BrianatCambridge's picture
Upload folder using huggingface_hub
20e46c9 verified
Epoch 0 | Batch 1999/57412 | Loss: 0.396 | time 254.98479223251343
Epoch 0 | Batch 3999/57412 | Loss: 0.598 | time 501.5320391654968
Epoch 0 | Batch 5999/57412 | Loss: 0.412 | time 750.8397879600525
Epoch 0 | Batch 7999/57412 | Loss: 0.531 | time 995.1165738105774
Epoch 0 | Batch 9999/57412 | Loss: 0.520 | time 1229.0222663879395
Epoch 0 | Batch 11999/57412 | Loss: 0.254 | time 1469.9271094799042
Epoch 0 | Batch 13999/57412 | Loss: 0.196 | time 1717.8649022579193
Epoch 0 | Batch 15999/57412 | Loss: 0.377 | time 1954.6233386993408
Epoch 0 | Batch 17999/57412 | Loss: 0.139 | time 2201.718013048172
Epoch 0 | Batch 19999/57412 | Loss: 0.463 | time 2447.0384137630463
Saving at Step 20000
Epoch 0 | Batch 21999/57412 | Loss: 0.535 | time 2681.026620388031
Epoch 0 | Batch 23999/57412 | Loss: 0.824 | time 2921.255893945694
Epoch 0 | Batch 25999/57412 | Loss: 0.301 | time 3162.3500986099243
Epoch 0 | Batch 27999/57412 | Loss: 0.410 | time 3391.048465490341
Epoch 0 | Batch 29999/57412 | Loss: 0.424 | time 3623.9968135356903
Epoch 0 | Batch 31999/57412 | Loss: 0.396 | time 3860.8973116874695
Epoch 0 | Batch 33999/57412 | Loss: 0.295 | time 4091.862708568573
Epoch 0 | Batch 35999/57412 | Loss: 0.307 | time 4324.332058429718
Epoch 0 | Batch 37999/57412 | Loss: 0.617 | time 4553.862748861313
Epoch 0 | Batch 39999/57412 | Loss: 0.430 | time 4785.509864807129
Saving at Step 40000
Epoch 0 | Batch 41999/57412 | Loss: 0.504 | time 5022.830712080002
Epoch 0 | Batch 43999/57412 | Loss: 0.477 | time 5255.034224510193
Epoch 0 | Batch 45999/57412 | Loss: 0.414 | time 5486.204306602478
Epoch 0 | Batch 47999/57412 | Loss: 0.428 | time 5717.4952936172485
Epoch 0 | Batch 49999/57412 | Loss: 0.477 | time 5952.2831490039825
Epoch 0 | Batch 51999/57412 | Loss: 0.531 | time 6181.074133634567
Epoch 0 | Batch 53999/57412 | Loss: 0.393 | time 6419.50391626358
Epoch 0 | Batch 55999/57412 | Loss: 0.359 | time 6664.515767812729
=========================================================================================
=========================================================================================
=========================================================================================
Epoch 0 | Validation Samples 5082 | Validation Acc: 0.4183392365210547 | time 332.23271918296814
=========================================================================================
Saving at Epoch 0 | Learning rate: 2e-05
Epoch 1 | Batch 1999/57412 | Loss: 0.363 | time 232.890930891037
Epoch 1 | Batch 3999/57412 | Loss: 0.465 | time 466.0495946407318
Epoch 1 | Batch 5999/57412 | Loss: 0.412 | time 708.206280708313
Epoch 1 | Batch 7999/57412 | Loss: 0.523 | time 942.3334040641785
Epoch 1 | Batch 9999/57412 | Loss: 0.330 | time 1178.746410369873
Epoch 1 | Batch 11999/57412 | Loss: 0.404 | time 1416.6608390808105
Epoch 1 | Batch 13999/57412 | Loss: 0.715 | time 1654.1416494846344
Epoch 1 | Batch 15999/57412 | Loss: 0.719 | time 1889.1311526298523
Epoch 1 | Batch 17999/57412 | Loss: 0.475 | time 2125.462559223175
Epoch 1 | Batch 19999/57412 | Loss: 0.177 | time 2361.2632160186768
Saving at Step 20000
Epoch 1 | Batch 21999/57412 | Loss: 0.367 | time 2603.959163427353
Epoch 1 | Batch 23999/57412 | Loss: 0.605 | time 2841.8973412513733
Epoch 1 | Batch 25999/57412 | Loss: 0.211 | time 3078.931995868683
Epoch 1 | Batch 27999/57412 | Loss: 0.250 | time 3316.8245265483856
Epoch 1 | Batch 29999/57412 | Loss: 0.586 | time 3553.8761093616486
Epoch 1 | Batch 31999/57412 | Loss: 0.455 | time 3786.8724162578583
Epoch 1 | Batch 33999/57412 | Loss: 0.412 | time 4023.375405073166
Epoch 1 | Batch 35999/57412 | Loss: 0.336 | time 4255.024896144867
Epoch 1 | Batch 37999/57412 | Loss: 0.418 | time 4491.783854007721
Epoch 1 | Batch 39999/57412 | Loss: 0.570 | time 4726.813305139542
Saving at Step 40000
Epoch 1 | Batch 41999/57412 | Loss: 0.625 | time 4969.020828008652
Epoch 1 | Batch 43999/57412 | Loss: 0.637 | time 5207.284041404724
Epoch 1 | Batch 45999/57412 | Loss: 0.332 | time 5439.013684988022
Epoch 1 | Batch 47999/57412 | Loss: 0.228 | time 5669.40888094902
Epoch 1 | Batch 49999/57412 | Loss: 0.414 | time 5904.615941286087
Epoch 1 | Batch 51999/57412 | Loss: 0.396 | time 6133.846004247665
Epoch 1 | Batch 53999/57412 | Loss: 0.422 | time 6365.930061101913
Epoch 1 | Batch 55999/57412 | Loss: 0.516 | time 6590.622798681259
=========================================================================================
=========================================================================================
=========================================================================================
Epoch 1 | Validation Samples 5082 | Validation Acc: 0.4594647776465958 | time 331.0848898887634
=========================================================================================
Saving at Epoch 1 | Learning rate: 2e-05
Epoch 2 | Batch 1999/57412 | Loss: 0.516 | time 228.50831747055054
Epoch 2 | Batch 3999/57412 | Loss: 0.275 | time 459.5048680305481
Epoch 2 | Batch 5999/57412 | Loss: 0.594 | time 687.5822954177856
Epoch 2 | Batch 7999/57412 | Loss: 0.520 | time 918.0298216342926
Epoch 2 | Batch 9999/57412 | Loss: 0.227 | time 1152.4581677913666
Epoch 2 | Batch 11999/57412 | Loss: 0.436 | time 1387.710283756256
Epoch 2 | Batch 13999/57412 | Loss: 0.482 | time 1626.314659833908
Epoch 2 | Batch 15999/57412 | Loss: 0.605 | time 1857.9773828983307
Epoch 2 | Batch 17999/57412 | Loss: 0.469 | time 2088.4553101062775
Epoch 2 | Batch 19999/57412 | Loss: 0.316 | time 2319.654955148697
Saving at Step 20000
Epoch 2 | Batch 21999/57412 | Loss: 0.350 | time 2568.6353709697723
Epoch 2 | Batch 23999/57412 | Loss: 0.371 | time 2803.5683188438416
Epoch 2 | Batch 25999/57412 | Loss: 0.379 | time 3040.185362815857
Epoch 2 | Batch 27999/57412 | Loss: 0.398 | time 3273.572656393051
Epoch 2 | Batch 29999/57412 | Loss: 0.400 | time 3508.264441013336
Epoch 2 | Batch 31999/57412 | Loss: 0.754 | time 3739.2418954372406
Epoch 2 | Batch 33999/57412 | Loss: 0.543 | time 3967.3613052368164
Epoch 2 | Batch 35999/57412 | Loss: 0.307 | time 4198.480511665344
Epoch 2 | Batch 37999/57412 | Loss: 0.316 | time 4441.841189146042
Epoch 2 | Batch 39999/57412 | Loss: 0.250 | time 4690.8165464401245
Saving at Step 40000
Epoch 2 | Batch 41999/57412 | Loss: 0.449 | time 4937.942239999771
Epoch 2 | Batch 43999/57412 | Loss: 0.201 | time 5178.583641767502
Epoch 2 | Batch 45999/57412 | Loss: 0.387 | time 5415.440551519394
Epoch 2 | Batch 47999/57412 | Loss: 0.149 | time 5657.493630886078
Epoch 2 | Batch 49999/57412 | Loss: 0.312 | time 5897.974587440491
Epoch 2 | Batch 51999/57412 | Loss: 0.220 | time 6128.398864030838
Epoch 2 | Batch 53999/57412 | Loss: 0.684 | time 6358.333668231964
Epoch 2 | Batch 55999/57412 | Loss: 0.621 | time 6586.893479824066
=========================================================================================
=========================================================================================
=========================================================================================
Epoch 2 | Validation Samples 5082 | Validation Acc: 0.4822904368358914 | time 319.2647862434387
=========================================================================================
Saving at Epoch 2 | Learning rate: 2e-05
Epoch 3 | Batch 1999/57412 | Loss: 0.465 | time 235.30760025978088
Epoch 3 | Batch 3999/57412 | Loss: 0.197 | time 465.1452827453613
Epoch 3 | Batch 5999/57412 | Loss: 0.365 | time 704.030993938446
Epoch 3 | Batch 7999/57412 | Loss: 0.046 | time 940.4018654823303
Epoch 3 | Batch 9999/57412 | Loss: 0.328 | time 1179.0897731781006
Epoch 3 | Batch 11999/57412 | Loss: 0.496 | time 1415.592700958252
Epoch 3 | Batch 13999/57412 | Loss: 0.252 | time 1650.0549399852753
Epoch 3 | Batch 15999/57412 | Loss: 0.149 | time 1883.0535883903503
Epoch 3 | Batch 17999/57412 | Loss: 0.479 | time 2116.9367537498474
Epoch 3 | Batch 19999/57412 | Loss: 0.527 | time 2357.039795398712
Saving at Step 20000
Epoch 3 | Batch 21999/57412 | Loss: 0.193 | time 2592.145604133606
Epoch 3 | Batch 23999/57412 | Loss: 0.373 | time 2825.750539779663
Epoch 3 | Batch 25999/57412 | Loss: 0.324 | time 3063.385856151581
Epoch 3 | Batch 27999/57412 | Loss: 0.543 | time 3301.0782577991486
Epoch 3 | Batch 29999/57412 | Loss: 0.208 | time 3536.4568316936493
Epoch 3 | Batch 31999/57412 | Loss: 0.383 | time 3766.9469695091248
Epoch 3 | Batch 33999/57412 | Loss: 0.248 | time 4001.1584606170654
Epoch 3 | Batch 35999/57412 | Loss: 0.195 | time 4238.465028762817
Epoch 3 | Batch 37999/57412 | Loss: 0.346 | time 4475.129605770111
Epoch 3 | Batch 39999/57412 | Loss: 0.273 | time 4705.53656744957
Saving at Step 40000
Epoch 3 | Batch 41999/57412 | Loss: 0.469 | time 4949.908352136612
Epoch 3 | Batch 43999/57412 | Loss: 0.074 | time 5190.864463329315
Epoch 3 | Batch 45999/57412 | Loss: 0.352 | time 5427.7940464019775
Epoch 3 | Batch 47999/57412 | Loss: 0.342 | time 5659.307080745697
Epoch 3 | Batch 49999/57412 | Loss: 0.434 | time 5888.111845731735
Epoch 3 | Batch 51999/57412 | Loss: 0.168 | time 6111.985684394836
Epoch 3 | Batch 53999/57412 | Loss: 0.316 | time 6335.961695194244
Epoch 3 | Batch 55999/57412 | Loss: 0.328 | time 6566.0662751197815
=========================================================================================
=========================================================================================
=========================================================================================
Epoch 3 | Validation Samples 5082 | Validation Acc: 0.5556867375049194 | time 311.10527992248535
=========================================================================================
Saving at Epoch 3 | Learning rate: 2e-05
Epoch 4 | Batch 1999/57412 | Loss: 0.486 | time 230.6996374130249
Epoch 4 | Batch 3999/57412 | Loss: 0.404 | time 459.80813455581665
Epoch 4 | Batch 5999/57412 | Loss: 0.158 | time 700.2022352218628
Epoch 4 | Batch 7999/57412 | Loss: 0.621 | time 926.9909446239471
Epoch 4 | Batch 9999/57412 | Loss: 0.447 | time 1161.9302620887756
Epoch 4 | Batch 11999/57412 | Loss: 0.320 | time 1399.862356185913
Epoch 4 | Batch 13999/57412 | Loss: 0.439 | time 1631.9697835445404
Epoch 4 | Batch 15999/57412 | Loss: 0.237 | time 1863.7826466560364
Epoch 4 | Batch 17999/57412 | Loss: 0.414 | time 2093.7483241558075
Epoch 4 | Batch 19999/57412 | Loss: 0.562 | time 2325.1370437145233
Saving at Step 20000
Epoch 4 | Batch 21999/57412 | Loss: 0.467 | time 2575.4963347911835
Epoch 4 | Batch 23999/57412 | Loss: 0.424 | time 2809.878782272339
Epoch 4 | Batch 25999/57412 | Loss: 0.398 | time 3044.8267843723297
Epoch 4 | Batch 27999/57412 | Loss: 0.375 | time 3277.2568085193634
Epoch 4 | Batch 29999/57412 | Loss: 0.202 | time 3512.824246406555
Epoch 4 | Batch 31999/57412 | Loss: 0.354 | time 3745.535721063614
Epoch 4 | Batch 33999/57412 | Loss: 0.201 | time 3976.723613023758
Epoch 4 | Batch 35999/57412 | Loss: 0.439 | time 4207.477236509323
Epoch 4 | Batch 37999/57412 | Loss: 0.307 | time 4436.664123535156
Epoch 4 | Batch 39999/57412 | Loss: 0.586 | time 4668.947089672089
Saving at Step 40000
Epoch 4 | Batch 41999/57412 | Loss: 0.299 | time 4914.486303567886
Epoch 4 | Batch 43999/57412 | Loss: 0.543 | time 5150.6524221897125
Epoch 4 | Batch 45999/57412 | Loss: 0.461 | time 5397.079828500748
Epoch 4 | Batch 47999/57412 | Loss: 0.406 | time 5641.841084957123
Epoch 4 | Batch 49999/57412 | Loss: 0.220 | time 5877.933323144913
Epoch 4 | Batch 51999/57412 | Loss: 0.393 | time 6122.285271883011
Epoch 4 | Batch 53999/57412 | Loss: 0.408 | time 6359.644855499268
Epoch 4 | Batch 55999/57412 | Loss: 0.355 | time 6599.34525179863
=========================================================================================
=========================================================================================
=========================================================================================
Epoch 4 | Validation Samples 5082 | Validation Acc: 0.6044864226682408 | time 336.8937237262726
=========================================================================================
Saving at Epoch 4 | Learning rate: 2e-05