45799/57412 | Loss: 0.281 | time 5387.222385168076 |
| Epoch 4 | Batch 45999/57412 | Loss: 0.015 | time 5409.8304142951965 |
| Epoch 4 | Batch 46199/57412 | Loss: 0.430 | time 5432.540081501007 |
| Epoch 4 | Batch 46399/57412 | Loss: 0.176 | time 5454.6674110889435 |
| Epoch 4 | Batch 46599/57412 | Loss: 0.303 | time 5479.104860544205 |
| Epoch 4 | Batch 46799/57412 | Loss: 0.574 | time 5502.284702062607 |
| Epoch 4 | Batch 46999/57412 | Loss: 0.336 | time 5525.844783067703 |
| Epoch 4 | Batch 47199/57412 | Loss: 0.244 | time 5548.839049339294 |
| Epoch 4 | Batch 47399/57412 | Loss: 0.293 | time 5570.526427030563 |
| Epoch 4 | Batch 47599/57412 | Loss: 0.309 | time 5593.873394012451 |
| Epoch 4 | Batch 47799/57412 | Loss: 0.279 | time 5616.658677339554 |
| Epoch 4 | Batch 47999/57412 | Loss: 0.160 | time 5641.3259863853455 |
| Epoch 4 | Batch 48199/57412 | Loss: 0.590 | time 5665.801714897156 |
| Epoch 4 | Batch 48399/57412 | Loss: 0.246 | time 5688.378141403198 |
| Epoch 4 | Batch 48599/57412 | Loss: 0.132 | time 5712.401171207428 |
| Epoch 4 | Batch 48799/57412 | Loss: 0.367 | time 5734.431068897247 |
| Epoch 4 | Batch 48999/57412 | Loss: 0.416 | time 5757.795754432678 |
| Epoch 4 | Batch 49199/57412 | Loss: 0.285 | time 5781.580869436264 |
| Epoch 4 | Batch 49399/57412 | Loss: 0.354 | time 5804.848086833954 |
| Epoch 4 | Batch 49599/57412 | Loss: 0.305 | time 5829.247303009033 |
| Epoch 4 | Batch 49799/57412 | Loss: 0.198 | time 5852.816997051239 |
| Epoch 4 | Batch 49999/57412 | Loss: 0.273 | time 5875.822581529617 |
| Epoch 4 | Batch 50199/57412 | Loss: 0.105 | time 5899.100542545319 |
| Epoch 4 | Batch 50399/57412 | Loss: 0.307 | time 5922.785616636276 |
| Epoch 4 | Batch 50599/57412 | Loss: 0.314 | time 5945.467596769333 |
| Epoch 4 | Batch 50799/57412 | Loss: 0.277 | time 5969.418765068054 |
| Epoch 4 | Batch 50999/57412 | Loss: 0.516 | time 5993.987643957138 |
| Epoch 4 | Batch 51199/57412 | Loss: 0.471 | time 6016.973550081253 |
| Epoch 4 | Batch 51399/57412 | Loss: 0.299 | time 6041.786082267761 |
| Epoch 4 | Batch 51599/57412 | Loss: 0.324 | time 6065.257765293121 |
| Epoch 4 | Batch 51799/57412 | Loss: 0.379 | time 6088.6483409404755 |
| Epoch 4 | Batch 51999/57412 | Loss: 0.523 | time 6112.344381570816 |
| Epoch 4 | Batch 52199/57412 | Loss: 0.365 | time 6137.49622631073 |
| Epoch 4 | Batch 52399/57412 | Loss: 0.301 | time 6161.612456083298 |
| Epoch 4 | Batch 52599/57412 | Loss: 0.531 | time 6185.796411514282 |
| Epoch 4 | Batch 52799/57412 | Loss: 0.153 | time 6210.185425758362 |
| Epoch 4 | Batch 52999/57412 | Loss: 0.383 | time 6233.554478645325 |
| Epoch 4 | Batch 53199/57412 | Loss: 0.416 | time 6256.3999979496 |
| Epoch 4 | Batch 53399/57412 | Loss: 0.097 | time 6279.832350730896 |
| Epoch 4 | Batch 53599/57412 | Loss: 0.246 | time 6302.871659994125 |
| Epoch 4 | Batch 53799/57412 | Loss: 0.279 | time 6326.684661149979 |
| Epoch 4 | Batch 53999/57412 | Loss: 0.193 | time 6351.752961158752 |
| Epoch 4 | Batch 54199/57412 | Loss: 0.289 | time 6376.668070316315 |
| Epoch 4 | Batch 54399/57412 | Loss: 0.258 | time 6400.772945165634 |
| Epoch 4 | Batch 54599/57412 | Loss: 0.320 | time 6424.4567193984985 |
| Epoch 4 | Batch 54799/57412 | Loss: 0.373 | time 6446.847034215927 |
| Epoch 4 | Batch 54999/57412 | Loss: 0.508 | time 6470.591621398926 |
| Epoch 4 | Batch 55199/57412 | Loss: 0.307 | time 6494.510283708572 |
| Epoch 4 | Batch 55399/57412 | Loss: 0.369 | time 6517.595104455948 |
| Epoch 4 | Batch 55599/57412 | Loss: 0.391 | time 6541.369531393051 |
| Epoch 4 | Batch 55799/57412 | Loss: 0.500 | time 6565.771498441696 |
| Epoch 4 | Batch 55999/57412 | Loss: 0.076 | time 6589.267009973526 |
| Epoch 4 | Batch 56199/57412 | Loss: 0.243 | time 6613.8078026771545 |
| Epoch 4 | Batch 56399/57412 | Loss: 0.100 | time 6637.340825796127 |
| Epoch 4 | Batch 56599/57412 | Loss: 0.307 | time 6663.581857919693 |
| Epoch 4 | Batch 56799/57412 | Loss: 0.430 | time 6686.730523586273 |
| Epoch 4 | Batch 56999/57412 | Loss: 0.163 | time 6711.476532697678 |
| Epoch 4 | Batch 57199/57412 | Loss: 0.354 | time 6735.625875711441 |
| Epoch 4 | Batch 57399/57412 | Loss: 0.220 | time 6759.039487123489 |
| ========================================================================================= |
| ========================================================================================= |
| ========================================================================================= |
| Epoch 4 | Validation Samples 5082 | Validation Acc: 0.7160566706021252 | time 320.7283499240875 |
| ========================================================================================= |
| Saving at Epoch 4 | Learning rate: 2e-05 |
|
|
