ViV1T / viv1t_001 /output.log
bryanlimy's picture
rename folders
e148497
Use bfloat16 for core module.
Use parallel attention and MLP in ViViT.
Epoch 001/400
Train loss: 113924008.00 correlation: 0.0107
Validation loss: 200107760.00 correlation: 0.0263
Elapse: 605.46s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 002/400
Train loss: 97896768.00 correlation: 0.0340
Validation loss: 199235856.00 correlation: 0.0378
Elapse: 556.86s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 003/400
Train loss: 96883144.00 correlation: 0.0443
Validation loss: 198767200.00 correlation: 0.0405
Elapse: 560.31s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 004/400
Train loss: 96317000.00 correlation: 0.0499
Validation loss: 198194576.00 correlation: 0.0449
Elapse: 560.78s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 005/400
Train loss: 95592096.00 correlation: 0.0577
Validation loss: 197382576.00 correlation: 0.0493
Elapse: 563.47s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 006/400
Train loss: 94455696.00 correlation: 0.0692
Validation loss: 195917728.00 correlation: 0.0602
Elapse: 565.83s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 007/400
Train loss: 93257144.00 correlation: 0.0821
Validation loss: 193956304.00 correlation: 0.0721
Elapse: 566.60s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 008/400
Train loss: 92014728.00 correlation: 0.0951
Validation loss: 192039424.00 correlation: 0.0839
Elapse: 566.05s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 009/400
Train loss: 91067848.00 correlation: 0.1053
Validation loss: 190701360.00 correlation: 0.0935
Elapse: 563.36s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 010/400
Train loss: 90106464.00 correlation: 0.1149
Validation loss: 189259712.00 correlation: 0.1026
Elapse: 562.44s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 011/400
Train loss: 89183952.00 correlation: 0.1243
Validation loss: 187589920.00 correlation: 0.1130
Elapse: 560.31s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 012/400
Train loss: 88171952.00 correlation: 0.1348
Validation loss: 186131232.00 correlation: 0.1219
Elapse: 559.08s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 013/400
Train loss: 87220088.00 correlation: 0.1442
Validation loss: 184294144.00 correlation: 0.1328
Elapse: 564.57s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 014/400
Train loss: 86305888.00 correlation: 0.1535
Validation loss: 182766432.00 correlation: 0.1425
Elapse: 564.31s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 015/400
Train loss: 85306032.00 correlation: 0.1632
Validation loss: 181567952.00 correlation: 0.1505
Elapse: 562.48s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 016/400
Train loss: 84622736.00 correlation: 0.1701
Validation loss: 180351280.00 correlation: 0.1587
Elapse: 561.75s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 017/400
Train loss: 83839824.00 correlation: 0.1780
Validation loss: 179434864.00 correlation: 0.1644
Elapse: 561.90s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 018/400
Train loss: 83213000.00 correlation: 0.1840
Validation loss: 178816208.00 correlation: 0.1695
Elapse: 562.07s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 019/400
Train loss: 82673560.00 correlation: 0.1895
Validation loss: 178169568.00 correlation: 0.1740
Elapse: 562.05s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 020/400
Train loss: 82081376.00 correlation: 0.1949
Validation loss: 177435600.00 correlation: 0.1787
Elapse: 561.87s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 021/400
Train loss: 81644440.00 correlation: 0.1992
Validation loss: 176873408.00 correlation: 0.1824
Elapse: 562.31s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 022/400
Train loss: 81300944.00 correlation: 0.2027
Validation loss: 176298464.00 correlation: 0.1855
Elapse: 562.57s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 023/400
Train loss: 80864992.00 correlation: 0.2072
Validation loss: 175630304.00 correlation: 0.1905
Elapse: 562.55s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 024/400
Train loss: 80516536.00 correlation: 0.2102
Validation loss: 175099184.00 correlation: 0.1937
Elapse: 561.74s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 025/400
Train loss: 80141312.00 correlation: 0.2140
Validation loss: 174953952.00 correlation: 0.1954
Elapse: 563.63s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 026/400
Train loss: 79914320.00 correlation: 0.2162
Validation loss: 174397456.00 correlation: 0.1975
Elapse: 563.20s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 027/400
Train loss: 79520984.00 correlation: 0.2200
Validation loss: 174134560.00 correlation: 0.2014
Elapse: 562.71s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 028/400
Train loss: 79212208.00 correlation: 0.2229
Validation loss: 173471920.00 correlation: 0.2046
Elapse: 562.79s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 029/400
Train loss: 78973040.00 correlation: 0.2252
Validation loss: 173183520.00 correlation: 0.2064
Elapse: 563.08s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 030/400
Train loss: 78870784.00 correlation: 0.2264
Validation loss: 172775088.00 correlation: 0.2096
Elapse: 562.82s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 031/400
Train loss: 78490200.00 correlation: 0.2296
Validation loss: 172664992.00 correlation: 0.2107
Elapse: 563.05s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 032/400
Train loss: 78317200.00 correlation: 0.2318
Validation loss: 172019824.00 correlation: 0.2132
Elapse: 562.89s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 033/400
Train loss: 78137128.00 correlation: 0.2333
Validation loss: 171897024.00 correlation: 0.2150
Elapse: 562.71s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 034/400
Train loss: 77950032.00 correlation: 0.2354
Validation loss: 171802656.00 correlation: 0.2157
Elapse: 562.68s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 035/400
Train loss: 77766832.00 correlation: 0.2367
Validation loss: 171322160.00 correlation: 0.2180
Elapse: 561.92s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 036/400
Train loss: 77637216.00 correlation: 0.2384
Validation loss: 171034880.00 correlation: 0.2209
Elapse: 562.15s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 037/400
Train loss: 77482304.00 correlation: 0.2397
Validation loss: 171174608.00 correlation: 0.2203
Elapse: 564.93s
Epoch 038/400
Train loss: 77429408.00 correlation: 0.2402
Validation loss: 170815888.00 correlation: 0.2218
Elapse: 564.45s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 039/400
Train loss: 77292792.00 correlation: 0.2412
Validation loss: 170692576.00 correlation: 0.2224
Elapse: 563.94s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 040/400
Train loss: 77122352.00 correlation: 0.2426
Validation loss: 170492064.00 correlation: 0.2256
Elapse: 565.03s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 041/400
Train loss: 77124456.00 correlation: 0.2430
Validation loss: 170532256.00 correlation: 0.2245
Elapse: 564.39s
Epoch 042/400
Train loss: 76964592.00 correlation: 0.2445
Validation loss: 170406688.00 correlation: 0.2257
Elapse: 565.12s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 043/400
Train loss: 76869008.00 correlation: 0.2454
Validation loss: 170592080.00 correlation: 0.2248
Elapse: 563.98s
Epoch 044/400
Train loss: 76798544.00 correlation: 0.2461
Validation loss: 170285664.00 correlation: 0.2253
Elapse: 564.88s
Epoch 045/400
Train loss: 76706736.00 correlation: 0.2469
Validation loss: 170137504.00 correlation: 0.2276
Elapse: 564.78s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 046/400
Train loss: 76661312.00 correlation: 0.2474
Validation loss: 170203360.00 correlation: 0.2258
Elapse: 563.90s
Epoch 047/400
Train loss: 76653904.00 correlation: 0.2479
Validation loss: 169920448.00 correlation: 0.2283
Elapse: 565.79s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 048/400
Train loss: 76510704.00 correlation: 0.2491
Validation loss: 169954384.00 correlation: 0.2279
Elapse: 566.41s
Epoch 049/400
Train loss: 76493416.00 correlation: 0.2492
Validation loss: 169840224.00 correlation: 0.2290
Elapse: 566.13s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 050/400
Train loss: 76429016.00 correlation: 0.2495
Validation loss: 170036560.00 correlation: 0.2272
Elapse: 567.15s
Epoch 051/400
Train loss: 76434184.00 correlation: 0.2500
Validation loss: 169668848.00 correlation: 0.2298
Elapse: 563.78s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 052/400
Train loss: 76368824.00 correlation: 0.2505
Validation loss: 169862640.00 correlation: 0.2292
Elapse: 563.52s
Epoch 053/400
Train loss: 76235120.00 correlation: 0.2516
Validation loss: 169524560.00 correlation: 0.2315
Elapse: 565.24s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 054/400
Train loss: 76236176.00 correlation: 0.2518
Validation loss: 169235696.00 correlation: 0.2324
Elapse: 564.43s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 055/400
Train loss: 76272208.00 correlation: 0.2514
Validation loss: 169566032.00 correlation: 0.2303
Elapse: 564.62s
Epoch 056/400
Train loss: 76095872.00 correlation: 0.2528
Validation loss: 169257008.00 correlation: 0.2329
Elapse: 564.52s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 057/400
Train loss: 76150160.00 correlation: 0.2523
Validation loss: 169306656.00 correlation: 0.2323
Elapse: 564.42s
Epoch 058/400
Train loss: 75971168.00 correlation: 0.2540
Validation loss: 169281824.00 correlation: 0.2326
Elapse: 565.79s
Epoch 059/400
Train loss: 76016632.00 correlation: 0.2537
Validation loss: 169175984.00 correlation: 0.2330
Elapse: 567.53s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 060/400
Train loss: 75961288.00 correlation: 0.2543
Validation loss: 169244832.00 correlation: 0.2312
Elapse: 564.74s
Epoch 061/400
Train loss: 75918368.00 correlation: 0.2545
Validation loss: 169277088.00 correlation: 0.2310
Elapse: 564.73s
Epoch 062/400
Train loss: 75869072.00 correlation: 0.2550
Validation loss: 169328944.00 correlation: 0.2320
Elapse: 564.39s
Epoch 063/400
Train loss: 75762192.00 correlation: 0.2561
Validation loss: 169258112.00 correlation: 0.2316
Elapse: 564.19s
Epoch 064/400
Train loss: 75825328.00 correlation: 0.2556
Validation loss: 169071840.00 correlation: 0.2332
Elapse: 564.93s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 065/400
Train loss: 75713256.00 correlation: 0.2566
Validation loss: 169214000.00 correlation: 0.2329
Elapse: 564.84s
Epoch 066/400
Train loss: 75745160.00 correlation: 0.2561
Validation loss: 169041248.00 correlation: 0.2327
Elapse: 564.78s
Epoch 067/400
Train loss: 75648152.00 correlation: 0.2572
Validation loss: 169051616.00 correlation: 0.2327
Elapse: 565.20s
Epoch 068/400
Train loss: 75614368.00 correlation: 0.2577
Validation loss: 169221120.00 correlation: 0.2327
Elapse: 564.71s
Epoch 069/400
Train loss: 75625776.00 correlation: 0.2574
Validation loss: 169122256.00 correlation: 0.2330
Elapse: 564.46s
Loaded checkpoint from epoch 64 (correlation: 0.2332).
Reduce learning rate of core to 1.4400e-03 (num. reduce: 1).
Reduce learning rate of readouts to 1.0800e-03 (num. reduce: 1).
Reduce learning rate of shifters to 1.0800e-03 (num. reduce: 1).
Epoch 070/400
Train loss: 73759920.00 correlation: 0.2731
Validation loss: 167411552.00 correlation: 0.2444
Elapse: 565.27s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 071/400
Train loss: 73162624.00 correlation: 0.2785
Validation loss: 167276400.00 correlation: 0.2454
Elapse: 565.31s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 072/400
Train loss: 73062080.00 correlation: 0.2797
Validation loss: 167303840.00 correlation: 0.2453
Elapse: 565.70s
Epoch 073/400
Train loss: 72859248.00 correlation: 0.2818
Validation loss: 167279840.00 correlation: 0.2448
Elapse: 566.21s
Epoch 074/400
Train loss: 72867944.00 correlation: 0.2819
Validation loss: 167328304.00 correlation: 0.2445
Elapse: 566.10s
Epoch 075/400
Train loss: 72806064.00 correlation: 0.2823
Validation loss: 167367392.00 correlation: 0.2443
Elapse: 566.39s
Epoch 076/400
Train loss: 72707504.00 correlation: 0.2836
Validation loss: 167176992.00 correlation: 0.2450
Elapse: 566.85s
Loaded checkpoint from epoch 71 (correlation: 0.2454).
Reduce learning rate of core to 4.3200e-04 (num. reduce: 1).
Reduce learning rate of readouts to 3.2400e-04 (num. reduce: 1).
Reduce learning rate of shifters to 3.2400e-04 (num. reduce: 1).
Epoch 077/400
Train loss: 72382144.00 correlation: 0.2856
Validation loss: 166842576.00 correlation: 0.2479
Elapse: 566.50s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 078/400
Train loss: 72078376.00 correlation: 0.2881
Validation loss: 166822496.00 correlation: 0.2483
Elapse: 567.42s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 079/400
Train loss: 71958248.00 correlation: 0.2896
Validation loss: 166738624.00 correlation: 0.2488
Elapse: 567.48s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 080/400
Train loss: 71895880.00 correlation: 0.2900
Validation loss: 166642640.00 correlation: 0.2490
Elapse: 567.47s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 081/400
Train loss: 71789800.00 correlation: 0.2911
Validation loss: 166705248.00 correlation: 0.2491
Elapse: 566.97s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 082/400
Train loss: 71770424.00 correlation: 0.2915
Validation loss: 166628368.00 correlation: 0.2487
Elapse: 567.08s
Epoch 083/400
Train loss: 71742960.00 correlation: 0.2915
Validation loss: 166623744.00 correlation: 0.2491
Elapse: 567.32s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 084/400
Train loss: 71726080.00 correlation: 0.2915
Validation loss: 166593008.00 correlation: 0.2489
Elapse: 566.52s
Epoch 085/400
Train loss: 71608472.00 correlation: 0.2930
Validation loss: 166582448.00 correlation: 0.2494
Elapse: 566.39s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 086/400
Train loss: 71622656.00 correlation: 0.2929
Validation loss: 166651488.00 correlation: 0.2488
Elapse: 567.27s
Epoch 087/400
Train loss: 71550640.00 correlation: 0.2936
Validation loss: 166630768.00 correlation: 0.2490
Elapse: 566.70s
Epoch 088/400
Train loss: 71491864.00 correlation: 0.2941
Validation loss: 166610080.00 correlation: 0.2493
Elapse: 567.65s
Epoch 089/400
Train loss: 71458584.00 correlation: 0.2944
Validation loss: 166546464.00 correlation: 0.2492
Elapse: 566.68s
Epoch 090/400
Train loss: 71462792.00 correlation: 0.2945
Validation loss: 166668320.00 correlation: 0.2492
Elapse: 566.84s
Loaded checkpoint from epoch 85 (correlation: 0.2494).
Reduce learning rate of core to 1.2960e-04 (num. reduce: 1).
Reduce learning rate of readouts to 9.7200e-05 (num. reduce: 1).
Reduce learning rate of shifters to 9.7200e-05 (num. reduce: 1).
Epoch 091/400
Train loss: 71288968.00 correlation: 0.2953
Validation loss: 166494880.00 correlation: 0.2500
Elapse: 567.12s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 092/400
Train loss: 71275904.00 correlation: 0.2958
Validation loss: 166504256.00 correlation: 0.2498
Elapse: 566.80s
Epoch 093/400
Train loss: 71229392.00 correlation: 0.2962
Validation loss: 166490080.00 correlation: 0.2502
Elapse: 567.91s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 094/400
Train loss: 71289360.00 correlation: 0.2955
Validation loss: 166481824.00 correlation: 0.2501
Elapse: 567.35s
Epoch 095/400
Train loss: 71209928.00 correlation: 0.2962
Validation loss: 166463984.00 correlation: 0.2501
Elapse: 567.29s
Epoch 096/400
Train loss: 71155936.00 correlation: 0.2969
Validation loss: 166447504.00 correlation: 0.2502
Elapse: 567.66s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 097/400
Train loss: 71163080.00 correlation: 0.2967
Validation loss: 166476448.00 correlation: 0.2501
Elapse: 567.23s
Epoch 098/400
Train loss: 71090216.00 correlation: 0.2976
Validation loss: 166447952.00 correlation: 0.2502
Elapse: 567.15s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 099/400
Train loss: 71146160.00 correlation: 0.2971
Validation loss: 166436720.00 correlation: 0.2504
Elapse: 566.92s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 100/400
Train loss: 71064016.00 correlation: 0.2977
Validation loss: 166422944.00 correlation: 0.2502
Elapse: 567.03s
Epoch 101/400
Train loss: 71056432.00 correlation: 0.2979
Validation loss: 166403840.00 correlation: 0.2505
Elapse: 567.52s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 102/400
Train loss: 71051576.00 correlation: 0.2978
Validation loss: 166443872.00 correlation: 0.2504
Elapse: 567.13s
Epoch 103/400
Train loss: 71046608.00 correlation: 0.2979
Validation loss: 166479680.00 correlation: 0.2502
Elapse: 566.67s
Epoch 104/400
Train loss: 71038120.00 correlation: 0.2981
Validation loss: 166454112.00 correlation: 0.2501
Elapse: 564.93s
Epoch 105/400
Train loss: 70978192.00 correlation: 0.2988
Validation loss: 166414432.00 correlation: 0.2503
Elapse: 564.91s
Epoch 106/400
Train loss: 70976960.00 correlation: 0.2990
Validation loss: 166474720.00 correlation: 0.2500
Elapse: 566.07s
Loaded checkpoint from epoch 101 (correlation: 0.2505).
Reduce learning rate of core to 3.8880e-05 (num. reduce: 1).
Reduce learning rate of readouts to 2.9160e-05 (num. reduce: 1).
Reduce learning rate of shifters to 2.9160e-05 (num. reduce: 1).
Epoch 107/400
Train loss: 70972624.00 correlation: 0.2986
Validation loss: 166398144.00 correlation: 0.2506
Elapse: 566.75s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 108/400
Train loss: 70964176.00 correlation: 0.2987
Validation loss: 166382512.00 correlation: 0.2506
Elapse: 566.44s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 109/400
Train loss: 70976968.00 correlation: 0.2987
Validation loss: 166404432.00 correlation: 0.2505
Elapse: 567.32s
Epoch 110/400
Train loss: 70951736.00 correlation: 0.2985
Validation loss: 166396896.00 correlation: 0.2506
Elapse: 567.77s
Epoch 111/400
Train loss: 70895944.00 correlation: 0.2997
Validation loss: 166385856.00 correlation: 0.2506
Elapse: 566.89s
Epoch 112/400
Train loss: 70880200.00 correlation: 0.2995
Validation loss: 166380672.00 correlation: 0.2507
Elapse: 566.94s
Checkpoint saved to /home/storage/runs/vivit_ensemble/002/ckpt/model_state.pt.
Epoch 113/400
Train loss: 70861304.00 correlation: 0.2998
Validation loss: 166391184.00 correlation: 0.2506
Elapse: 566.54s
Epoch 114/400
Train loss: 70921912.00 correlation: 0.2989
Validation loss: 166417776.00 correlation: 0.2504
Elapse: 567.26s
Epoch 115/400
Train loss: 70914312.00 correlation: 0.2989
Validation loss: 166403376.00 correlation: 0.2506
Elapse: 566.92s
Epoch 116/400
Train loss: 70903784.00 correlation: 0.2992
Validation loss: 166403360.00 correlation: 0.2506
Elapse: 567.11s
Epoch 117/400
Train loss: 70866896.00 correlation: 0.2994
Validation loss: 166382688.00 correlation: 0.2505
Elapse: 566.94s
Loaded checkpoint from epoch 112 (correlation: 0.2507).
Reduce learning rate of core to 1.1664e-05 (num. reduce: 1).
Reduce learning rate of readouts to 8.7480e-06 (num. reduce: 1).
Reduce learning rate of shifters to 8.7480e-06 (num. reduce: 1).
Epoch 118/400
Train loss: 70926000.00 correlation: 0.2989
Validation loss: 166380768.00 correlation: 0.2506
Elapse: 567.50s
Epoch 119/400
Train loss: 70900808.00 correlation: 0.2993
Validation loss: 166386016.00 correlation: 0.2506
Elapse: 567.35s
Epoch 120/400
Train loss: 70881328.00 correlation: 0.2993
Validation loss: 166387568.00 correlation: 0.2506
Elapse: 567.65s
Epoch 121/400
Train loss: 70937256.00 correlation: 0.2984
Validation loss: 166393568.00 correlation: 0.2506
Elapse: 566.97s
Epoch 122/400
Train loss: 70855888.00 correlation: 0.2996
Validation loss: 166378592.00 correlation: 0.2507
Elapse: 567.50s
Loaded checkpoint from epoch 112 (correlation: 0.2507).
Reduce learning rate of core to 3.4992e-06 (num. reduce: 2).
Reduce learning rate of readouts to 2.6244e-06 (num. reduce: 2).
Reduce learning rate of shifters to 2.6244e-06 (num. reduce: 2).
Epoch 123/400
Train loss: 70938496.00 correlation: 0.2988
Validation loss: 166375792.00 correlation: 0.2507
Elapse: 567.50s
Epoch 124/400
Train loss: 70966320.00 correlation: 0.2986
Validation loss: 166383664.00 correlation: 0.2506
Elapse: 567.77s
Epoch 125/400
Train loss: 70872080.00 correlation: 0.2990
Validation loss: 166383328.00 correlation: 0.2506
Elapse: 567.87s
Epoch 126/400
Train loss: 70866264.00 correlation: 0.2997
Validation loss: 166383136.00 correlation: 0.2506
Elapse: 567.21s
Epoch 127/400
Train loss: 70836352.00 correlation: 0.2996
Validation loss: 166386816.00 correlation: 0.2506
Elapse: 566.90s
Model has not improved after 2 LR reductions.
Loaded checkpoint from epoch 112 (correlation: 0.2507).
ValidationA: 0.2487 B: 0.2789 C: 0.2720 D: 0.2341 E: 0.2367 F: 0.2340 G: 0.2555 H: 0.2351 I: 0.2533 J: 0.2589 average: 0.2507
Results saved to /home/storage/runs/vivit_ensemble/002.