| /athenahomes/gabrijel/miniconda3/envs/track-generator/lib/python3.11/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: '/athenahomes/gabrijel/miniconda3/envs/track-generator/lib/python3.11/site-packages/torchvision/image.so: undefined symbol: _ZN3c1017RegisterOperatorsD1Ev'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source? |
| warn( |
| Schedule: cosine |
| Cfg: False |
| Output path: /scratch/shared/beegfs/gabrijel/m2l/mini |
| Patch Size: 4 |
| Device: cuda:3 |
| ===================================================================================== |
| Layer (type:depth-idx) Param # |
| ===================================================================================== |
| DiT 18,816 |
| ├─PatchEmbed: 1-1 -- |
| │ └─Conv2d: 2-1 6,528 |
| ├─TimestepEmbedder: 1-2 -- |
| │ └─Mlp: 2-2 -- |
| │ │ └─Linear: 3-1 98,688 |
| │ │ └─SiLU: 3-2 -- |
| │ │ └─Linear: 3-3 147,840 |
| ├─ModuleList: 1-3 -- |
| │ └─DiTBlock: 2-3 -- |
| │ │ └─LayerNorm: 3-4 -- |
| │ │ └─MultiheadAttention: 3-5 591,360 |
| │ │ └─LayerNorm: 3-6 -- |
| │ │ └─Mlp: 3-7 1,181,568 |
| │ │ └─Sequential: 3-8 887,040 |
| │ └─DiTBlock: 2-4 -- |
| │ │ └─LayerNorm: 3-9 -- |
| │ │ └─MultiheadAttention: 3-10 591,360 |
| │ │ └─LayerNorm: 3-11 -- |
| │ │ └─Mlp: 3-12 1,181,568 |
| │ │ └─Sequential: 3-13 887,040 |
| │ └─DiTBlock: 2-5 -- |
| │ │ └─LayerNorm: 3-14 -- |
| │ │ └─MultiheadAttention: 3-15 591,360 |
| │ │ └─LayerNorm: 3-16 -- |
| │ │ └─Mlp: 3-17 1,181,568 |
| │ │ └─Sequential: 3-18 887,040 |
| │ └─DiTBlock: 2-6 -- |
| │ │ └─LayerNorm: 3-19 -- |
| │ │ └─MultiheadAttention: 3-20 591,360 |
| │ │ └─LayerNorm: 3-21 -- |
| │ │ └─Mlp: 3-22 1,181,568 |
| │ │ └─Sequential: 3-23 887,040 |
| │ └─DiTBlock: 2-7 -- |
| │ │ └─LayerNorm: 3-24 -- |
| │ │ └─MultiheadAttention: 3-25 591,360 |
| │ │ └─LayerNorm: 3-26 -- |
| │ │ └─Mlp: 3-27 1,181,568 |
| │ │ └─Sequential: 3-28 887,040 |
| │ └─DiTBlock: 2-8 -- |
| │ │ └─LayerNorm: 3-29 -- |
| │ │ └─MultiheadAttention: 3-30 591,360 |
| │ │ └─LayerNorm: 3-31 -- |
| │ │ └─Mlp: 3-32 1,181,568 |
| │ │ └─Sequential: 3-33 887,040 |
| ├─FinalLayer: 1-4 -- |
| │ └─LayerNorm: 2-9 -- |
| │ └─Linear: 2-10 6,160 |
| │ └─Sequential: 2-11 -- |
| │ │ └─SiLU: 3-34 -- |
| │ │ └─Linear: 3-35 295,680 |
| ├─Unpatchify: 1-5 -- |
| ===================================================================================== |
| Total params: 16,533,520 |
| Trainable params: 16,514,704 |
| Non-trainable params: 18,816 |
| ===================================================================================== |
|
|
| EPOCH: 1 |
| Loss at step 0: 1.003037691116333 |
| Loss at step 50: 0.27661845088005066 |
| Loss at step 100: 0.1424816995859146 |
| Loss at step 150: 0.149506077170372 |
| Loss at step 200: 0.12479458749294281 |
| Loss at step 250: 0.11426719278097153 |
| Loss at step 300: 0.1121714636683464 |
| Loss at step 350: 0.1525452435016632 |
| Loss at step 400: 0.13195544481277466 |
| Loss at step 450: 0.13427531719207764 |
| Loss at step 500: 0.11618398129940033 |
| Loss at step 550: 0.10502675175666809 |
| Loss at step 600: 0.0915534719824791 |
| Loss at step 650: 0.09573911130428314 |
| Loss at step 700: 0.1133173331618309 |
| Loss at step 750: 0.1329326033592224 |
| Loss at step 800: 0.10761799663305283 |
| Loss at step 850: 0.11455381661653519 |
| Loss at step 900: 0.12381163984537125 |
| Mean training loss after epoch 1: 0.1539469359716627 |
|
|
| EPOCH: 2 |
| Loss at step 0: 0.10565946251153946 |
| Loss at step 50: 0.10705439746379852 |
| Loss at step 100: 0.10383749008178711 |
| Loss at step 150: 0.06799700856208801 |
| Loss at step 200: 0.10335274040699005 |
| Loss at step 250: 0.06992191821336746 |
| Loss at step 300: 0.08639557659626007 |
| Loss at step 350: 0.075557179749012 |
| Loss at step 400: 0.08633825182914734 |
| Loss at step 450: 0.07788940519094467 |
| Loss at step 500: 0.07782382518053055 |
| Loss at step 550: 0.08034727722406387 |
| Loss at step 600: 0.08342839032411575 |
| Loss at step 650: 0.07229728251695633 |
| Loss at step 700: 0.1125161275267601 |
| Loss at step 750: 0.07861600071191788 |
| Loss at step 800: 0.07409120351076126 |
| Loss at step 850: 0.078849658370018 |
| Loss at step 900: 0.094577357172966 |
| Mean training loss after epoch 2: 0.0890754868369748 |
|
|
| EPOCH: 3 |
| Loss at step 0: 0.09418325871229172 |
| Loss at step 50: 0.07725059241056442 |
| Loss at step 100: 0.07888046652078629 |
| Loss at step 150: 0.07631003856658936 |
| Loss at step 200: 0.08261995762586594 |
| Loss at step 250: 0.08296551555395126 |
| Loss at step 300: 0.06751654297113419 |
| Loss at step 350: 0.056437570601701736 |
| Loss at step 400: 0.08921732008457184 |
| Loss at step 450: 0.07375114411115646 |
| Loss at step 500: 0.0711858719587326 |
| Loss at step 550: 0.07505157589912415 |
| Loss at step 600: 0.08232348412275314 |
| Loss at step 650: 0.06679479032754898 |
| Loss at step 700: 0.08148904889822006 |
| Loss at step 750: 0.08080136030912399 |
| Loss at step 800: 0.08312830328941345 |
| Loss at step 850: 0.07133378833532333 |
| Loss at step 900: 0.09072605520486832 |
| Mean training loss after epoch 3: 0.0814713569544653 |
|
|
| EPOCH: 4 |
| Loss at step 0: 0.07232233136892319 |
| Loss at step 50: 0.07750030606985092 |
| Loss at step 100: 0.07092481851577759 |
| Loss at step 150: 0.07385428249835968 |
| Loss at step 200: 0.0765489712357521 |
| Loss at step 250: 0.08501236885786057 |
| Loss at step 300: 0.0753922238945961 |
| Loss at step 350: 0.09025567770004272 |
| Loss at step 400: 0.07097692042589188 |
| Loss at step 450: 0.07708340883255005 |
| Loss at step 500: 0.09128562361001968 |
| Loss at step 550: 0.06018142029643059 |
| Loss at step 600: 0.09149488806724548 |
| Loss at step 650: 0.06561129540205002 |
| Loss at step 700: 0.06980118900537491 |
| Loss at step 750: 0.08550712466239929 |
| Loss at step 800: 0.09240806102752686 |
| Loss at step 850: 0.07110103964805603 |
| Loss at step 900: 0.07148086279630661 |
| Mean training loss after epoch 4: 0.0787551273812236 |
|
|
| EPOCH: 5 |
| Loss at step 0: 0.06837005913257599 |
| Loss at step 50: 0.08118490874767303 |
| Loss at step 100: 0.07443185895681381 |
| Loss at step 150: 0.07052610069513321 |
| Loss at step 200: 0.06819907575845718 |
| Loss at step 250: 0.07496345043182373 |
| Loss at step 300: 0.06800892949104309 |
| Loss at step 350: 0.06302531063556671 |
| Loss at step 400: 0.08613581210374832 |
| Loss at step 450: 0.08716695010662079 |
| Loss at step 500: 0.06726646423339844 |
| Loss at step 550: 0.08373793959617615 |
| Loss at step 600: 0.06744256615638733 |
| Loss at step 650: 0.06358864903450012 |
| Loss at step 700: 0.07609168440103531 |
| Loss at step 750: 0.08561772853136063 |
| Loss at step 800: 0.08098775148391724 |
| Loss at step 850: 0.06596729904413223 |
| Loss at step 900: 0.09818398952484131 |
| Mean training loss after epoch 5: 0.07621333939132532 |
|
|
| EPOCH: 6 |
| Loss at step 0: 0.07262420654296875 |
| Loss at step 50: 0.08250504732131958 |
| Loss at step 100: 0.0771770104765892 |
| Loss at step 150: 0.10342646390199661 |
| Loss at step 200: 0.058930426836013794 |
| Loss at step 250: 0.07842208445072174 |
| Loss at step 300: 0.07260201871395111 |
| Loss at step 350: 0.08667268604040146 |
| Loss at step 400: 0.07582581788301468 |
| Loss at step 450: 0.06609180569648743 |
| Loss at step 500: 0.06838462501764297 |
| Loss at step 550: 0.0686202421784401 |
| Loss at step 600: 0.07450604438781738 |
| Loss at step 650: 0.06995474547147751 |
| Loss at step 700: 0.09052643924951553 |
| Loss at step 750: 0.06410960853099823 |
| Loss at step 800: 0.06000955030322075 |
| Loss at step 850: 0.058537937700748444 |
| Loss at step 900: 0.05873030051589012 |
| Mean training loss after epoch 6: 0.07148564022296527 |
|
|
| EPOCH: 7 |
| Loss at step 0: 0.08014527708292007 |
| Loss at step 50: 0.06883670389652252 |
| Loss at step 100: 0.05498388782143593 |
| Loss at step 150: 0.049575336277484894 |
| Loss at step 200: 0.07284802198410034 |
| Loss at step 250: 0.05627802759408951 |
| Loss at step 300: 0.08170035481452942 |
| Loss at step 350: 0.07256447523832321 |
| Loss at step 400: 0.0588550940155983 |
| Loss at step 450: 0.07103971391916275 |
| Loss at step 500: 0.07324987649917603 |
| Loss at step 550: 0.06593427062034607 |
| Loss at step 600: 0.0544898696243763 |
| Loss at step 650: 0.07234711199998856 |
| Loss at step 700: 0.0544438362121582 |
| Loss at step 750: 0.057596754282712936 |
| Loss at step 800: 0.05363701656460762 |
| Loss at step 850: 0.06061040610074997 |
| Loss at step 900: 0.050478413701057434 |
| Mean training loss after epoch 7: 0.061715417988360055 |
|
|
| EPOCH: 8 |
| Loss at step 0: 0.06242280825972557 |
| Loss at step 50: 0.06322493404150009 |
| Loss at step 100: 0.06131700053811073 |
| Loss at step 150: 0.0666111558675766 |
| Loss at step 200: 0.05284688621759415 |
| Loss at step 250: 0.05521265044808388 |
| Loss at step 300: 0.0472453273832798 |
| Loss at step 350: 0.05958862975239754 |
| Loss at step 400: 0.05755053088068962 |
| Loss at step 450: 0.05801082402467728 |
| Loss at step 500: 0.05060916021466255 |
| Loss at step 550: 0.07028011977672577 |
| Loss at step 600: 0.06796812266111374 |
| Loss at step 650: 0.0757439136505127 |
| Loss at step 700: 0.07297901809215546 |
| Loss at step 750: 0.04786565154790878 |
| Loss at step 800: 0.04735714942216873 |
| Loss at step 850: 0.050904281437397 |
| Loss at step 900: 0.06405418366193771 |
| Mean training loss after epoch 8: 0.05874664374569586 |
|
|
| EPOCH: 9 |
| Loss at step 0: 0.05896533280611038 |
| Loss at step 50: 0.045270130038261414 |
| Loss at step 100: 0.06076472997665405 |
| Loss at step 150: 0.06235165521502495 |
| Loss at step 200: 0.05219578742980957 |
| Loss at step 250: 0.08207492530345917 |
| Loss at step 300: 0.0685364380478859 |
| Loss at step 350: 0.05931883305311203 |
| Loss at step 400: 0.08336658030748367 |
| Loss at step 450: 0.0543605200946331 |
| Loss at step 500: 0.06624364852905273 |
| Loss at step 550: 0.05275069177150726 |
| Loss at step 600: 0.08901050686836243 |
| Loss at step 650: 0.059423163533210754 |
| Loss at step 700: 0.05075065791606903 |
| Loss at step 750: 0.06457599997520447 |
| Loss at step 800: 0.07323766499757767 |
| Loss at step 850: 0.05342152714729309 |
| Loss at step 900: 0.048495594412088394 |
| Mean training loss after epoch 9: 0.057180338576912626 |
|
|
| EPOCH: 10 |
| Loss at step 0: 0.05593691021203995 |
| Loss at step 50: 0.0404839850962162 |
| Loss at step 100: 0.05115879699587822 |
| Loss at step 150: 0.06267639249563217 |
| Loss at step 200: 0.05636795982718468 |
| Loss at step 250: 0.04891287907958031 |
| Loss at step 300: 0.05107974261045456 |
| Loss at step 350: 0.04856284707784653 |
| Loss at step 400: 0.04412970319390297 |
| Loss at step 450: 0.060254551470279694 |
| Loss at step 500: 0.051702968776226044 |
| Loss at step 550: 0.05409117415547371 |
| Loss at step 600: 0.051070645451545715 |
| Loss at step 650: 0.052753277122974396 |
| Loss at step 700: 0.047309085726737976 |
| Loss at step 750: 0.06630449742078781 |
| Loss at step 800: 0.05757919326424599 |
| Loss at step 850: 0.06830720603466034 |
| Loss at step 900: 0.06392861902713776 |
| Mean training loss after epoch 10: 0.055850355824364276 |
|
|
| EPOCH: 11 |
| Loss at step 0: 0.05673862248659134 |
| Loss at step 50: 0.07288111001253128 |
| Loss at step 100: 0.055279891937971115 |
| Loss at step 150: 0.0829085037112236 |
| Loss at step 200: 0.05173470079898834 |
| Loss at step 250: 0.07002110034227371 |
| Loss at step 300: 0.041786838322877884 |
| Loss at step 350: 0.05969792976975441 |
| Loss at step 400: 0.05089668184518814 |
| Loss at step 450: 0.05513685569167137 |
| Loss at step 500: 0.06826495379209518 |
| Loss at step 550: 0.050237782299518585 |
| Loss at step 600: 0.045402273535728455 |
| Loss at step 650: 0.037458840757608414 |
| Loss at step 700: 0.047057949006557465 |
| Loss at step 750: 0.04279854893684387 |
| Loss at step 800: 0.05576750636100769 |
| Loss at step 850: 0.06578527390956879 |
| Loss at step 900: 0.053406111896038055 |
| Mean training loss after epoch 11: 0.05481477373682741 |
|
|
| EPOCH: 12 |
| Loss at step 0: 0.07697451114654541 |
| Loss at step 50: 0.07025545835494995 |
| Loss at step 100: 0.04223531857132912 |
| Loss at step 150: 0.042284440249204636 |
| Loss at step 200: 0.05219224467873573 |
| Loss at step 250: 0.04662042483687401 |
| Loss at step 300: 0.04564218968153 |
| Loss at step 350: 0.05778402090072632 |
| Loss at step 400: 0.04940832406282425 |
| Loss at step 450: 0.06825084984302521 |
| Loss at step 500: 0.047397445887327194 |
| Loss at step 550: 0.05655599385499954 |
| Loss at step 600: 0.038294047117233276 |
| Loss at step 650: 0.05758621171116829 |
| Loss at step 700: 0.05851387232542038 |
| Loss at step 750: 0.04812590777873993 |
| Loss at step 800: 0.05467564985156059 |
| Loss at step 850: 0.03674978390336037 |
| Loss at step 900: 0.048017699271440506 |
| Mean training loss after epoch 12: 0.05442187320719014 |
|
|
| EPOCH: 13 |
| Loss at step 0: 0.05527647212147713 |
| Loss at step 50: 0.05649548023939133 |
| Loss at step 100: 0.043758414685726166 |
| Loss at step 150: 0.03759016841650009 |
| Loss at step 200: 0.04890500754117966 |
| Loss at step 250: 0.05938975140452385 |
| Loss at step 300: 0.06087874621152878 |
| Loss at step 350: 0.05134967714548111 |
| Loss at step 400: 0.05763311684131622 |
| Loss at step 450: 0.055892378091812134 |
| Loss at step 500: 0.0512239933013916 |
| Loss at step 550: 0.04450003057718277 |
| Loss at step 600: 0.04901410639286041 |
| Loss at step 650: 0.05313221737742424 |
| Loss at step 700: 0.05558817461133003 |
| Loss at step 750: 0.0535665825009346 |
| Loss at step 800: 0.08542494475841522 |
| Loss at step 850: 0.05859728157520294 |
| Loss at step 900: 0.06013469025492668 |
| Mean training loss after epoch 13: 0.05281311084530247 |
|
|
| EPOCH: 14 |
| Loss at step 0: 0.0647217184305191 |
| Loss at step 50: 0.05076614022254944 |
| Loss at step 100: 0.03850740194320679 |
| Loss at step 150: 0.055288080126047134 |
| Loss at step 200: 0.040912121534347534 |
| Loss at step 250: 0.04788177087903023 |
| Loss at step 300: 0.06873010843992233 |
| Loss at step 350: 0.05255446210503578 |
| Loss at step 400: 0.0451323427259922 |
| Loss at step 450: 0.04819340631365776 |
| Loss at step 500: 0.059297483414411545 |
| Loss at step 550: 0.045939523726701736 |
| Loss at step 600: 0.06561025977134705 |
| Loss at step 650: 0.04447270184755325 |
| Loss at step 700: 0.06097499653697014 |
| Loss at step 750: 0.059666190296411514 |
| Loss at step 800: 0.04177537560462952 |
| Loss at step 850: 0.0449681282043457 |
| Loss at step 900: 0.05002983286976814 |
| Mean training loss after epoch 14: 0.052944248951256656 |
|
|
| EPOCH: 15 |
| Loss at step 0: 0.037823475897312164 |
| Loss at step 50: 0.07952471077442169 |
| Loss at step 100: 0.04094187170267105 |
| Loss at step 150: 0.060179922729730606 |
| Loss at step 200: 0.04950674995779991 |
| Loss at step 250: 0.06051012873649597 |
| Loss at step 300: 0.051027752459049225 |
| Loss at step 350: 0.08003382384777069 |
| Loss at step 400: 0.054900161921978 |
| Loss at step 450: 0.03677508607506752 |
| Loss at step 500: 0.07743576914072037 |
| Loss at step 550: 0.05444803088903427 |
| Loss at step 600: 0.05422133579850197 |
| Loss at step 650: 0.04789598658680916 |
| Loss at step 700: 0.043794967234134674 |
| Loss at step 750: 0.04442184790968895 |
| Loss at step 800: 0.04888957738876343 |
| Loss at step 850: 0.03554798662662506 |
| Loss at step 900: 0.054901354014873505 |
| Mean training loss after epoch 15: 0.05222853389916135 |
|
|
| EPOCH: 16 |
| Loss at step 0: 0.04499037191271782 |
| Loss at step 50: 0.04059568792581558 |
| Loss at step 100: 0.03880501538515091 |
| Loss at step 150: 0.04736030101776123 |
| Loss at step 200: 0.054259732365608215 |
| Loss at step 250: 0.056340526789426804 |
| Loss at step 300: 0.04246404767036438 |
| Loss at step 350: 0.05467449873685837 |
| Loss at step 400: 0.038072679191827774 |
| Loss at step 450: 0.05155707895755768 |
| Loss at step 500: 0.04925785958766937 |
| Loss at step 550: 0.04731006547808647 |
| Loss at step 600: 0.05020074173808098 |
| Loss at step 650: 0.04935452342033386 |
| Loss at step 700: 0.05027550831437111 |
| Loss at step 750: 0.04625604674220085 |
| Loss at step 800: 0.07112246006727219 |
| Loss at step 850: 0.04084265977144241 |
| Loss at step 900: 0.05504539981484413 |
| Mean training loss after epoch 16: 0.05157376733670103 |
|
|
| EPOCH: 17 |
| Loss at step 0: 0.07536608725786209 |
| Loss at step 50: 0.042563579976558685 |
| Loss at step 100: 0.035157352685928345 |
| Loss at step 150: 0.05038196220993996 |
| Loss at step 200: 0.049875665456056595 |
| Loss at step 250: 0.03781870752573013 |
| Loss at step 300: 0.056721512228250504 |
| Loss at step 350: 0.04967416077852249 |
| Loss at step 400: 0.03947357088327408 |
| Loss at step 450: 0.053190600126981735 |
| Loss at step 500: 0.0426432341337204 |
| Loss at step 550: 0.05922308564186096 |
| Loss at step 600: 0.054390799254179 |
| Loss at step 650: 0.08069872856140137 |
| Loss at step 700: 0.04753753915429115 |
| Loss at step 750: 0.036334265023469925 |
| Loss at step 800: 0.04727853834629059 |
| Loss at step 850: 0.0459805391728878 |
| Loss at step 900: 0.0718865841627121 |
| Mean training loss after epoch 17: 0.05112017348392813 |
|
|
| EPOCH: 18 |
| Loss at step 0: 0.04541482776403427 |
| Loss at step 50: 0.06357073783874512 |
| Loss at step 100: 0.05133732780814171 |
| Loss at step 150: 0.048947181552648544 |
| Loss at step 200: 0.04920993372797966 |
| Loss at step 250: 0.07129635661840439 |
| Loss at step 300: 0.041890259832143784 |
| Loss at step 350: 0.04461813345551491 |
| Loss at step 400: 0.06786632537841797 |
| Loss at step 450: 0.03864361345767975 |
| Loss at step 500: 0.05946999043226242 |
| Loss at step 550: 0.047408659011125565 |
| Loss at step 600: 0.04371762275695801 |
| Loss at step 650: 0.05137581750750542 |
| Loss at step 700: 0.039694201201200485 |
| Loss at step 750: 0.0421057753264904 |
| Loss at step 800: 0.056266531348228455 |
| Loss at step 850: 0.04970719292759895 |
| Loss at step 900: 0.04250839352607727 |
| Mean training loss after epoch 18: 0.05031145928002624 |
|
|
| EPOCH: 19 |
| Loss at step 0: 0.04314997047185898 |
| Loss at step 50: 0.041050687432289124 |
| Loss at step 100: 0.04350142553448677 |
| Loss at step 150: 0.04735090211033821 |
| Loss at step 200: 0.042304959148168564 |
| Loss at step 250: 0.0446118600666523 |
| Loss at step 300: 0.057678062468767166 |
| Loss at step 350: 0.05146310478448868 |
| Loss at step 400: 0.04377828165888786 |
| Loss at step 450: 0.03493072837591171 |
| Loss at step 500: 0.06859597563743591 |
| Loss at step 550: 0.046260152012109756 |
| Loss at step 600: 0.043077658861875534 |
| Loss at step 650: 0.055410489439964294 |
| Loss at step 700: 0.05613521486520767 |
| Loss at step 750: 0.051349785178899765 |
| Loss at step 800: 0.04863078147172928 |
| Loss at step 850: 0.03905788064002991 |
| Loss at step 900: 0.050505876541137695 |
| Mean training loss after epoch 19: 0.04981041669662891 |
|
|
| EPOCH: 20 |
| Loss at step 0: 0.05344609543681145 |
| Loss at step 50: 0.04180081561207771 |
| Loss at step 100: 0.04135894775390625 |
| Loss at step 150: 0.04881883040070534 |
| Loss at step 200: 0.05157637596130371 |
| Loss at step 250: 0.06887549161911011 |
| Loss at step 300: 0.04278310760855675 |
| Loss at step 350: 0.05660872906446457 |
| Loss at step 400: 0.04872731491923332 |
| Loss at step 450: 0.04364878311753273 |
| Loss at step 500: 0.06668026745319366 |
| Loss at step 550: 0.046359237283468246 |
| Loss at step 600: 0.03858613967895508 |
| Loss at step 650: 0.0463811419904232 |
| Loss at step 700: 0.0656844824552536 |
| Loss at step 750: 0.04059052839875221 |
| Loss at step 800: 0.04606105759739876 |
| Loss at step 850: 0.04518028721213341 |
| Loss at step 900: 0.0448128879070282 |
| Mean training loss after epoch 20: 0.04971912971088119 |
|
|
| EPOCH: 21 |
| Loss at step 0: 0.04147127643227577 |
| Loss at step 50: 0.06674125045537949 |
| Loss at step 100: 0.046588215976953506 |
| Loss at step 150: 0.04992952197790146 |
| Loss at step 200: 0.05683210864663124 |
| Loss at step 250: 0.06446004658937454 |
| Loss at step 300: 0.07351561635732651 |
| Loss at step 350: 0.048821087926626205 |
| Loss at step 400: 0.06500155478715897 |
| Loss at step 450: 0.045991450548172 |
| Loss at step 500: 0.04785887897014618 |
| Loss at step 550: 0.048360276967287064 |
| Loss at step 600: 0.06245497241616249 |
| Loss at step 650: 0.04141472652554512 |
| Loss at step 700: 0.04987093433737755 |
| Loss at step 750: 0.0448041595518589 |
| Loss at step 800: 0.039200447499752045 |
| Loss at step 850: 0.061862435191869736 |
| Loss at step 900: 0.0374799408018589 |
| Mean training loss after epoch 21: 0.04948336795083622 |
|
|
| EPOCH: 22 |
| Loss at step 0: 0.04231114313006401 |
| Loss at step 50: 0.051757242530584335 |
| Loss at step 100: 0.052523884922266006 |
| Loss at step 150: 0.04511893540620804 |
| Loss at step 200: 0.047847066074609756 |
| Loss at step 250: 0.03973916918039322 |
| Loss at step 300: 0.04602658003568649 |
| Loss at step 350: 0.04559744521975517 |
| Loss at step 400: 0.038105208426713943 |
| Loss at step 450: 0.04447225108742714 |
| Loss at step 500: 0.052992742508649826 |
| Loss at step 550: 0.04886539652943611 |
| Loss at step 600: 0.04174038767814636 |
| Loss at step 650: 0.0403764434158802 |
| Loss at step 700: 0.037883441895246506 |
| Loss at step 750: 0.04011991620063782 |
| Loss at step 800: 0.04163298383355141 |
| Loss at step 850: 0.0440143346786499 |
| Loss at step 900: 0.04741881787776947 |
| Mean training loss after epoch 22: 0.04886792954812045 |
|
|
| EPOCH: 23 |
| Loss at step 0: 0.043948426842689514 |
| Loss at step 50: 0.04617948457598686 |
| Loss at step 100: 0.041180990636348724 |
| Loss at step 150: 0.07108350843191147 |
| Loss at step 200: 0.05402369797229767 |
| Loss at step 250: 0.04757066071033478 |
| Loss at step 300: 0.04691469669342041 |
| Loss at step 350: 0.056709371507167816 |
| Loss at step 400: 0.08028485625982285 |
| Loss at step 450: 0.03732737526297569 |
| Loss at step 500: 0.0625799298286438 |
| Loss at step 550: 0.06315683573484421 |
| Loss at step 600: 0.06575877964496613 |
| Loss at step 650: 0.03729895502328873 |
| Loss at step 700: 0.04726678505539894 |
| Loss at step 750: 0.052553772926330566 |
| Loss at step 800: 0.04982905834913254 |
| Loss at step 850: 0.04016149044036865 |
| Loss at step 900: 0.0371665395796299 |
| Mean training loss after epoch 23: 0.04921432044793929 |
|
|
| EPOCH: 24 |
| Loss at step 0: 0.06877944618463516 |
| Loss at step 50: 0.04670701548457146 |
| Loss at step 100: 0.05103413760662079 |
| Loss at step 150: 0.04245966300368309 |
| Loss at step 200: 0.0422213077545166 |
| Loss at step 250: 0.05819988250732422 |
| Loss at step 300: 0.03808581456542015 |
| Loss at step 350: 0.044201094657182693 |
| Loss at step 400: 0.04023940488696098 |
| Loss at step 450: 0.03730885311961174 |
| Loss at step 500: 0.0679292157292366 |
| Loss at step 550: 0.04315111041069031 |
| Loss at step 600: 0.040241699665784836 |
| Loss at step 650: 0.048743754625320435 |
| Loss at step 700: 0.06133408471941948 |
| Loss at step 750: 0.035759568214416504 |
| Loss at step 800: 0.07543523609638214 |
| Loss at step 850: 0.041768599301576614 |
| Loss at step 900: 0.0580260194838047 |
| Mean training loss after epoch 24: 0.04907009113572045 |
|
|
| EPOCH: 25 |
| Loss at step 0: 0.04890033230185509 |
| Loss at step 50: 0.04610893502831459 |
| Loss at step 100: 0.062320057302713394 |
| Loss at step 150: 0.04533512517809868 |
| Loss at step 200: 0.04547395929694176 |
| Loss at step 250: 0.05030956491827965 |
| Loss at step 300: 0.06314963847398758 |
| Loss at step 350: 0.04957621917128563 |
| Loss at step 400: 0.04225068539381027 |
| Loss at step 450: 0.060565605759620667 |
| Loss at step 500: 0.04107939824461937 |
| Loss at step 550: 0.04405638575553894 |
| Loss at step 600: 0.035555195063352585 |
| Loss at step 650: 0.04223763942718506 |
| Loss at step 700: 0.05385289713740349 |
| Loss at step 750: 0.04406140744686127 |
| Loss at step 800: 0.04753494635224342 |
| Loss at step 850: 0.03908270597457886 |
| Loss at step 900: 0.056173697113990784 |
| Mean training loss after epoch 25: 0.04854819401185205 |
|
|
| EPOCH: 26 |
| Loss at step 0: 0.04513881728053093 |
| Loss at step 50: 0.04251381754875183 |
| Loss at step 100: 0.056283511221408844 |
| Loss at step 150: 0.03885107487440109 |
| Loss at step 200: 0.04387012869119644 |
| Loss at step 250: 0.08682840317487717 |
| Loss at step 300: 0.042384058237075806 |
| Loss at step 350: 0.04147539287805557 |
| Loss at step 400: 0.037699781358242035 |
| Loss at step 450: 0.08088959008455276 |
| Loss at step 500: 0.045566800981760025 |
| Loss at step 550: 0.04850791022181511 |
| Loss at step 600: 0.03581390157341957 |
| Loss at step 650: 0.036787454038858414 |
| Loss at step 700: 0.04770814999938011 |
| Loss at step 750: 0.044783566147089005 |
| Loss at step 800: 0.031522780656814575 |
| Loss at step 850: 0.04192390292882919 |
| Loss at step 900: 0.053765568882226944 |
| Mean training loss after epoch 26: 0.048748386362547684 |
|
|
| EPOCH: 27 |
| Loss at step 0: 0.062231797724962234 |
| Loss at step 50: 0.05955164507031441 |
| Loss at step 100: 0.039682600647211075 |
| Loss at step 150: 0.06752344220876694 |
| Loss at step 200: 0.061354734003543854 |
| Loss at step 250: 0.037967052310705185 |
| Loss at step 300: 0.04776729643344879 |
| Loss at step 350: 0.0504353865981102 |
| Loss at step 400: 0.04052473604679108 |
| Loss at step 450: 0.05573800951242447 |
| Loss at step 500: 0.04068378359079361 |
| Loss at step 550: 0.03938471898436546 |
| Loss at step 600: 0.04409739375114441 |
| Loss at step 650: 0.06390330195426941 |
| Loss at step 700: 0.04190912842750549 |
| Loss at step 750: 0.04705492779612541 |
| Loss at step 800: 0.05942341312766075 |
| Loss at step 850: 0.054906755685806274 |
| Loss at step 900: 0.04091959819197655 |
| Mean training loss after epoch 27: 0.04785118063590102 |
|
|
| EPOCH: 28 |
| Loss at step 0: 0.040065884590148926 |
| Loss at step 50: 0.041816335171461105 |
| Loss at step 100: 0.04768802598118782 |
| Loss at step 150: 0.04224841296672821 |
| Loss at step 200: 0.0600610077381134 |
| Loss at step 250: 0.04674086347222328 |
| Loss at step 300: 0.03749881684780121 |
| Loss at step 350: 0.05725927650928497 |
| Loss at step 400: 0.03463057056069374 |
| Loss at step 450: 0.04139943793416023 |
| Loss at step 500: 0.036785632371902466 |
| Loss at step 550: 0.037280261516571045 |
| Loss at step 600: 0.042486049234867096 |
| Loss at step 650: 0.039891812950372696 |
| Loss at step 700: 0.045055538415908813 |
| Loss at step 750: 0.044806648045778275 |
| Loss at step 800: 0.035296935588121414 |
| Loss at step 850: 0.03997200354933739 |
| Loss at step 900: 0.037372902035713196 |
| Mean training loss after epoch 28: 0.04802111568433771 |
|
|
| EPOCH: 29 |
| Loss at step 0: 0.04029547795653343 |
| Loss at step 50: 0.0373159721493721 |
| Loss at step 100: 0.03195224702358246 |
| Loss at step 150: 0.03967324644327164 |
| Loss at step 200: 0.043493278324604034 |
| Loss at step 250: 0.0421735905110836 |
| Loss at step 300: 0.05834754556417465 |
| Loss at step 350: 0.04977051913738251 |
| Loss at step 400: 0.04890020564198494 |
| Loss at step 450: 0.06144270300865173 |
| Loss at step 500: 0.031416650861501694 |
| Loss at step 550: 0.05089873448014259 |
| Loss at step 600: 0.0559379979968071 |
| Loss at step 650: 0.0431060828268528 |
| Loss at step 700: 0.07970152795314789 |
| Loss at step 750: 0.056120067834854126 |
| Loss at step 800: 0.03743378818035126 |
| Loss at step 850: 0.04540432617068291 |
| Loss at step 900: 0.03750884160399437 |
| Mean training loss after epoch 29: 0.0479990868275163 |
|
|
| EPOCH: 30 |
| Loss at step 0: 0.06987270712852478 |
| Loss at step 50: 0.04634237289428711 |
| Loss at step 100: 0.04522722586989403 |
| Loss at step 150: 0.037521373480558395 |
| Loss at step 200: 0.04052366316318512 |
| Loss at step 250: 0.047215189784765244 |
| Loss at step 300: 0.03429022431373596 |
| Loss at step 350: 0.042049288749694824 |
| Loss at step 400: 0.04067468270659447 |
| Loss at step 450: 0.04121248796582222 |
| Loss at step 500: 0.041117824614048004 |
| Loss at step 550: 0.04192543774843216 |
| Loss at step 600: 0.0396721288561821 |
| Loss at step 650: 0.0341549776494503 |
| Loss at step 700: 0.0468466654419899 |
| Loss at step 750: 0.047854892909526825 |
| Loss at step 800: 0.04301496222615242 |
| Loss at step 850: 0.06360778212547302 |
| Loss at step 900: 0.04396238178014755 |
| Mean training loss after epoch 30: 0.047711722947943055 |
|
|
| EPOCH: 31 |
| Loss at step 0: 0.046192727982997894 |
| Loss at step 50: 0.056457825005054474 |
| Loss at step 100: 0.04437893256545067 |
| Loss at step 150: 0.03571672737598419 |
| Loss at step 200: 0.041461531072854996 |
| Loss at step 250: 0.038135726004838943 |
| Loss at step 300: 0.04517757520079613 |
| Loss at step 350: 0.03788148984313011 |
| Loss at step 400: 0.04530879482626915 |
| Loss at step 450: 0.04953639209270477 |
| Loss at step 500: 0.04382310435175896 |
| Loss at step 550: 0.04811768978834152 |
| Loss at step 600: 0.05481934919953346 |
| Loss at step 650: 0.04873424395918846 |
| Loss at step 700: 0.03968692570924759 |
| Loss at step 750: 0.0569743812084198 |
| Loss at step 800: 0.034903742372989655 |
| Loss at step 850: 0.041342843323946 |
| Loss at step 900: 0.04470614343881607 |
| Mean training loss after epoch 31: 0.04718852380135738 |
|
|
| EPOCH: 32 |
| Loss at step 0: 0.04153997823596001 |
| Loss at step 50: 0.05805188789963722 |
| Loss at step 100: 0.041979819536209106 |
| Loss at step 150: 0.056203462183475494 |
| Loss at step 200: 0.0658581554889679 |
| Loss at step 250: 0.047670237720012665 |
| Loss at step 300: 0.04966076835989952 |
| Loss at step 350: 0.03970366343855858 |
| Loss at step 400: 0.04131742939352989 |
| Loss at step 450: 0.0559941828250885 |
| Loss at step 500: 0.0515303798019886 |
| Loss at step 550: 0.03879852220416069 |
| Loss at step 600: 0.04554179310798645 |
| Loss at step 650: 0.041446492075920105 |
| Loss at step 700: 0.045973826199769974 |
| Loss at step 750: 0.05146137624979019 |
| Loss at step 800: 0.050301820039749146 |
| Loss at step 850: 0.05406748503446579 |
| Loss at step 900: 0.047541599720716476 |
| Mean training loss after epoch 32: 0.04734661062555844 |
|
|
| EPOCH: 33 |
| Loss at step 0: 0.04255426675081253 |
| Loss at step 50: 0.045197855681180954 |
| Loss at step 100: 0.04791636765003204 |
| Loss at step 150: 0.053649891167879105 |
| Loss at step 200: 0.04649053514003754 |
| Loss at step 250: 0.06254279613494873 |
| Loss at step 300: 0.04070403426885605 |
| Loss at step 350: 0.06130688264966011 |
| Loss at step 400: 0.033172477036714554 |
| Loss at step 450: 0.0585973784327507 |
| Loss at step 500: 0.03862085938453674 |
| Loss at step 550: 0.05066192150115967 |
| Loss at step 600: 0.04806030914187431 |
| Loss at step 650: 0.040029771625995636 |
| Loss at step 700: 0.05965670570731163 |
| Loss at step 750: 0.04943963512778282 |
| Loss at step 800: 0.04053017124533653 |
| Loss at step 850: 0.0498194620013237 |
| Loss at step 900: 0.0408606119453907 |
| Mean training loss after epoch 33: 0.0466234264577598 |
|
|
| EPOCH: 34 |
| Loss at step 0: 0.03834282234311104 |
| Loss at step 50: 0.04476098716259003 |
| Loss at step 100: 0.03126645088195801 |
| Loss at step 150: 0.04436739161610603 |
| Loss at step 200: 0.035322658717632294 |
| Loss at step 250: 0.04000015929341316 |
| Loss at step 300: 0.03503677248954773 |
| Loss at step 350: 0.036471109837293625 |
| Loss at step 400: 0.041077565401792526 |
| Loss at step 450: 0.05009006708860397 |
| Loss at step 500: 0.03923311457037926 |
| Loss at step 550: 0.027845796197652817 |
| Loss at step 600: 0.044510193169116974 |
| Loss at step 650: 0.04472494497895241 |
| Loss at step 700: 0.036923471838235855 |
| Loss at step 750: 0.03947778046131134 |
| Loss at step 800: 0.04731610417366028 |
| Loss at step 850: 0.0408109687268734 |
| Loss at step 900: 0.04986851289868355 |
| Mean training loss after epoch 34: 0.046696712082224104 |
|
|
| EPOCH: 35 |
| Loss at step 0: 0.05347377434372902 |
| Loss at step 50: 0.04718519747257233 |
| Loss at step 100: 0.03765302151441574 |
| Loss at step 150: 0.04094993695616722 |
| Loss at step 200: 0.03967397287487984 |
| Loss at step 250: 0.04097428172826767 |
| Loss at step 300: 0.044033098965883255 |
| Loss at step 350: 0.04170210286974907 |
| Loss at step 400: 0.0495026521384716 |
| Loss at step 450: 0.04580632597208023 |
| Loss at step 500: 0.03558526188135147 |
| Loss at step 550: 0.03231578320264816 |
| Loss at step 600: 0.03808858245611191 |
| Loss at step 650: 0.07824570685625076 |
| Loss at step 700: 0.054205626249313354 |
| Loss at step 750: 0.04045689105987549 |
| Loss at step 800: 0.05172043293714523 |
| Loss at step 850: 0.04387115314602852 |
| Loss at step 900: 0.041183892637491226 |
| Mean training loss after epoch 35: 0.046853596549123716 |
|
|
| EPOCH: 36 |
| Loss at step 0: 0.033817023038864136 |
| Loss at step 50: 0.044775836169719696 |
| Loss at step 100: 0.04400419443845749 |
| Loss at step 150: 0.03964809328317642 |
| Loss at step 200: 0.0711473822593689 |
| Loss at step 250: 0.05198086053133011 |
| Loss at step 300: 0.04352946951985359 |
| Loss at step 350: 0.04083481431007385 |
| Loss at step 400: 0.05258826166391373 |
| Loss at step 450: 0.05412430688738823 |
| Loss at step 500: 0.036205705255270004 |
| Loss at step 550: 0.051293227821588516 |
| Loss at step 600: 0.039234597235918045 |
| Loss at step 650: 0.04175081476569176 |
| Loss at step 700: 0.044233884662389755 |
| Loss at step 750: 0.047170866280794144 |
| Loss at step 800: 0.06287672370672226 |
| Loss at step 850: 0.045048587024211884 |
| Loss at step 900: 0.047904081642627716 |
| Mean training loss after epoch 36: 0.04643794561007511 |
|
|
| EPOCH: 37 |
| Loss at step 0: 0.04380636289715767 |
| Loss at step 50: 0.05075736716389656 |
| Loss at step 100: 0.03939420357346535 |
| Loss at step 150: 0.041874587535858154 |
| Loss at step 200: 0.03474506363272667 |
| Loss at step 250: 0.04335370287299156 |
| Loss at step 300: 0.03736516833305359 |
| Loss at step 350: 0.04953424260020256 |
| Loss at step 400: 0.034158531576395035 |
| Loss at step 450: 0.09222576022148132 |
| Loss at step 500: 0.03409889340400696 |
| Loss at step 550: 0.05518524721264839 |
| Loss at step 600: 0.04299849644303322 |
| Loss at step 650: 0.03976324200630188 |
| Loss at step 700: 0.04704044759273529 |
| Loss at step 750: 0.05765819549560547 |
| Loss at step 800: 0.06325271725654602 |
| Loss at step 850: 0.041912227869033813 |
| Loss at step 900: 0.03947821259498596 |
| Mean training loss after epoch 37: 0.046509189958146006 |
|
|
| EPOCH: 38 |
| Loss at step 0: 0.04105685278773308 |
| Loss at step 50: 0.03853827714920044 |
| Loss at step 100: 0.0615091510117054 |
| Loss at step 150: 0.037095218896865845 |
| Loss at step 200: 0.07169077545404434 |
| Loss at step 250: 0.04093547165393829 |
| Loss at step 300: 0.03914790228009224 |
| Loss at step 350: 0.041914209723472595 |
| Loss at step 400: 0.04600575566291809 |
| Loss at step 450: 0.04167331010103226 |
| Loss at step 500: 0.05215302109718323 |
| Loss at step 550: 0.057997964322566986 |
| Loss at step 600: 0.04561407491564751 |
| Loss at step 650: 0.035688042640686035 |
| Loss at step 700: 0.041937340050935745 |
| Loss at step 750: 0.042008861899375916 |
| Loss at step 800: 0.043819013983011246 |
| Loss at step 850: 0.06488332897424698 |
| Loss at step 900: 0.04783814772963524 |
| Mean training loss after epoch 38: 0.04656100857343628 |
|
|
| EPOCH: 39 |
| Loss at step 0: 0.046811822801828384 |
| Loss at step 50: 0.03458044305443764 |
| Loss at step 100: 0.03519456088542938 |
| Loss at step 150: 0.04023684188723564 |
| Loss at step 200: 0.0593542717397213 |
| Loss at step 250: 0.04351789876818657 |
| Loss at step 300: 0.03666255623102188 |
| Loss at step 350: 0.042385295033454895 |
| Loss at step 400: 0.04535805433988571 |
| Loss at step 450: 0.040133584290742874 |
| Loss at step 500: 0.04300827905535698 |
| Loss at step 550: 0.03428584709763527 |
| Loss at step 600: 0.043937813490629196 |
| Loss at step 650: 0.04107518866658211 |
| Loss at step 700: 0.0656203031539917 |
| Loss at step 750: 0.04333982616662979 |
| Loss at step 800: 0.04911601543426514 |
| Loss at step 850: 0.04030092433094978 |
| Loss at step 900: 0.037120744585990906 |
| Mean training loss after epoch 39: 0.045584303543352876 |
|
|
| EPOCH: 40 |
| Loss at step 0: 0.043902307748794556 |
| Loss at step 50: 0.04687623307108879 |
| Loss at step 100: 0.02807818166911602 |
| Loss at step 150: 0.0314083956182003 |
| Loss at step 200: 0.03794383257627487 |
| Loss at step 250: 0.06447502225637436 |
| Loss at step 300: 0.03832196444272995 |
| Loss at step 350: 0.04955507069826126 |
| Loss at step 400: 0.04458783194422722 |
| Loss at step 450: 0.06182854250073433 |
| Loss at step 500: 0.034158360213041306 |
| Loss at step 550: 0.05126744136214256 |
| Loss at step 600: 0.04173079878091812 |
| Loss at step 650: 0.047507818788290024 |
| Loss at step 700: 0.039553675800561905 |
| Loss at step 750: 0.036442361772060394 |
| Loss at step 800: 0.03931163251399994 |
| Loss at step 850: 0.03706500306725502 |
| Loss at step 900: 0.041083257645368576 |
| Mean training loss after epoch 40: 0.04583525314712639 |
|
|
| EPOCH: 41 |
| Loss at step 0: 0.04119272530078888 |
| Loss at step 50: 0.03651197627186775 |
| Loss at step 100: 0.04471709579229355 |
| Loss at step 150: 0.047409866005182266 |
| Loss at step 200: 0.057044386863708496 |
| Loss at step 250: 0.03891262039542198 |
| Loss at step 300: 0.058914415538311005 |
| Loss at step 350: 0.04932182654738426 |
| Loss at step 400: 0.03527601808309555 |
| Loss at step 450: 0.05954860895872116 |
| Loss at step 500: 0.05854950472712517 |
| Loss at step 550: 0.04339790716767311 |
| Loss at step 600: 0.04158823564648628 |
| Loss at step 650: 0.07653175294399261 |
| Loss at step 700: 0.06495469808578491 |
| Loss at step 750: 0.043791264295578 |
| Loss at step 800: 0.036664631217718124 |
| Loss at step 850: 0.04060354083776474 |
| Loss at step 900: 0.05105862393975258 |
| Mean training loss after epoch 41: 0.04578510606919588 |
|
|
| EPOCH: 42 |
| Loss at step 0: 0.036811619997024536 |
| Loss at step 50: 0.03967383876442909 |
| Loss at step 100: 0.046618491411209106 |
| Loss at step 150: 0.04137549176812172 |
| Loss at step 200: 0.04536155238747597 |
| Loss at step 250: 0.0344829335808754 |
| Loss at step 300: 0.04808695986866951 |
| Loss at step 350: 0.034518640488386154 |
| Loss at step 400: 0.040801674127578735 |
| Loss at step 450: 0.04482355713844299 |
| Loss at step 500: 0.05345488339662552 |
| Loss at step 550: 0.042088404297828674 |
| Loss at step 600: 0.03969200327992439 |
| Loss at step 650: 0.03745196387171745 |
| Loss at step 700: 0.05043499544262886 |
| Loss at step 750: 0.03882772848010063 |
| Loss at step 800: 0.040638964623212814 |
| Loss at step 850: 0.059426199644804 |
| Loss at step 900: 0.055314578115940094 |
| Mean training loss after epoch 42: 0.0457447647595647 |
|
|
| EPOCH: 43 |
| Loss at step 0: 0.039296090602874756 |
| Loss at step 50: 0.03757215291261673 |
| Loss at step 100: 0.045208077877759933 |
| Loss at step 150: 0.05504849553108215 |
| Loss at step 200: 0.03858547285199165 |
| Loss at step 250: 0.052830155938863754 |
| Loss at step 300: 0.03214762732386589 |
| Loss at step 350: 0.054610736668109894 |
| Loss at step 400: 0.072967529296875 |
| Loss at step 450: 0.045324429869651794 |
| Loss at step 500: 0.05956553295254707 |
| Loss at step 550: 0.040348950773477554 |
| Loss at step 600: 0.038659922778606415 |
| Loss at step 650: 0.04288933053612709 |
| Loss at step 700: 0.05102803558111191 |
| Loss at step 750: 0.04348177835345268 |
| Loss at step 800: 0.03679901733994484 |
| Loss at step 850: 0.04044809564948082 |
| Loss at step 900: 0.05503632500767708 |
| Mean training loss after epoch 43: 0.04555305257550816 |
|
|
| EPOCH: 44 |
| Loss at step 0: 0.03301229327917099 |
| Loss at step 50: 0.03727089241147041 |
| Loss at step 100: 0.042083825916051865 |
| Loss at step 150: 0.03446154296398163 |
| Loss at step 200: 0.0325627475976944 |
| Loss at step 250: 0.03536694869399071 |
| Loss at step 300: 0.03226042538881302 |
| Loss at step 350: 0.029795240610837936 |
| Loss at step 400: 0.06270813196897507 |
| Loss at step 450: 0.052609071135520935 |
| Loss at step 500: 0.04123945161700249 |
| Loss at step 550: 0.04485645145177841 |
| Loss at step 600: 0.03443359211087227 |
| Loss at step 650: 0.05156237632036209 |
| Loss at step 700: 0.05249813199043274 |
| Loss at step 750: 0.04118892922997475 |
| Loss at step 800: 0.042128149420022964 |
| Loss at step 850: 0.05222824215888977 |
| Loss at step 900: 0.04898892343044281 |
| Mean training loss after epoch 44: 0.045909754842567416 |
|
|
| EPOCH: 45 |
| Loss at step 0: 0.03726803511381149 |
| Loss at step 50: 0.041821129620075226 |
| Loss at step 100: 0.04788229241967201 |
| Loss at step 150: 0.07195736467838287 |
| Loss at step 200: 0.04190758243203163 |
| Loss at step 250: 0.04406964033842087 |
| Loss at step 300: 0.045066285878419876 |
| Loss at step 350: 0.08542950451374054 |
| Loss at step 400: 0.036356471478939056 |
| Loss at step 450: 0.05726455897092819 |
| Loss at step 500: 0.04086785390973091 |
| Loss at step 550: 0.04460451379418373 |
| Loss at step 600: 0.04124069958925247 |
| Loss at step 650: 0.03595242276787758 |
| Loss at step 700: 0.05711403861641884 |
| Loss at step 750: 0.04426044225692749 |
| Loss at step 800: 0.03155608847737312 |
| Loss at step 850: 0.04087338596582413 |
| Loss at step 900: 0.03957609087228775 |
| Mean training loss after epoch 45: 0.04559232201625798 |
|
|
| EPOCH: 46 |
| Loss at step 0: 0.03411095589399338 |
| Loss at step 50: 0.04670671746134758 |
| Loss at step 100: 0.05237060412764549 |
| Loss at step 150: 0.036262497305870056 |
| Loss at step 200: 0.057636260986328125 |
| Loss at step 250: 0.04624015465378761 |
| Loss at step 300: 0.0441092886030674 |
| Loss at step 350: 0.053948499262332916 |
| Loss at step 400: 0.0504290908575058 |
| Loss at step 450: 0.058462247252464294 |
| Loss at step 500: 0.03294298052787781 |
| Loss at step 550: 0.03885740414261818 |
| Loss at step 600: 0.03582470864057541 |
| Loss at step 650: 0.03290253505110741 |
| Loss at step 700: 0.034320082515478134 |
| Loss at step 750: 0.04563366621732712 |
| Loss at step 800: 0.037790633738040924 |
| Loss at step 850: 0.04004897549748421 |
| Loss at step 900: 0.05734499916434288 |
| Mean training loss after epoch 46: 0.045401750487892995 |
|
|
| EPOCH: 47 |
| Loss at step 0: 0.057515017688274384 |
| Loss at step 50: 0.05918058380484581 |
| Loss at step 100: 0.06794729083776474 |
| Loss at step 150: 0.04365697130560875 |
| Loss at step 200: 0.03533811867237091 |
| Loss at step 250: 0.0432097464799881 |
| Loss at step 300: 0.04337077960371971 |
| Loss at step 350: 0.05455371364951134 |
| Loss at step 400: 0.0480773001909256 |
| Loss at step 450: 0.03549404814839363 |
| Loss at step 500: 0.04109460487961769 |
| Loss at step 550: 0.05278811976313591 |
| Loss at step 600: 0.044051751494407654 |
| Loss at step 650: 0.03891625255346298 |
| Loss at step 700: 0.033852893859148026 |
| Loss at step 750: 0.05946379527449608 |
| Loss at step 800: 0.03839148208498955 |
| Loss at step 850: 0.05594377964735031 |
| Loss at step 900: 0.04837467148900032 |
| Mean training loss after epoch 47: 0.04540569746672218 |
|
|
| EPOCH: 48 |
| Loss at step 0: 0.03606181964278221 |
| Loss at step 50: 0.0393998920917511 |
| Loss at step 100: 0.0373966284096241 |
| Loss at step 150: 0.03726385161280632 |
| Loss at step 200: 0.053862374275922775 |
| Loss at step 250: 0.039281781762838364 |
| Loss at step 300: 0.052047569304704666 |
| Loss at step 350: 0.04218338802456856 |
| Loss at step 400: 0.054648831486701965 |
| Loss at step 450: 0.03355717286467552 |
| Loss at step 500: 0.04461154714226723 |
| Loss at step 550: 0.05380718782544136 |
| Loss at step 600: 0.03750714287161827 |
| Loss at step 650: 0.04356677457690239 |
| Loss at step 700: 0.056141939014196396 |
| Loss at step 750: 0.05722740665078163 |
| Loss at step 800: 0.05378589779138565 |
| Loss at step 850: 0.04318094253540039 |
| Loss at step 900: 0.03716384992003441 |
| Mean training loss after epoch 48: 0.04575384488261775 |
|
|
| EPOCH: 49 |
| Loss at step 0: 0.03405068814754486 |
| Loss at step 50: 0.0324208065867424 |
| Loss at step 100: 0.058766111731529236 |
| Loss at step 150: 0.050890326499938965 |
| Loss at step 200: 0.07835625112056732 |
| Loss at step 250: 0.06103396788239479 |
| Loss at step 300: 0.045262910425662994 |
| Loss at step 350: 0.039874423295259476 |
| Loss at step 400: 0.04232581704854965 |
| Loss at step 450: 0.0406830869615078 |
| Loss at step 500: 0.034022778272628784 |
| Loss at step 550: 0.042880669236183167 |
| Loss at step 600: 0.04260297492146492 |
| Loss at step 650: 0.03611469268798828 |
| Loss at step 700: 0.04044531658291817 |
| Loss at step 750: 0.04906051978468895 |
| Loss at step 800: 0.039296798408031464 |
| Loss at step 850: 0.07345636188983917 |
| Loss at step 900: 0.03570806607604027 |
| Mean training loss after epoch 49: 0.04555136628591938 |
|
|
| EPOCH: 50 |
| Loss at step 0: 0.03376990929245949 |
| Loss at step 50: 0.03915917128324509 |
| Loss at step 100: 0.045188069343566895 |
| Loss at step 150: 0.03396744653582573 |
| Loss at step 200: 0.037946101278066635 |
| Loss at step 250: 0.048230335116386414 |
| Loss at step 300: 0.03957950696349144 |
| Loss at step 350: 0.06345433741807938 |
| Loss at step 400: 0.05446555092930794 |
| Loss at step 450: 0.041967470198869705 |
| Loss at step 500: 0.06003414839506149 |
| Loss at step 550: 0.04475271329283714 |
| Loss at step 600: 0.0462694950401783 |
| Loss at step 650: 0.03884352743625641 |
| Loss at step 700: 0.03437794744968414 |
| Loss at step 750: 0.03485594317317009 |
| Loss at step 800: 0.03861059248447418 |
| Loss at step 850: 0.04278019815683365 |
| Loss at step 900: 0.036629848182201385 |
| Mean training loss after epoch 50: 0.04574327027675376 |
|
|
| EPOCH: 51 |
| Loss at step 0: 0.04773973673582077 |
| Loss at step 50: 0.04592078924179077 |
| Loss at step 100: 0.04153735190629959 |
| Loss at step 150: 0.04289252310991287 |
| Loss at step 200: 0.05378266051411629 |
| Loss at step 250: 0.05587004870176315 |
| Loss at step 300: 0.04163934290409088 |
| Loss at step 350: 0.0363738052546978 |
| Loss at step 400: 0.037308674305677414 |
| Loss at step 450: 0.03993014246225357 |
| Loss at step 500: 0.038397662341594696 |
| Loss at step 550: 0.03834199905395508 |
| Loss at step 600: 0.04541454836726189 |
| Loss at step 650: 0.03435764089226723 |
| Loss at step 700: 0.049492184072732925 |
| Loss at step 750: 0.06151849403977394 |
| Loss at step 800: 0.04823997616767883 |
| Loss at step 850: 0.05454571172595024 |
| Loss at step 900: 0.05095576122403145 |
| Mean training loss after epoch 51: 0.04479894658792883 |
|
|
| EPOCH: 52 |
| Loss at step 0: 0.05793466791510582 |
| Loss at step 50: 0.0341513492166996 |
| Loss at step 100: 0.031205492094159126 |
| Loss at step 150: 0.04038567468523979 |
| Loss at step 200: 0.04924270883202553 |
| Loss at step 250: 0.03751857206225395 |
| Loss at step 300: 0.03824424743652344 |
| Loss at step 350: 0.03565317019820213 |
| Loss at step 400: 0.03829163685441017 |
| Loss at step 450: 0.04630826786160469 |
| Loss at step 500: 0.036059457808732986 |
| Loss at step 550: 0.0380818210542202 |
| Loss at step 600: 0.05621505156159401 |
| Loss at step 650: 0.04004345089197159 |
| Loss at step 700: 0.04257235303521156 |
| Loss at step 750: 0.050609856843948364 |
| Loss at step 800: 0.03490433469414711 |
| Loss at step 850: 0.06108823046088219 |
| Loss at step 900: 0.03759215399622917 |
| Mean training loss after epoch 52: 0.04466682242781623 |
|
|
| EPOCH: 53 |
| Loss at step 0: 0.07153818756341934 |
| Loss at step 50: 0.037624310702085495 |
| Loss at step 100: 0.04795879125595093 |
| Loss at step 150: 0.042059168219566345 |
| Loss at step 200: 0.04592517390847206 |
| Loss at step 250: 0.047940898686647415 |
| Loss at step 300: 0.03760400414466858 |
| Loss at step 350: 0.058457329869270325 |
| Loss at step 400: 0.04059361293911934 |
| Loss at step 450: 0.03498782962560654 |
| Loss at step 500: 0.03895006701350212 |
| Loss at step 550: 0.03143053501844406 |
| Loss at step 600: 0.04286786541342735 |
| Loss at step 650: 0.02890370972454548 |
| Loss at step 700: 0.03054078109562397 |
| Loss at step 750: 0.0536319725215435 |
| Loss at step 800: 0.03379025310277939 |
| Loss at step 850: 0.041429825127124786 |
| Loss at step 900: 0.036586616188287735 |
| Mean training loss after epoch 53: 0.045109023038210516 |
|
|
| EPOCH: 54 |
| Loss at step 0: 0.034632205963134766 |
| Loss at step 50: 0.050900544971227646 |
| Loss at step 100: 0.055058639496564865 |
| Loss at step 150: 0.03736027702689171 |
| Loss at step 200: 0.03886985033750534 |
| Loss at step 250: 0.04547104611992836 |
| Loss at step 300: 0.03748707473278046 |
| Loss at step 350: 0.04005163908004761 |
| Loss at step 400: 0.04667902737855911 |
| Loss at step 450: 0.058137696236371994 |
| Loss at step 500: 0.036225397139787674 |
| Loss at step 550: 0.04980180785059929 |
| Loss at step 600: 0.0319833941757679 |
| Loss at step 650: 0.04132211208343506 |
| Loss at step 700: 0.031180264428257942 |
| Loss at step 750: 0.055976614356040955 |
| Loss at step 800: 0.040415141731500626 |
| Loss at step 850: 0.04537496343255043 |
| Loss at step 900: 0.06559032946825027 |
| Mean training loss after epoch 54: 0.0447263324688842 |
|
|
| EPOCH: 55 |
| Loss at step 0: 0.03894679620862007 |
| Loss at step 50: 0.055382825434207916 |
| Loss at step 100: 0.05500679835677147 |
| Loss at step 150: 0.03706794232130051 |
| Loss at step 200: 0.03788752481341362 |
| Loss at step 250: 0.044823382049798965 |
| Loss at step 300: 0.04081863537430763 |
| Loss at step 350: 0.051558658480644226 |
| Loss at step 400: 0.04245118051767349 |
| Loss at step 450: 0.032435957342386246 |
| Loss at step 500: 0.043013881891965866 |
| Loss at step 550: 0.06566750258207321 |
| Loss at step 600: 0.03832860663533211 |
| Loss at step 650: 0.04228802025318146 |
| Loss at step 700: 0.06671218574047089 |
| Loss at step 750: 0.03785625100135803 |
| Loss at step 800: 0.036910444498062134 |
| Loss at step 850: 0.0428953543305397 |
| Loss at step 900: 0.04201735183596611 |
| Mean training loss after epoch 55: 0.0454020186234067 |
|
|
| EPOCH: 56 |
| Loss at step 0: 0.04074004292488098 |
| Loss at step 50: 0.037089236080646515 |
| Loss at step 100: 0.038650039583444595 |
| Loss at step 150: 0.0470503568649292 |
| Loss at step 200: 0.0335656963288784 |
| Loss at step 250: 0.05130336433649063 |
| Loss at step 300: 0.041321609169244766 |
| Loss at step 350: 0.05360228195786476 |
| Loss at step 400: 0.054425470530986786 |
| Loss at step 450: 0.034286580979824066 |
| Loss at step 500: 0.05873318016529083 |
| Loss at step 550: 0.04183107987046242 |
| Loss at step 600: 0.04275026172399521 |
| Loss at step 650: 0.036619555205106735 |
| Loss at step 700: 0.04658738896250725 |
| Loss at step 750: 0.044796522706747055 |
| Loss at step 800: 0.06018178537487984 |
| Loss at step 850: 0.06020748242735863 |
| Loss at step 900: 0.04875624179840088 |
| Mean training loss after epoch 56: 0.04514047994947573 |
|
|
| EPOCH: 57 |
| Loss at step 0: 0.037959903478622437 |
| Loss at step 50: 0.045734651386737823 |
| Loss at step 100: 0.03870757669210434 |
| Loss at step 150: 0.053085923194885254 |
| Loss at step 200: 0.043345555663108826 |
| Loss at step 250: 0.04010351002216339 |
| Loss at step 300: 0.036371856927871704 |
| Loss at step 350: 0.037596240639686584 |
| Loss at step 400: 0.03472841903567314 |
| Loss at step 450: 0.043242860585451126 |
| Loss at step 500: 0.04697151854634285 |
| Loss at step 550: 0.03386232256889343 |
| Loss at step 600: 0.05848577991127968 |
| Loss at step 650: 0.037965212017297745 |
| Loss at step 700: 0.0595490001142025 |
| Loss at step 750: 0.07245443761348724 |
| Loss at step 800: 0.04417521506547928 |
| Loss at step 850: 0.03808566927909851 |
| Loss at step 900: 0.040217895060777664 |
| Mean training loss after epoch 57: 0.04452560655772686 |
|
|
| EPOCH: 58 |
| Loss at step 0: 0.04217497631907463 |
| Loss at step 50: 0.04118611663579941 |
| Loss at step 100: 0.03760745748877525 |
| Loss at step 150: 0.03788227215409279 |
| Loss at step 200: 0.036340679973363876 |
| Loss at step 250: 0.044277723878622055 |
| Loss at step 300: 0.07555166631937027 |
| Loss at step 350: 0.039096567779779434 |
| Loss at step 400: 0.0501922108232975 |
| Loss at step 450: 0.04151548445224762 |
| Loss at step 500: 0.04863559827208519 |
| Loss at step 550: 0.04006786644458771 |
| Loss at step 600: 0.03473867475986481 |
| Loss at step 650: 0.03917982801795006 |
| Loss at step 700: 0.038434259593486786 |
| Loss at step 750: 0.04143731668591499 |
| Loss at step 800: 0.04229950159788132 |
| Loss at step 850: 0.04580509290099144 |
| Loss at step 900: 0.04491458460688591 |
| Mean training loss after epoch 58: 0.044204890307015195 |
|
|
| EPOCH: 59 |
| Loss at step 0: 0.04443920776247978 |
| Loss at step 50: 0.038615334779024124 |
| Loss at step 100: 0.044301483780145645 |
| Loss at step 150: 0.04094228893518448 |
| Loss at step 200: 0.039516597986221313 |
| Loss at step 250: 0.04826976731419563 |
| Loss at step 300: 0.06281120330095291 |
| Loss at step 350: 0.04962306469678879 |
| Loss at step 400: 0.0349610410630703 |
| Loss at step 450: 0.04815901815891266 |
| Loss at step 500: 0.04071445018053055 |
| Loss at step 550: 0.03885610029101372 |
| Loss at step 600: 0.04688568040728569 |
| Loss at step 650: 0.03309250622987747 |
| Loss at step 700: 0.0372539721429348 |
| Loss at step 750: 0.06692992895841599 |
| Loss at step 800: 0.05041303485631943 |
| Loss at step 850: 0.049872804433107376 |
| Loss at step 900: 0.036385852843523026 |
| Mean training loss after epoch 59: 0.044467081796369955 |
|
|
| EPOCH: 60 |
| Loss at step 0: 0.04014415666460991 |
| Loss at step 50: 0.08164425194263458 |
| Loss at step 100: 0.039430782198905945 |
| Loss at step 150: 0.03921978920698166 |
| Loss at step 200: 0.04293350130319595 |
| Loss at step 250: 0.04878218099474907 |
| Loss at step 300: 0.039910636842250824 |
| Loss at step 350: 0.042387500405311584 |
| Loss at step 400: 0.056811921298503876 |
| Loss at step 450: 0.03985873982310295 |
| Loss at step 500: 0.046190857887268066 |
| Loss at step 550: 0.04720006883144379 |
| Loss at step 600: 0.04626326635479927 |
| Loss at step 650: 0.05943738669157028 |
| Loss at step 700: 0.07274050265550613 |
| Loss at step 750: 0.03711295872926712 |
| Loss at step 800: 0.04901483654975891 |
| Loss at step 850: 0.05160197988152504 |
| Loss at step 900: 0.04616200923919678 |
| Mean training loss after epoch 60: 0.04484942184487131 |
|
|
| EPOCH: 61 |
| Loss at step 0: 0.034708376973867416 |
| Loss at step 50: 0.04810373857617378 |
| Loss at step 100: 0.031009318307042122 |
| Loss at step 150: 0.04073278605937958 |
| Loss at step 200: 0.037685640156269073 |
| Loss at step 250: 0.03710269555449486 |
| Loss at step 300: 0.03902563825249672 |
| Loss at step 350: 0.03862561285495758 |
| Loss at step 400: 0.04300009831786156 |
| Loss at step 450: 0.059608832001686096 |
| Loss at step 500: 0.041195113211870193 |
| Loss at step 550: 0.05418688431382179 |
| Loss at step 600: 0.054096776992082596 |
| Loss at step 650: 0.040527116507291794 |
| Loss at step 700: 0.05888514593243599 |
| Loss at step 750: 0.053690355271101 |
| Loss at step 800: 0.04354054108262062 |
| Loss at step 850: 0.03404700756072998 |
| Loss at step 900: 0.06319855898618698 |
| Mean training loss after epoch 61: 0.04449913012924225 |
|
|
| EPOCH: 62 |
| Loss at step 0: 0.038908716291189194 |
| Loss at step 50: 0.034311648458242416 |
| Loss at step 100: 0.05498092994093895 |
| Loss at step 150: 0.0483626127243042 |
| Loss at step 200: 0.053626175969839096 |
| Loss at step 250: 0.07578056305646896 |
| Loss at step 300: 0.03906629979610443 |
| Loss at step 350: 0.04471297562122345 |
| Loss at step 400: 0.05710723251104355 |
| Loss at step 450: 0.06131591647863388 |
| Loss at step 500: 0.03654533997178078 |
| Loss at step 550: 0.03932672366499901 |
| Loss at step 600: 0.03606953099370003 |
| Loss at step 650: 0.03935694694519043 |
| Loss at step 700: 0.05097009614109993 |
| Loss at step 750: 0.05328104645013809 |
| Loss at step 800: 0.04115912318229675 |
| Loss at step 850: 0.05281233787536621 |
| Loss at step 900: 0.041887253522872925 |
| Mean training loss after epoch 62: 0.043596246160630354 |
|
|
| EPOCH: 63 |
| Loss at step 0: 0.029015716165304184 |
| Loss at step 50: 0.04414261877536774 |
| Loss at step 100: 0.03649734705686569 |
| Loss at step 150: 0.03882032260298729 |
| Loss at step 200: 0.04342050477862358 |
| Loss at step 250: 0.04559728503227234 |
| Loss at step 300: 0.05425933748483658 |
| Loss at step 350: 0.04868466034531593 |
| Loss at step 400: 0.03153557702898979 |
| Loss at step 450: 0.0450434684753418 |
| Loss at step 500: 0.04320262745022774 |
| Loss at step 550: 0.03585299476981163 |
| Loss at step 600: 0.06846386194229126 |
| Loss at step 650: 0.047943733632564545 |
| Loss at step 700: 0.03418514505028725 |
| Loss at step 750: 0.03347398713231087 |
| Loss at step 800: 0.05686632916331291 |
| Loss at step 850: 0.06825472414493561 |
| Loss at step 900: 0.04557757452130318 |
| Mean training loss after epoch 63: 0.04486468083052429 |
|
|
| EPOCH: 64 |
| Loss at step 0: 0.038345180451869965 |
| Loss at step 50: 0.04878535494208336 |
| Loss at step 100: 0.03994043916463852 |
| Loss at step 150: 0.03193584829568863 |
| Loss at step 200: 0.04250764101743698 |
| Loss at step 250: 0.042187292128801346 |
| Loss at step 300: 0.03934232145547867 |
| Loss at step 350: 0.0445597879588604 |
| Loss at step 400: 0.05384596064686775 |
| Loss at step 450: 0.05301209166646004 |
| Loss at step 500: 0.05442678555846214 |
| Loss at step 550: 0.04261470213532448 |
| Loss at step 600: 0.07980488240718842 |
| Loss at step 650: 0.04727376252412796 |
| Loss at step 700: 0.05658571049571037 |
| Loss at step 750: 0.033086925745010376 |
| Loss at step 800: 0.03404126316308975 |
| Loss at step 850: 0.03186289966106415 |
| Loss at step 900: 0.05820830911397934 |
| Mean training loss after epoch 64: 0.04441411661973068 |
|
|
| EPOCH: 65 |
| Loss at step 0: 0.06885399669408798 |
| Loss at step 50: 0.05599125474691391 |
| Loss at step 100: 0.04768454283475876 |
| Loss at step 150: 0.053254634141922 |
| Loss at step 200: 0.03931909427046776 |
| Loss at step 250: 0.036698125302791595 |
| Loss at step 300: 0.05797281861305237 |
| Loss at step 350: 0.04019729793071747 |
| Loss at step 400: 0.04408475384116173 |
| Loss at step 450: 0.04160203039646149 |
| Loss at step 500: 0.04312726855278015 |
| Loss at step 550: 0.03871629014611244 |
| Loss at step 600: 0.03298605978488922 |
| Loss at step 650: 0.036178793758153915 |
| Loss at step 700: 0.05048339441418648 |
| Loss at step 750: 0.05091601237654686 |
| Loss at step 800: 0.04056558758020401 |
| Loss at step 850: 0.03996409475803375 |
| Loss at step 900: 0.05364915356040001 |
| Mean training loss after epoch 65: 0.044274648685238636 |
|
|
| EPOCH: 66 |
| Loss at step 0: 0.04305730387568474 |
| Loss at step 50: 0.03969784080982208 |
| Loss at step 100: 0.035962287336587906 |
| Loss at step 150: 0.041350722312927246 |
| Loss at step 200: 0.04764682427048683 |
| Loss at step 250: 0.052914198487997055 |
| Loss at step 300: 0.052789609879255295 |
| Loss at step 350: 0.04125749692320824 |
| Loss at step 400: 0.05972205847501755 |
| Loss at step 450: 0.03646932169795036 |
| Loss at step 500: 0.04054495692253113 |
| Loss at step 550: 0.057925526052713394 |
| Loss at step 600: 0.05502486601471901 |
| Loss at step 650: 0.06423813849687576 |
| Loss at step 700: 0.041651081293821335 |
| Loss at step 750: 0.03151623159646988 |
| Loss at step 800: 0.037293124943971634 |
| Loss at step 850: 0.05984169617295265 |
| Loss at step 900: 0.036552462726831436 |
| Mean training loss after epoch 66: 0.04449663154963555 |
|
|
| EPOCH: 67 |
| Loss at step 0: 0.03824847564101219 |
| Loss at step 50: 0.03565288707613945 |
| Loss at step 100: 0.039270512759685516 |
| Loss at step 150: 0.03623337671160698 |
| Loss at step 200: 0.050690602511167526 |
| Loss at step 250: 0.03742752596735954 |
| Loss at step 300: 0.04467052221298218 |
| Loss at step 350: 0.03912796080112457 |
| Loss at step 400: 0.04786604270339012 |
| Loss at step 450: 0.03586800396442413 |
| Loss at step 500: 0.05089021101593971 |
| Loss at step 550: 0.045541878789663315 |
| Loss at step 600: 0.03662842512130737 |
| Loss at step 650: 0.03804018348455429 |
| Loss at step 700: 0.06523645669221878 |
| Loss at step 750: 0.043222736567258835 |
| Loss at step 800: 0.05228416249155998 |
| Loss at step 850: 0.044387586414813995 |
| Loss at step 900: 0.03197185695171356 |
| Mean training loss after epoch 67: 0.044540597860619965 |
|
|
| EPOCH: 68 |
| Loss at step 0: 0.03530889376997948 |
| Loss at step 50: 0.05193200334906578 |
| Loss at step 100: 0.04188147187232971 |
| Loss at step 150: 0.044340841472148895 |
| Loss at step 200: 0.039617713540792465 |
| Loss at step 250: 0.029141953215003014 |
| Loss at step 300: 0.059967041015625 |
| Loss at step 350: 0.0396573543548584 |
| Loss at step 400: 0.05593840777873993 |
| Loss at step 450: 0.0386933758854866 |
| Loss at step 500: 0.034154921770095825 |
| Loss at step 550: 0.03156885504722595 |
| Loss at step 600: 0.036909282207489014 |
| Loss at step 650: 0.04749900475144386 |
| Loss at step 700: 0.05062949284911156 |
| Loss at step 750: 0.04943801835179329 |
| Loss at step 800: 0.04978271201252937 |
| Loss at step 850: 0.04305383935570717 |
| Loss at step 900: 0.04517525061964989 |
| Mean training loss after epoch 68: 0.04403649984416105 |
|
|
| EPOCH: 69 |
| Loss at step 0: 0.04713786393404007 |
| Loss at step 50: 0.03688153252005577 |
| Loss at step 100: 0.04463440924882889 |
| Loss at step 150: 0.05203818157315254 |
| Loss at step 200: 0.03601740673184395 |
| Loss at step 250: 0.05964969843626022 |
| Loss at step 300: 0.04440165311098099 |
| Loss at step 350: 0.03362405300140381 |
| Loss at step 400: 0.03362003713846207 |
| Loss at step 450: 0.05673094093799591 |
| Loss at step 500: 0.05858657509088516 |
| Loss at step 550: 0.0479864664375782 |
| Loss at step 600: 0.04205408692359924 |
| Loss at step 650: 0.04245060309767723 |
| Loss at step 700: 0.047696929425001144 |
| Loss at step 750: 0.054844941943883896 |
| Loss at step 800: 0.0449918769299984 |
| Loss at step 850: 0.038500912487506866 |
| Loss at step 900: 0.036807581782341 |
| Mean training loss after epoch 69: 0.044288720746935684 |
|
|
| EPOCH: 70 |
| Loss at step 0: 0.033517539501190186 |
| Loss at step 50: 0.03352891281247139 |
| Loss at step 100: 0.03971768915653229 |
| Loss at step 150: 0.037568893283605576 |
| Loss at step 200: 0.066883884370327 |
| Loss at step 250: 0.04035981744527817 |
| Loss at step 300: 0.04199424386024475 |
| Loss at step 350: 0.04756559059023857 |
| Loss at step 400: 0.03550230711698532 |
| Loss at step 450: 0.05051594227552414 |
| Loss at step 500: 0.03901894763112068 |
| Loss at step 550: 0.04193321615457535 |
| Loss at step 600: 0.039895620197057724 |
| Loss at step 650: 0.03211652860045433 |
| Loss at step 700: 0.040562212467193604 |
| Loss at step 750: 0.0446825809776783 |
| Loss at step 800: 0.0380869135260582 |
| Loss at step 850: 0.04290211945772171 |
| Loss at step 900: 0.0694216936826706 |
| Mean training loss after epoch 70: 0.0440076610275995 |
|
|
| EPOCH: 71 |
| Loss at step 0: 0.062086571007966995 |
| Loss at step 50: 0.03848670795559883 |
| Loss at step 100: 0.04382505640387535 |
| Loss at step 150: 0.05180523172020912 |
| Loss at step 200: 0.057804908603429794 |
| Loss at step 250: 0.0323488749563694 |
| Loss at step 300: 0.034228235483169556 |
| Loss at step 350: 0.03648495674133301 |
| Loss at step 400: 0.06315921247005463 |
| Loss at step 450: 0.04083339869976044 |
| Loss at step 500: 0.07379228621721268 |
| Loss at step 550: 0.03787954896688461 |
| Loss at step 600: 0.05260041356086731 |
| Loss at step 650: 0.059406060725450516 |
| Loss at step 700: 0.037485867738723755 |
| Loss at step 750: 0.05460409075021744 |
| Loss at step 800: 0.034081242978572845 |
| Loss at step 850: 0.04085755720734596 |
| Loss at step 900: 0.040349770337343216 |
| Mean training loss after epoch 71: 0.04361870941092401 |
|
|
| EPOCH: 72 |
| Loss at step 0: 0.029254566878080368 |
| Loss at step 50: 0.04164859652519226 |
| Loss at step 100: 0.055361270904541016 |
| Loss at step 150: 0.034679099917411804 |
| Loss at step 200: 0.057330336421728134 |
| Loss at step 250: 0.03006073087453842 |
| Loss at step 300: 0.04034830629825592 |
| Loss at step 350: 0.0592540018260479 |
| Loss at step 400: 0.050872188061475754 |
| Loss at step 450: 0.04640054702758789 |
| Loss at step 500: 0.04122794792056084 |
| Loss at step 550: 0.03518728166818619 |
| Loss at step 600: 0.03226521611213684 |
| Loss at step 650: 0.05286054685711861 |
| Loss at step 700: 0.056503985077142715 |
| Loss at step 750: 0.056091148406267166 |
| Loss at step 800: 0.0377902016043663 |
| Loss at step 850: 0.03336455300450325 |
| Loss at step 900: 0.053792499005794525 |
| Mean training loss after epoch 72: 0.04410509114215242 |
|
|
| EPOCH: 73 |
| Loss at step 0: 0.038552019745111465 |
| Loss at step 50: 0.05634484440088272 |
| Loss at step 100: 0.04676920548081398 |
| Loss at step 150: 0.03462210297584534 |
| Loss at step 200: 0.03854707255959511 |
| Loss at step 250: 0.03935917094349861 |
| Loss at step 300: 0.03562043607234955 |
| Loss at step 350: 0.053162477910518646 |
| Loss at step 400: 0.037162214517593384 |
| Loss at step 450: 0.04485498368740082 |
| Loss at step 500: 0.03977523371577263 |
| Loss at step 550: 0.04587831348180771 |
| Loss at step 600: 0.03899795562028885 |
| Loss at step 650: 0.04515599459409714 |
| Loss at step 700: 0.04103762283921242 |
| Loss at step 750: 0.039450667798519135 |
| Loss at step 800: 0.03686264902353287 |
| Loss at step 850: 0.04491258040070534 |
| Loss at step 900: 0.0488051176071167 |
| Mean training loss after epoch 73: 0.04355169675632644 |
|
|
| EPOCH: 74 |
| Loss at step 0: 0.04988710954785347 |
| Loss at step 50: 0.027748363092541695 |
| Loss at step 100: 0.04155075177550316 |
| Loss at step 150: 0.03735537454485893 |
| Loss at step 200: 0.032628364861011505 |
| Loss at step 250: 0.03547734394669533 |
| Loss at step 300: 0.037861354649066925 |
| Loss at step 350: 0.04635513946413994 |
| Loss at step 400: 0.048436835408210754 |
| Loss at step 450: 0.032090190798044205 |
| Loss at step 500: 0.03548084571957588 |
| Loss at step 550: 0.03390015661716461 |
| Loss at step 600: 0.047148313373327255 |
| Loss at step 650: 0.05798763036727905 |
| Loss at step 700: 0.04385918006300926 |
| Loss at step 750: 0.03126253932714462 |
| Loss at step 800: 0.036953262984752655 |
| Loss at step 850: 0.04472682997584343 |
| Loss at step 900: 0.04753943160176277 |
| Mean training loss after epoch 74: 0.0431889588994258 |
|
|
| EPOCH: 75 |
| Loss at step 0: 0.04028462618589401 |
| Loss at step 50: 0.0357113741338253 |
| Loss at step 100: 0.036158740520477295 |
| Loss at step 150: 0.03210321068763733 |
| Loss at step 200: 0.05170506611466408 |
| Loss at step 250: 0.03830462694168091 |
| Loss at step 300: 0.03860298916697502 |
| Loss at step 350: 0.04723210632801056 |
| Loss at step 400: 0.03571812063455582 |
| Loss at step 450: 0.039360903203487396 |
| Loss at step 500: 0.05320487916469574 |
| Loss at step 550: 0.04517296701669693 |
| Loss at step 600: 0.06515920162200928 |
| Loss at step 650: 0.04200473055243492 |
| Loss at step 700: 0.04345053806900978 |
| Loss at step 750: 0.0318668894469738 |
| Loss at step 800: 0.03969453275203705 |
| Loss at step 850: 0.03803049027919769 |
| Loss at step 900: 0.05230528861284256 |
| Mean training loss after epoch 75: 0.04395143901194527 |
|
|
| EPOCH: 76 |
| Loss at step 0: 0.05301972106099129 |
| Loss at step 50: 0.03598293662071228 |
| Loss at step 100: 0.034400515258312225 |
| Loss at step 150: 0.056835055351257324 |
| Loss at step 200: 0.034850649535655975 |
| Loss at step 250: 0.05500826612114906 |
| Loss at step 300: 0.034377194941043854 |
| Loss at step 350: 0.04097644239664078 |
| Loss at step 400: 0.040672894567251205 |
| Loss at step 450: 0.04145702347159386 |
| Loss at step 500: 0.04885398969054222 |
| Loss at step 550: 0.048190776258707047 |
| Loss at step 600: 0.05448161065578461 |
| Loss at step 650: 0.03598455339670181 |
| Loss at step 700: 0.039347682148218155 |
| Loss at step 750: 0.04168093949556351 |
| Loss at step 800: 0.05553968623280525 |
| Loss at step 850: 0.041357167065143585 |
| Loss at step 900: 0.05268019810318947 |
| Mean training loss after epoch 76: 0.04369923243247497 |
|
|
| EPOCH: 77 |
| Loss at step 0: 0.05658970773220062 |
| Loss at step 50: 0.04428544640541077 |
| Loss at step 100: 0.043661028146743774 |
| Loss at step 150: 0.0697946846485138 |
| Loss at step 200: 0.03846244886517525 |
| Loss at step 250: 0.03738200664520264 |
| Loss at step 300: 0.0409412682056427 |
| Loss at step 350: 0.03415956720709801 |
| Loss at step 400: 0.04453108832240105 |
| Loss at step 450: 0.07758844643831253 |
| Loss at step 500: 0.04702244699001312 |
| Loss at step 550: 0.048941150307655334 |
| Loss at step 600: 0.033970486372709274 |
| Loss at step 650: 0.042242471128702164 |
| Loss at step 700: 0.035817310214042664 |
| Loss at step 750: 0.039164334535598755 |
| Loss at step 800: 0.049258679151535034 |
| Loss at step 850: 0.03764650970697403 |
| Loss at step 900: 0.037753425538539886 |
| Mean training loss after epoch 77: 0.04369013253122822 |
|
|
| EPOCH: 78 |
| Loss at step 0: 0.03975120559334755 |
| Loss at step 50: 0.054156817495822906 |
| Loss at step 100: 0.037778813391923904 |
| Loss at step 150: 0.04597756266593933 |
| Loss at step 200: 0.05560242384672165 |
| Loss at step 250: 0.038838405162096024 |
| Loss at step 300: 0.04676831513643265 |
| Loss at step 350: 0.0348576083779335 |
| Loss at step 400: 0.0406932570040226 |
| Loss at step 450: 0.03799765929579735 |
| Loss at step 500: 0.056081756949424744 |
| Loss at step 550: 0.036105211824178696 |
| Loss at step 600: 0.041250597685575485 |
| Loss at step 650: 0.05187376216053963 |
| Loss at step 700: 0.039942413568496704 |
| Loss at step 750: 0.03421250730752945 |
| Loss at step 800: 0.04465855285525322 |
| Loss at step 850: 0.04682165011763573 |
| Loss at step 900: 0.035167813301086426 |
| Mean training loss after epoch 78: 0.043499736446958745 |
|
|
| EPOCH: 79 |
| Loss at step 0: 0.04953160509467125 |
| Loss at step 50: 0.032434191554784775 |
| Loss at step 100: 0.060896556824445724 |
| Loss at step 150: 0.04916340485215187 |
| Loss at step 200: 0.04294075071811676 |
| Loss at step 250: 0.03761627897620201 |
| Loss at step 300: 0.05153829976916313 |
| Loss at step 350: 0.04079978168010712 |
| Loss at step 400: 0.03249897435307503 |
| Loss at step 450: 0.04026184603571892 |
| Loss at step 500: 0.04575842246413231 |
| Loss at step 550: 0.04249049723148346 |
| Loss at step 600: 0.043605584651231766 |
| Loss at step 650: 0.036947764456272125 |
| Loss at step 700: 0.04640074074268341 |
| Loss at step 750: 0.05125853419303894 |
| Loss at step 800: 0.053849928081035614 |
| Loss at step 850: 0.05566077306866646 |
| Loss at step 900: 0.045117028057575226 |
| Mean training loss after epoch 79: 0.04346864549383553 |
|
|
| EPOCH: 80 |
| Loss at step 0: 0.06714697927236557 |
| Loss at step 50: 0.04404205456376076 |
| Loss at step 100: 0.04033733159303665 |
| Loss at step 150: 0.05670079216361046 |
| Loss at step 200: 0.043069370090961456 |
| Loss at step 250: 0.06097843125462532 |
| Loss at step 300: 0.028987368568778038 |
| Loss at step 350: 0.039095696061849594 |
| Loss at step 400: 0.05482931435108185 |
| Loss at step 450: 0.03586386889219284 |
| Loss at step 500: 0.041629496961832047 |
| Loss at step 550: 0.03990510106086731 |
| Loss at step 600: 0.054544877260923386 |
| Loss at step 650: 0.03798103332519531 |
| Loss at step 700: 0.04047699645161629 |
| Loss at step 750: 0.037949416786432266 |
| Loss at step 800: 0.0359608419239521 |
| Loss at step 850: 0.054671015590429306 |
| Loss at step 900: 0.038619618862867355 |
| Mean training loss after epoch 80: 0.043645817013993574 |
|
|
| EPOCH: 81 |
| Loss at step 0: 0.041961781680583954 |
| Loss at step 50: 0.040087319910526276 |
| Loss at step 100: 0.05675582215189934 |
| Loss at step 150: 0.05277867987751961 |
| Loss at step 200: 0.03047415055334568 |
| Loss at step 250: 0.03613193705677986 |
| Loss at step 300: 0.043731123208999634 |
| Loss at step 350: 0.04356745257973671 |
| Loss at step 400: 0.03996061906218529 |
| Loss at step 450: 0.03311565890908241 |
| Loss at step 500: 0.04055457562208176 |
| Loss at step 550: 0.05245433375239372 |
| Loss at step 600: 0.038351498544216156 |
| Loss at step 650: 0.05595117434859276 |
| Loss at step 700: 0.03396110609173775 |
| Loss at step 750: 0.0379931665956974 |
| Loss at step 800: 0.03337615355849266 |
| Loss at step 850: 0.0378812812268734 |
| Loss at step 900: 0.03898458927869797 |
| Mean training loss after epoch 81: 0.0442062398152692 |
|
|
| EPOCH: 82 |
| Loss at step 0: 0.06833680719137192 |
| Loss at step 50: 0.04500220715999603 |
| Loss at step 100: 0.03772377222776413 |
| Loss at step 150: 0.053365763276815414 |
| Loss at step 200: 0.03693930804729462 |
| Loss at step 250: 0.04482037574052811 |
| Loss at step 300: 0.049105748534202576 |
| Loss at step 350: 0.04576117917895317 |
| Loss at step 400: 0.03354388102889061 |
| Loss at step 450: 0.04362288862466812 |
| Loss at step 500: 0.037595588713884354 |
| Loss at step 550: 0.06382860988378525 |
| Loss at step 600: 0.03677208349108696 |
| Loss at step 650: 0.05604931339621544 |
| Loss at step 700: 0.05555162578821182 |
| Loss at step 750: 0.027041040360927582 |
| Loss at step 800: 0.04180339723825455 |
| Loss at step 850: 0.07140879333019257 |
| Loss at step 900: 0.04118446260690689 |
| Mean training loss after epoch 82: 0.04348311074443463 |
|
|
| EPOCH: 83 |
| Loss at step 0: 0.03259948268532753 |
| Loss at step 50: 0.05375359207391739 |
| Loss at step 100: 0.03813811019062996 |
| Loss at step 150: 0.04393210634589195 |
| Loss at step 200: 0.05306578427553177 |
| Loss at step 250: 0.05070703849196434 |
| Loss at step 300: 0.042646415531635284 |
| Loss at step 350: 0.052681658416986465 |
| Loss at step 400: 0.039171285927295685 |
| Loss at step 450: 0.050351619720458984 |
| Loss at step 500: 0.036338597536087036 |
| Loss at step 550: 0.04029726982116699 |
| Loss at step 600: 0.04205435514450073 |
| Loss at step 650: 0.0543961375951767 |
| Loss at step 700: 0.027284376323223114 |
| Loss at step 750: 0.036568958312273026 |
| Loss at step 800: 0.04350341111421585 |
| Loss at step 850: 0.05256534367799759 |
| Loss at step 900: 0.03388596326112747 |
| Mean training loss after epoch 83: 0.04297601644084779 |
|
|
| EPOCH: 84 |
| Loss at step 0: 0.04750841483473778 |
| Loss at step 50: 0.04352778568863869 |
| Loss at step 100: 0.03605922311544418 |
| Loss at step 150: 0.040445245802402496 |
| Loss at step 200: 0.044017449021339417 |
| Loss at step 250: 0.036079443991184235 |
| Loss at step 300: 0.03524768352508545 |
| Loss at step 350: 0.058157481253147125 |
| Loss at step 400: 0.03466608375310898 |
| Loss at step 450: 0.04185834154486656 |
| Loss at step 500: 0.039541035890579224 |
| Loss at step 550: 0.03365383669734001 |
| Loss at step 600: 0.032277606427669525 |
| Loss at step 650: 0.038126878440380096 |
| Loss at step 700: 0.06580884009599686 |
| Loss at step 750: 0.029152007773518562 |
| Loss at step 800: 0.05536936596035957 |
| Loss at step 850: 0.03807036951184273 |
| Loss at step 900: 0.03619582951068878 |
| Mean training loss after epoch 84: 0.043629243781269866 |
|
|
| EPOCH: 85 |
| Loss at step 0: 0.049788717180490494 |
| Loss at step 50: 0.038022421300411224 |
| Loss at step 100: 0.032273098826408386 |
| Loss at step 150: 0.03269397094845772 |
| Loss at step 200: 0.035025231540203094 |
| Loss at step 250: 0.05248252674937248 |
| Loss at step 300: 0.03256775811314583 |
| Loss at step 350: 0.03763484209775925 |
| Loss at step 400: 0.04265531152486801 |
| Loss at step 450: 0.03436531126499176 |
| Loss at step 500: 0.03978480398654938 |
| Loss at step 550: 0.03378177434206009 |
| Loss at step 600: 0.06930387765169144 |
| Loss at step 650: 0.03307141363620758 |
| Loss at step 700: 0.053521107882261276 |
| Loss at step 750: 0.053345534950494766 |
| Loss at step 800: 0.037429336458444595 |
| Loss at step 850: 0.043063025921583176 |
| Loss at step 900: 0.03620496019721031 |
| Mean training loss after epoch 85: 0.04344467626118075 |
|
|
| EPOCH: 86 |
| Loss at step 0: 0.07604561746120453 |
| Loss at step 50: 0.03270219638943672 |
| Loss at step 100: 0.048404015600681305 |
| Loss at step 150: 0.03520134463906288 |
| Loss at step 200: 0.04617032781243324 |
| Loss at step 250: 0.04424238204956055 |
| Loss at step 300: 0.05500755086541176 |
| Loss at step 350: 0.03861868381500244 |
| Loss at step 400: 0.041497714817523956 |
| Loss at step 450: 0.040724895894527435 |
| Loss at step 500: 0.05506446212530136 |
| Loss at step 550: 0.050323523581027985 |
| Loss at step 600: 0.04069007933139801 |
| Loss at step 650: 0.04165627062320709 |
| Loss at step 700: 0.03901602327823639 |
| Loss at step 750: 0.05073634907603264 |
| Loss at step 800: 0.057049982249736786 |
| Loss at step 850: 0.041814662516117096 |
| Loss at step 900: 0.050912611186504364 |
| Mean training loss after epoch 86: 0.04339223529603372 |
|
|
| EPOCH: 87 |
| Loss at step 0: 0.03832190856337547 |
| Loss at step 50: 0.03612116724252701 |
| Loss at step 100: 0.046430062502622604 |
| Loss at step 150: 0.0354364775121212 |
| Loss at step 200: 0.04589875787496567 |
| Loss at step 250: 0.046340517699718475 |
| Loss at step 300: 0.03358977660536766 |
| Loss at step 350: 0.05608464032411575 |
| Loss at step 400: 0.03659365326166153 |
| Loss at step 450: 0.031970780342817307 |
| Loss at step 500: 0.05655407905578613 |
| Loss at step 550: 0.035570062696933746 |
| Loss at step 600: 0.050886280834674835 |
| Loss at step 650: 0.035846032202243805 |
| Loss at step 700: 0.033220771700143814 |
| Loss at step 750: 0.05016414076089859 |
| Loss at step 800: 0.03249731287360191 |
| Loss at step 850: 0.051579348742961884 |
| Loss at step 900: 0.04383271932601929 |
| Mean training loss after epoch 87: 0.043103055565802656 |
|
|
| EPOCH: 88 |
| Loss at step 0: 0.059718307107686996 |
| Loss at step 50: 0.03696296736598015 |
| Loss at step 100: 0.05220537632703781 |
| Loss at step 150: 0.059678904712200165 |
| Loss at step 200: 0.036867622286081314 |
| Loss at step 250: 0.03667591139674187 |
| Loss at step 300: 0.034308452159166336 |
| Loss at step 350: 0.04058242589235306 |
| Loss at step 400: 0.03458724543452263 |
| Loss at step 450: 0.03771441802382469 |
| Loss at step 500: 0.03839738294482231 |
| Loss at step 550: 0.04058973491191864 |
| Loss at step 600: 0.039897527545690536 |
| Loss at step 650: 0.04819701239466667 |
| Loss at step 700: 0.039945267140865326 |
| Loss at step 750: 0.039837781339883804 |
| Loss at step 800: 0.03418014571070671 |
| Loss at step 850: 0.0561252199113369 |
| Loss at step 900: 0.03720290958881378 |
| Mean training loss after epoch 88: 0.043711327423434906 |
|
|
| EPOCH: 89 |
| Loss at step 0: 0.0622403658926487 |
| Loss at step 50: 0.03501512110233307 |
| Loss at step 100: 0.03378031775355339 |
| Loss at step 150: 0.03363748639822006 |
| Loss at step 200: 0.04308519512414932 |
| Loss at step 250: 0.04319875314831734 |
| Loss at step 300: 0.06976722180843353 |
| Loss at step 350: 0.03761878237128258 |
| Loss at step 400: 0.03650447353720665 |
| Loss at step 450: 0.03223029896616936 |
| Loss at step 500: 0.0345304012298584 |
| Loss at step 550: 0.04749958589673042 |
| Loss at step 600: 0.04270829260349274 |
| Loss at step 650: 0.05911377817392349 |
| Loss at step 700: 0.03142249584197998 |
| Loss at step 750: 0.04553729668259621 |
| Loss at step 800: 0.039227329194545746 |
| Loss at step 850: 0.0500837042927742 |
| Loss at step 900: 0.038529522716999054 |
| Mean training loss after epoch 89: 0.043296830720707044 |
|
|
| EPOCH: 90 |
| Loss at step 0: 0.03306594118475914 |
| Loss at step 50: 0.04650263860821724 |
| Loss at step 100: 0.04312523081898689 |
| Loss at step 150: 0.034545619040727615 |
| Loss at step 200: 0.04598233103752136 |
| Loss at step 250: 0.03660457953810692 |
| Loss at step 300: 0.04116194322705269 |
| Loss at step 350: 0.05145394057035446 |
| Loss at step 400: 0.05626528337597847 |
| Loss at step 450: 0.042840324342250824 |
| Loss at step 500: 0.03150201216340065 |
| Loss at step 550: 0.03885718807578087 |
| Loss at step 600: 0.04799865931272507 |
| Loss at step 650: 0.03733006864786148 |
| Loss at step 700: 0.03675772249698639 |
| Loss at step 750: 0.0403936468064785 |
| Loss at step 800: 0.060140881687402725 |
| Loss at step 850: 0.056603480130434036 |
| Loss at step 900: 0.03747309371829033 |
| Mean training loss after epoch 90: 0.04427998538202505 |
|
|
| EPOCH: 91 |
| Loss at step 0: 0.04910198971629143 |
| Loss at step 50: 0.04037807509303093 |
| Loss at step 100: 0.03771810233592987 |
| Loss at step 150: 0.034219253808259964 |
| Loss at step 200: 0.04347480833530426 |
| Loss at step 250: 0.047169867902994156 |
| Loss at step 300: 0.041771892458200455 |
| Loss at step 350: 0.038890402764081955 |
| Loss at step 400: 0.03771951049566269 |
| Loss at step 450: 0.027332905679941177 |
| Loss at step 500: 0.0472487211227417 |
| Loss at step 550: 0.03344062715768814 |
| Loss at step 600: 0.039007991552352905 |
| Loss at step 650: 0.03601592034101486 |
| Loss at step 700: 0.03445816412568092 |
| Loss at step 750: 0.040622398257255554 |
| Loss at step 800: 0.03840038552880287 |
| Loss at step 850: 0.05284346267580986 |
| Loss at step 900: 0.046714216470718384 |
| Mean training loss after epoch 91: 0.04356519312166901 |
|
|
| EPOCH: 92 |
| Loss at step 0: 0.07394585013389587 |
| Loss at step 50: 0.03972001373767853 |
| Loss at step 100: 0.03388422727584839 |
| Loss at step 150: 0.043385058641433716 |
| Loss at step 200: 0.0548548698425293 |
| Loss at step 250: 0.03832437843084335 |
| Loss at step 300: 0.05707750469446182 |
| Loss at step 350: 0.041515324264764786 |
| Loss at step 400: 0.031445201486349106 |
| Loss at step 450: 0.06388327479362488 |
| Loss at step 500: 0.037730999290943146 |
| Loss at step 550: 0.05023184418678284 |
| Loss at step 600: 0.0382477305829525 |
| Loss at step 650: 0.043892908841371536 |
| Loss at step 700: 0.03901918977499008 |
| Loss at step 750: 0.04354777932167053 |
| Loss at step 800: 0.040302760899066925 |
| Loss at step 850: 0.04244803264737129 |
| Loss at step 900: 0.05645808205008507 |
| Mean training loss after epoch 92: 0.042986952499158855 |
|
|
| EPOCH: 93 |
| Loss at step 0: 0.03539813309907913 |
| Loss at step 50: 0.048581965267658234 |
| Loss at step 100: 0.04163219407200813 |
| Loss at step 150: 0.03968052938580513 |
| Loss at step 200: 0.03193528577685356 |
| Loss at step 250: 0.0411832369863987 |
| Loss at step 300: 0.03856983408331871 |
| Loss at step 350: 0.038298722356557846 |
| Loss at step 400: 0.04309234768152237 |
| Loss at step 450: 0.040360040962696075 |
| Loss at step 500: 0.04634615406394005 |
| Loss at step 550: 0.06071379780769348 |
| Loss at step 600: 0.04664158821105957 |
| Loss at step 650: 0.03374982625246048 |
| Loss at step 700: 0.052322570234537125 |
| Loss at step 750: 0.04002566263079643 |
| Loss at step 800: 0.03766759857535362 |
| Loss at step 850: 0.05379931628704071 |
| Loss at step 900: 0.037691425532102585 |
| Mean training loss after epoch 93: 0.043927824297455204 |
|
|
| EPOCH: 94 |
| Loss at step 0: 0.06536293029785156 |
| Loss at step 50: 0.030876314267516136 |
| Loss at step 100: 0.03512885048985481 |
| Loss at step 150: 0.05157601460814476 |
| Loss at step 200: 0.07189564406871796 |
| Loss at step 250: 0.038722675293684006 |
| Loss at step 300: 0.03872859477996826 |
| Loss at step 350: 0.03599252551794052 |
| Loss at step 400: 0.042109329253435135 |
| Loss at step 450: 0.04384028911590576 |
| Loss at step 500: 0.036185961216688156 |
| Loss at step 550: 0.062151797115802765 |
| Loss at step 600: 0.03850793465971947 |
| Loss at step 650: 0.034010306000709534 |
| Loss at step 700: 0.04265720024704933 |
| Loss at step 750: 0.038343705236911774 |
| Loss at step 800: 0.0559229701757431 |
| Loss at step 850: 0.037223100662231445 |
| Loss at step 900: 0.03685881197452545 |
| Mean training loss after epoch 94: 0.04346638769983673 |
|
|
| EPOCH: 95 |
| Loss at step 0: 0.07410895079374313 |
| Loss at step 50: 0.030644426122307777 |
| Loss at step 100: 0.03632679581642151 |
| Loss at step 150: 0.04430529475212097 |
| Loss at step 200: 0.03993538022041321 |
| Loss at step 250: 0.033615030348300934 |
| Loss at step 300: 0.03780102729797363 |
| Loss at step 350: 0.03041430376470089 |
| Loss at step 400: 0.04900200292468071 |
| Loss at step 450: 0.04082078859210014 |
| Loss at step 500: 0.04287412762641907 |
| Loss at step 550: 0.04590524733066559 |
| Loss at step 600: 0.03649120777845383 |
| Loss at step 650: 0.032129764556884766 |
| Loss at step 700: 0.07543421536684036 |
| Loss at step 750: 0.037203382700681686 |
| Loss at step 800: 0.048887595534324646 |
| Loss at step 850: 0.036267053335905075 |
| Loss at step 900: 0.036513570696115494 |
| Mean training loss after epoch 95: 0.04287138573928619 |
|
|
| EPOCH: 96 |
| Loss at step 0: 0.03439642861485481 |
| Loss at step 50: 0.032979678362607956 |
| Loss at step 100: 0.03874444216489792 |
| Loss at step 150: 0.03373781591653824 |
| Loss at step 200: 0.03582676500082016 |
| Loss at step 250: 0.03851651772856712 |
| Loss at step 300: 0.03581823408603668 |
| Loss at step 350: 0.036770474165678024 |
| Loss at step 400: 0.03284694626927376 |
| Loss at step 450: 0.04079399257898331 |
| Loss at step 500: 0.04206319525837898 |
| Loss at step 550: 0.03856229409575462 |
| Loss at step 600: 0.04491313174366951 |
| Loss at step 650: 0.03701549395918846 |
| Loss at step 700: 0.053752969950437546 |
| Loss at step 750: 0.05399594083428383 |
| Loss at step 800: 0.0453498549759388 |
| Loss at step 850: 0.038302745670080185 |
| Loss at step 900: 0.04211289435625076 |
| Mean training loss after epoch 96: 0.0431959849816046 |
|
|
| EPOCH: 97 |
| Loss at step 0: 0.03641681745648384 |
| Loss at step 50: 0.03312516584992409 |
| Loss at step 100: 0.05817564204335213 |
| Loss at step 150: 0.03705442696809769 |
| Loss at step 200: 0.0319865457713604 |
| Loss at step 250: 0.040375035256147385 |
| Loss at step 300: 0.059476666152477264 |
| Loss at step 350: 0.048445623368024826 |
| Loss at step 400: 0.03463175892829895 |
| Loss at step 450: 0.037774257361888885 |
| Loss at step 500: 0.0523386187851429 |
| Loss at step 550: 0.0657649040222168 |
| Loss at step 600: 0.04200076311826706 |
| Loss at step 650: 0.0315200611948967 |
| Loss at step 700: 0.04242639243602753 |
| Loss at step 750: 0.03878338262438774 |
| Loss at step 800: 0.05410474166274071 |
| Loss at step 850: 0.03390758857131004 |
| Loss at step 900: 0.051311835646629333 |
| Mean training loss after epoch 97: 0.04318891692040826 |
|
|
| EPOCH: 98 |
| Loss at step 0: 0.046574775129556656 |
| Loss at step 50: 0.05641099810600281 |
| Loss at step 100: 0.03675806149840355 |
| Loss at step 150: 0.04064656049013138 |
| Loss at step 200: 0.041404858231544495 |
| Loss at step 250: 0.046953871846199036 |
| Loss at step 300: 0.03851976990699768 |
| Loss at step 350: 0.058378178626298904 |
| Loss at step 400: 0.04649720713496208 |
| Loss at step 450: 0.03760731220245361 |
| Loss at step 500: 0.03642148897051811 |
| Loss at step 550: 0.050206299871206284 |
| Loss at step 600: 0.04067458212375641 |
| Loss at step 650: 0.04598100110888481 |
| Loss at step 700: 0.06879585981369019 |
| Loss at step 750: 0.039883676916360855 |
| Loss at step 800: 0.05262867361307144 |
| Loss at step 850: 0.03709295392036438 |
| Loss at step 900: 0.03144507482647896 |
| Mean training loss after epoch 98: 0.043273623160986124 |
|
|
| EPOCH: 99 |
| Loss at step 0: 0.05169651284813881 |
| Loss at step 50: 0.04774666577577591 |
| Loss at step 100: 0.04283786192536354 |
| Loss at step 150: 0.05346093699336052 |
| Loss at step 200: 0.06267943233251572 |
| Loss at step 250: 0.029984652996063232 |
| Loss at step 300: 0.04070792719721794 |
| Loss at step 350: 0.042360469698905945 |
| Loss at step 400: 0.04965902864933014 |
| Loss at step 450: 0.034532371908426285 |
| Loss at step 500: 0.031091198325157166 |
| Loss at step 550: 0.0353475958108902 |
| Loss at step 600: 0.03562759980559349 |
| Loss at step 650: 0.05511445179581642 |
| Loss at step 700: 0.04308035969734192 |
| Loss at step 750: 0.039943136274814606 |
| Loss at step 800: 0.03640330582857132 |
| Loss at step 850: 0.052483201026916504 |
| Loss at step 900: 0.038298215717077255 |
| Mean training loss after epoch 99: 0.04306189255761122 |
|
|
| EPOCH: 100 |
| Loss at step 0: 0.03868278115987778 |
| Loss at step 50: 0.033273063600063324 |
| Loss at step 100: 0.06904282420873642 |
| Loss at step 150: 0.04511026293039322 |
| Loss at step 200: 0.03361647576093674 |
| Loss at step 250: 0.04274805635213852 |
| Loss at step 300: 0.04436466470360756 |
| Loss at step 350: 0.06252449005842209 |
| Loss at step 400: 0.04299504682421684 |
| Loss at step 450: 0.0542532280087471 |
| Loss at step 500: 0.0670432448387146 |
| Loss at step 550: 0.0696505531668663 |
| Loss at step 600: 0.059786610305309296 |
| Loss at step 650: 0.041346460580825806 |
| Loss at step 700: 0.05354555323719978 |
| Loss at step 750: 0.048287492245435715 |
| Loss at step 800: 0.03784614056348801 |
| Loss at step 850: 0.03387393057346344 |
| Loss at step 900: 0.03863274306058884 |
| Mean training loss after epoch 100: 0.04288483024842894 |
|
|
| EPOCH: 101 |
| Loss at step 0: 0.0320252887904644 |
| Loss at step 50: 0.03704806789755821 |
| Loss at step 100: 0.04036383330821991 |
| Loss at step 150: 0.04344441741704941 |
| Loss at step 200: 0.036835040897130966 |
| Loss at step 250: 0.03519628942012787 |
| Loss at step 300: 0.038933347910642624 |
| Loss at step 350: 0.040308840572834015 |
| Loss at step 400: 0.06893625110387802 |
| Loss at step 450: 0.03442879393696785 |
| Loss at step 500: 0.06981664896011353 |
| Loss at step 550: 0.05139239877462387 |
| Loss at step 600: 0.045361023396253586 |
| Loss at step 650: 0.03686761483550072 |
| Loss at step 700: 0.03900550678372383 |
| Loss at step 750: 0.033268269151449203 |
| Loss at step 800: 0.036813121289014816 |
| Loss at step 850: 0.03814901039004326 |
| Loss at step 900: 0.05475571006536484 |
| Mean training loss after epoch 101: 0.04321747353828665 |
|
|
| EPOCH: 102 |
| Loss at step 0: 0.05565916746854782 |
| Loss at step 50: 0.0469290092587471 |
| Loss at step 100: 0.03592206537723541 |
| Loss at step 150: 0.03643639758229256 |
| Loss at step 200: 0.06259587407112122 |
| Loss at step 250: 0.0463993102312088 |
| Loss at step 300: 0.03069981187582016 |
| Loss at step 350: 0.06601865589618683 |
| Loss at step 400: 0.03276149928569794 |
| Loss at step 450: 0.02996491827070713 |
| Loss at step 500: 0.0397891029715538 |
| Loss at step 550: 0.04958747699856758 |
| Loss at step 600: 0.05864010751247406 |
| Loss at step 650: 0.04672401770949364 |
| Loss at step 700: 0.04763220250606537 |
| Loss at step 750: 0.03473920747637749 |
| Loss at step 800: 0.037696756422519684 |
| Loss at step 850: 0.0555654838681221 |
| Loss at step 900: 0.05053749680519104 |
| Mean training loss after epoch 102: 0.043392531085274876 |
|
|
| EPOCH: 103 |
| Loss at step 0: 0.03874469920992851 |
| Loss at step 50: 0.05485078692436218 |
| Loss at step 100: 0.03334563225507736 |
| Loss at step 150: 0.03602603077888489 |
| Loss at step 200: 0.05366092175245285 |
| Loss at step 250: 0.05623263493180275 |
| Loss at step 300: 0.04819313436746597 |
| Loss at step 350: 0.05220837891101837 |
| Loss at step 400: 0.03978671878576279 |
| Loss at step 450: 0.05293075740337372 |
| Loss at step 500: 0.03283224254846573 |
| Loss at step 550: 0.03934606537222862 |
| Loss at step 600: 0.039841100573539734 |
| Loss at step 650: 0.038126807659864426 |
| Loss at step 700: 0.04651801660656929 |
| Loss at step 750: 0.03205832839012146 |
| Loss at step 800: 0.03260406479239464 |
| Loss at step 850: 0.04802095517516136 |
| Loss at step 900: 0.07081270962953568 |
| Mean training loss after epoch 103: 0.04249289364758522 |
|
|
| EPOCH: 104 |
| Loss at step 0: 0.03115161508321762 |
| Loss at step 50: 0.03379862755537033 |
| Loss at step 100: 0.041275929659605026 |
| Loss at step 150: 0.039807382971048355 |
| Loss at step 200: 0.04076811298727989 |
| Loss at step 250: 0.03608894720673561 |
| Loss at step 300: 0.03752580285072327 |
| Loss at step 350: 0.046833109110593796 |
| Loss at step 400: 0.051159657537937164 |
| Loss at step 450: 0.03741542249917984 |
| Loss at step 500: 0.038290105760097504 |
| Loss at step 550: 0.041247326880693436 |
| Loss at step 600: 0.04058607295155525 |
| Loss at step 650: 0.03616589680314064 |
| Loss at step 700: 0.04535121098160744 |
| Loss at step 750: 0.03494952619075775 |
| Loss at step 800: 0.03949456289410591 |
| Loss at step 850: 0.040163662284612656 |
| Loss at step 900: 0.03252166882157326 |
| Mean training loss after epoch 104: 0.043015945414855664 |
|
|
| EPOCH: 105 |
| Loss at step 0: 0.036242932081222534 |
| Loss at step 50: 0.03272474184632301 |
| Loss at step 100: 0.04492475464940071 |
| Loss at step 150: 0.045704569667577744 |
| Loss at step 200: 0.036552704870700836 |
| Loss at step 250: 0.041963860392570496 |
| Loss at step 300: 0.03921288624405861 |
| Loss at step 350: 0.04874899238348007 |
| Loss at step 400: 0.04045712947845459 |
| Loss at step 450: 0.030772091820836067 |
| Loss at step 500: 0.05106096714735031 |
| Loss at step 550: 0.027127612382173538 |
| Loss at step 600: 0.05295705795288086 |
| Loss at step 650: 0.04706268757581711 |
| Loss at step 700: 0.03845628723502159 |
| Loss at step 750: 0.03371085971593857 |
| Loss at step 800: 0.04142371937632561 |
| Loss at step 850: 0.052407145500183105 |
| Loss at step 900: 0.037026021629571915 |
| Mean training loss after epoch 105: 0.04317484780558264 |
|
|
| EPOCH: 106 |
| Loss at step 0: 0.029048895463347435 |
| Loss at step 50: 0.04750058054924011 |
| Loss at step 100: 0.03844717517495155 |
| Loss at step 150: 0.032337259501218796 |
| Loss at step 200: 0.03150492161512375 |
| Loss at step 250: 0.05508747696876526 |
| Loss at step 300: 0.04623647406697273 |
| Loss at step 350: 0.03420146927237511 |
| Loss at step 400: 0.054901327937841415 |
| Loss at step 450: 0.044081784784793854 |
| Loss at step 500: 0.05747245252132416 |
| Loss at step 550: 0.04383234679698944 |
| Loss at step 600: 0.03540794178843498 |
| Loss at step 650: 0.049649763852357864 |
| Loss at step 700: 0.03573313355445862 |
| Loss at step 750: 0.044091228395700455 |
| Loss at step 800: 0.03443525359034538 |
| Loss at step 850: 0.04229153320193291 |
| Loss at step 900: 0.058335255831480026 |
| Mean training loss after epoch 106: 0.04258429711219916 |
|
|
| EPOCH: 107 |
| Loss at step 0: 0.0381803885102272 |
| Loss at step 50: 0.03685297816991806 |
| Loss at step 100: 0.034790556877851486 |
| Loss at step 150: 0.05032115802168846 |
| Loss at step 200: 0.033282600343227386 |
| Loss at step 250: 0.04043008014559746 |
| Loss at step 300: 0.03712046146392822 |
| Loss at step 350: 0.047968342900276184 |
| Loss at step 400: 0.043439656496047974 |
| Loss at step 450: 0.043730784207582474 |
| Loss at step 500: 0.03656981140375137 |
| Loss at step 550: 0.042178064584732056 |
| Loss at step 600: 0.04220736026763916 |
| Loss at step 650: 0.03219039365649223 |
| Loss at step 700: 0.03509557992219925 |
| Loss at step 750: 0.04011481627821922 |
| Loss at step 800: 0.029533641412854195 |
| Loss at step 850: 0.0380370207130909 |
| Loss at step 900: 0.03835464268922806 |
| Mean training loss after epoch 107: 0.042455973341933954 |
|
|
| EPOCH: 108 |
| Loss at step 0: 0.053215205669403076 |
| Loss at step 50: 0.03992394357919693 |
| Loss at step 100: 0.03500846028327942 |
| Loss at step 150: 0.05199325084686279 |
| Loss at step 200: 0.03509242460131645 |
| Loss at step 250: 0.033662501722574234 |
| Loss at step 300: 0.0520416721701622 |
| Loss at step 350: 0.03401225432753563 |
| Loss at step 400: 0.04268940910696983 |
| Loss at step 450: 0.036937568336725235 |
| Loss at step 500: 0.053192269057035446 |
| Loss at step 550: 0.029121991246938705 |
| Loss at step 600: 0.035446252673864365 |
| Loss at step 650: 0.05657226964831352 |
| Loss at step 700: 0.040138814598321915 |
| Loss at step 750: 0.042686160653829575 |
| Loss at step 800: 0.03484735265374184 |
| Loss at step 850: 0.03614305332303047 |
| Loss at step 900: 0.04524796083569527 |
| Mean training loss after epoch 108: 0.04264132607021311 |
|
|
| EPOCH: 109 |
| Loss at step 0: 0.03732350468635559 |
| Loss at step 50: 0.03490496426820755 |
| Loss at step 100: 0.039503492414951324 |
| Loss at step 150: 0.03732747957110405 |
| Loss at step 200: 0.036649059504270554 |
| Loss at step 250: 0.04840293154120445 |
| Loss at step 300: 0.04136611893773079 |
| Loss at step 350: 0.03097102977335453 |
| Loss at step 400: 0.03997035697102547 |
| Loss at step 450: 0.040247365832328796 |
| Loss at step 500: 0.033872444182634354 |
| Loss at step 550: 0.043919648975133896 |
| Loss at step 600: 0.03842050954699516 |
| Loss at step 650: 0.03633081167936325 |
| Loss at step 700: 0.03185936436057091 |
| Loss at step 750: 0.031556472182273865 |
| Loss at step 800: 0.04246347025036812 |
| Loss at step 850: 0.03699137642979622 |
| Loss at step 900: 0.038073956966400146 |
| Mean training loss after epoch 109: 0.042871232231908134 |
|
|
| EPOCH: 110 |
| Loss at step 0: 0.044787343591451645 |
| Loss at step 50: 0.04396039992570877 |
| Loss at step 100: 0.03304486349225044 |
| Loss at step 150: 0.0329318642616272 |
| Loss at step 200: 0.03712211921811104 |
| Loss at step 250: 0.03624889254570007 |
| Loss at step 300: 0.039950963109731674 |
| Loss at step 350: 0.05183451250195503 |
| Loss at step 400: 0.033772554248571396 |
| Loss at step 450: 0.03334091603755951 |
| Loss at step 500: 0.04075763002038002 |
| Loss at step 550: 0.06279843300580978 |
| Loss at step 600: 0.05321997031569481 |
| Loss at step 650: 0.037117939442396164 |
| Loss at step 700: 0.03801078721880913 |
| Loss at step 750: 0.032297197729349136 |
| Loss at step 800: 0.049598902463912964 |
| Loss at step 850: 0.06499156355857849 |
| Loss at step 900: 0.034701526165008545 |
| Mean training loss after epoch 110: 0.04253625459095308 |
|
|
| EPOCH: 111 |
| Loss at step 0: 0.04842279478907585 |
| Loss at step 50: 0.030813124030828476 |
| Loss at step 100: 0.03582942485809326 |
| Loss at step 150: 0.04401121661067009 |
| Loss at step 200: 0.03099776618182659 |
| Loss at step 250: 0.039740726351737976 |
| Loss at step 300: 0.049139585345983505 |
| Loss at step 350: 0.032573338598012924 |
| Loss at step 400: 0.0320868156850338 |
| Loss at step 450: 0.03244072571396828 |
| Loss at step 500: 0.036660704761743546 |
| Loss at step 550: 0.05817306786775589 |
| Loss at step 600: 0.046441372483968735 |
| Loss at step 650: 0.04924897104501724 |
| Loss at step 700: 0.032720550894737244 |
| Loss at step 750: 0.03674580901861191 |
| Loss at step 800: 0.0348411463201046 |
| Loss at step 850: 0.0344211645424366 |
| Loss at step 900: 0.03544338792562485 |
| Mean training loss after epoch 111: 0.04321895086808182 |
|
|
| EPOCH: 112 |
| Loss at step 0: 0.02806468866765499 |
| Loss at step 50: 0.05090704187750816 |
| Loss at step 100: 0.04020220413804054 |
| Loss at step 150: 0.04250871017575264 |
| Loss at step 200: 0.041078586131334305 |
| Loss at step 250: 0.049902867525815964 |
| Loss at step 300: 0.03141561150550842 |
| Loss at step 350: 0.04201183468103409 |
| Loss at step 400: 0.055819183588027954 |
| Loss at step 450: 0.03885535150766373 |
| Loss at step 500: 0.06769788265228271 |
| Loss at step 550: 0.039124418050050735 |
| Loss at step 600: 0.0550897940993309 |
| Loss at step 650: 0.05477220192551613 |
| Loss at step 700: 0.03775335103273392 |
| Loss at step 750: 0.03932926058769226 |
| Loss at step 800: 0.03385027498006821 |
| Loss at step 850: 0.03438795357942581 |
| Loss at step 900: 0.04016561433672905 |
| Mean training loss after epoch 112: 0.042442821055603056 |
|
|
| EPOCH: 113 |
| Loss at step 0: 0.03439236432313919 |
| Loss at step 50: 0.04079783707857132 |
| Loss at step 100: 0.04035534709692001 |
| Loss at step 150: 0.02992972545325756 |
| Loss at step 200: 0.03763053938746452 |
| Loss at step 250: 0.03948152810335159 |
| Loss at step 300: 0.05204397067427635 |
| Loss at step 350: 0.03488073870539665 |
| Loss at step 400: 0.039366453886032104 |
| Loss at step 450: 0.07258787006139755 |
| Loss at step 500: 0.03774897754192352 |
| Loss at step 550: 0.03692486137151718 |
| Loss at step 600: 0.03373141214251518 |
| Loss at step 650: 0.038297783583402634 |
| Loss at step 700: 0.03423190861940384 |
| Loss at step 750: 0.036938928067684174 |
| Loss at step 800: 0.02934008091688156 |
| Loss at step 850: 0.03978004679083824 |
| Loss at step 900: 0.05455329269170761 |
| Mean training loss after epoch 113: 0.04289359403357132 |
|
|
| EPOCH: 114 |
| Loss at step 0: 0.04929478093981743 |
| Loss at step 50: 0.036822691559791565 |
| Loss at step 100: 0.033721890300512314 |
| Loss at step 150: 0.034397274255752563 |
| Loss at step 200: 0.044635675847530365 |
| Loss at step 250: 0.04207928106188774 |
| Loss at step 300: 0.039686571806669235 |
| Loss at step 350: 0.05589856952428818 |
| Loss at step 400: 0.03962375968694687 |
| Loss at step 450: 0.02821052446961403 |
| Loss at step 500: 0.059994060546159744 |
| Loss at step 550: 0.035869430750608444 |
| Loss at step 600: 0.035884320735931396 |
| Loss at step 650: 0.03565568849444389 |
| Loss at step 700: 0.03159397095441818 |
| Loss at step 750: 0.038982052356004715 |
| Loss at step 800: 0.03397154062986374 |
| Loss at step 850: 0.03521180897951126 |
| Loss at step 900: 0.036220405250787735 |
| Mean training loss after epoch 114: 0.042887798532335236 |
|
|
| EPOCH: 115 |
| Loss at step 0: 0.03161928057670593 |
| Loss at step 50: 0.03745435178279877 |
| Loss at step 100: 0.03620803356170654 |
| Loss at step 150: 0.061530373990535736 |
| Loss at step 200: 0.03712744638323784 |
| Loss at step 250: 0.053335703909397125 |
| Loss at step 300: 0.03177780285477638 |
| Loss at step 350: 0.04157339781522751 |
| Loss at step 400: 0.04165922477841377 |
| Loss at step 450: 0.035079650580883026 |
| Loss at step 500: 0.048865292221307755 |
| Loss at step 550: 0.03243790939450264 |
| Loss at step 600: 0.04753963276743889 |
| Loss at step 650: 0.03399026021361351 |
| Loss at step 700: 0.04165761172771454 |
| Loss at step 750: 0.03889331966638565 |
| Loss at step 800: 0.0683445930480957 |
| Loss at step 850: 0.029029441997408867 |
| Loss at step 900: 0.0362982414662838 |
| Mean training loss after epoch 115: 0.04242513539480058 |
|
|
| EPOCH: 116 |
| Loss at step 0: 0.04920484870672226 |
| Loss at step 50: 0.05231902003288269 |
| Loss at step 100: 0.05306987091898918 |
| Loss at step 150: 0.05240674689412117 |
| Loss at step 200: 0.030451014637947083 |
| Loss at step 250: 0.03890807926654816 |
| Loss at step 300: 0.05659289285540581 |
| Loss at step 350: 0.036650046706199646 |
| Loss at step 400: 0.0346701517701149 |
| Loss at step 450: 0.03688763082027435 |
| Loss at step 500: 0.036767538636922836 |
| Loss at step 550: 0.032335709780454636 |
| Loss at step 600: 0.05310001224279404 |
| Loss at step 650: 0.05066174641251564 |
| Loss at step 700: 0.05107351019978523 |
| Loss at step 750: 0.03427013009786606 |
| Loss at step 800: 0.04448677599430084 |
| Loss at step 850: 0.034152865409851074 |
| Loss at step 900: 0.03578594699501991 |
| Mean training loss after epoch 116: 0.04271252417583456 |
|
|
| EPOCH: 117 |
| Loss at step 0: 0.041914403438568115 |
| Loss at step 50: 0.03944316506385803 |
| Loss at step 100: 0.036425188183784485 |
| Loss at step 150: 0.052166588604450226 |
| Loss at step 200: 0.03263081610202789 |
| Loss at step 250: 0.05043771117925644 |
| Loss at step 300: 0.042433347553014755 |
| Loss at step 350: 0.04409367963671684 |
| Loss at step 400: 0.05648966133594513 |
| Loss at step 450: 0.03183215856552124 |
| Loss at step 500: 0.0522194467484951 |
| Loss at step 550: 0.03536583483219147 |
| Loss at step 600: 0.037089429795742035 |
| Loss at step 650: 0.03506674990057945 |
| Loss at step 700: 0.07340752333402634 |
| Loss at step 750: 0.03434182330965996 |
| Loss at step 800: 0.05077645555138588 |
| Loss at step 850: 0.03686201199889183 |
| Loss at step 900: 0.04532172158360481 |
| Mean training loss after epoch 117: 0.04204082982872785 |
|
|
| EPOCH: 118 |
| Loss at step 0: 0.0334637425839901 |
| Loss at step 50: 0.04286329075694084 |
| Loss at step 100: 0.039089079946279526 |
| Loss at step 150: 0.034637052565813065 |
| Loss at step 200: 0.03227698802947998 |
| Loss at step 250: 0.042662475258111954 |
| Loss at step 300: 0.04539114609360695 |
| Loss at step 350: 0.046212777495384216 |
| Loss at step 400: 0.04341699928045273 |
| Loss at step 450: 0.03234335035085678 |
| Loss at step 500: 0.03549254313111305 |
| Loss at step 550: 0.035518109798431396 |
| Loss at step 600: 0.05670395493507385 |
| Loss at step 650: 0.05329463258385658 |
| Loss at step 700: 0.05610251426696777 |
| Loss at step 750: 0.03034931793808937 |
| Loss at step 800: 0.05315934866666794 |
| Loss at step 850: 0.03849770873785019 |
| Loss at step 900: 0.043866049498319626 |
| Mean training loss after epoch 118: 0.04233125694540899 |
|
|
| EPOCH: 119 |
| Loss at step 0: 0.041059985756874084 |
| Loss at step 50: 0.03781072795391083 |
| Loss at step 100: 0.058324579149484634 |
| Loss at step 150: 0.05159909278154373 |
| Loss at step 200: 0.03973580151796341 |
| Loss at step 250: 0.034565798938274384 |
| Loss at step 300: 0.03539511188864708 |
| Loss at step 350: 0.04175569489598274 |
| Loss at step 400: 0.033448122441768646 |
| Loss at step 450: 0.04465514048933983 |
| Loss at step 500: 0.035944875329732895 |
| Loss at step 550: 0.03353596478700638 |
| Loss at step 600: 0.03798174858093262 |
| Loss at step 650: 0.03126048669219017 |
| Loss at step 700: 0.035585448145866394 |
| Loss at step 750: 0.04394625499844551 |
| Loss at step 800: 0.03175988048315048 |
| Loss at step 850: 0.048157915472984314 |
| Loss at step 900: 0.05564304068684578 |
| Mean training loss after epoch 119: 0.04225547817438396 |
|
|
| EPOCH: 120 |
| Loss at step 0: 0.03654998168349266 |
| Loss at step 50: 0.04451896995306015 |
| Loss at step 100: 0.03440498933196068 |
| Loss at step 150: 0.045890774577856064 |
| Loss at step 200: 0.06556855142116547 |
| Loss at step 250: 0.04978479444980621 |
| Loss at step 300: 0.05234746262431145 |
| Loss at step 350: 0.048763785511255264 |
| Loss at step 400: 0.03686203435063362 |
| Loss at step 450: 0.0392894446849823 |
| Loss at step 500: 0.02784913033246994 |
| Loss at step 550: 0.060626059770584106 |
| Loss at step 600: 0.028441544622182846 |
| Loss at step 650: 0.03267563879489899 |
| Loss at step 700: 0.07281859219074249 |
| Loss at step 750: 0.03018386848270893 |
| Loss at step 800: 0.05207233875989914 |
| Loss at step 850: 0.03853049501776695 |
| Loss at step 900: 0.03876703232526779 |
| Mean training loss after epoch 120: 0.04244164083558105 |
|
|
| EPOCH: 121 |
| Loss at step 0: 0.032931819558143616 |
| Loss at step 50: 0.03577417507767677 |
| Loss at step 100: 0.05528498440980911 |
| Loss at step 150: 0.043831828981637955 |
| Loss at step 200: 0.040254004299640656 |
| Loss at step 250: 0.03872199356555939 |
| Loss at step 300: 0.05530036985874176 |
| Loss at step 350: 0.047019802033901215 |
| Loss at step 400: 0.05679032951593399 |
| Loss at step 450: 0.051629483699798584 |
| Loss at step 500: 0.04071091488003731 |
| Loss at step 550: 0.032426148653030396 |
| Loss at step 600: 0.03054644539952278 |
| Loss at step 650: 0.035769470036029816 |
| Loss at step 700: 0.04064173623919487 |
| Loss at step 750: 0.03792702406644821 |
| Loss at step 800: 0.0382523275911808 |
| Loss at step 850: 0.04808889329433441 |
| Loss at step 900: 0.03021102026104927 |
| Mean training loss after epoch 121: 0.042562772758177984 |
|
|
| EPOCH: 122 |
| Loss at step 0: 0.03836292773485184 |
| Loss at step 50: 0.037795569747686386 |
| Loss at step 100: 0.037729810923337936 |
| Loss at step 150: 0.05760970711708069 |
| Loss at step 200: 0.03326259180903435 |
| Loss at step 250: 0.03908127546310425 |
| Loss at step 300: 0.05211610719561577 |
| Loss at step 350: 0.03992561250925064 |
| Loss at step 400: 0.03279055282473564 |
| Loss at step 450: 0.054718513041734695 |
| Loss at step 500: 0.028107738122344017 |
| Loss at step 550: 0.037835340946912766 |
| Loss at step 600: 0.0400877371430397 |
| Loss at step 650: 0.040323756635189056 |
| Loss at step 700: 0.050479333847761154 |
| Loss at step 750: 0.03604499623179436 |
| Loss at step 800: 0.061530329287052155 |
| Loss at step 850: 0.0346403643488884 |
| Loss at step 900: 0.042586684226989746 |
| Mean training loss after epoch 122: 0.042738208541277245 |
|
|
| EPOCH: 123 |
| Loss at step 0: 0.037863537669181824 |
| Loss at step 50: 0.04166312888264656 |
| Loss at step 100: 0.0541631244122982 |
| Loss at step 150: 0.03443480655550957 |
| Loss at step 200: 0.03567218780517578 |
| Loss at step 250: 0.07678977400064468 |
| Loss at step 300: 0.035183656960725784 |
| Loss at step 350: 0.04932482913136482 |
| Loss at step 400: 0.03603409230709076 |
| Loss at step 450: 0.045590754598379135 |
| Loss at step 500: 0.050587721168994904 |
| Loss at step 550: 0.03978006914258003 |
| Loss at step 600: 0.03327057883143425 |
| Loss at step 650: 0.032649945467710495 |
| Loss at step 700: 0.040164392441511154 |
| Loss at step 750: 0.054399918764829636 |
| Loss at step 800: 0.03373962640762329 |
| Loss at step 850: 0.037361081689596176 |
| Loss at step 900: 0.040843669325113297 |
| Mean training loss after epoch 123: 0.04323216850188241 |
|
|
| EPOCH: 124 |
| Loss at step 0: 0.054330259561538696 |
| Loss at step 50: 0.034485798329114914 |
| Loss at step 100: 0.03967555612325668 |
| Loss at step 150: 0.033644016832113266 |
| Loss at step 200: 0.08355532586574554 |
| Loss at step 250: 0.040038276463747025 |
| Loss at step 300: 0.059040505439043045 |
| Loss at step 350: 0.036361414939165115 |
| Loss at step 400: 0.032022830098867416 |
| Loss at step 450: 0.06843051314353943 |
| Loss at step 500: 0.03649810701608658 |
| Loss at step 550: 0.03632408380508423 |
| Loss at step 600: 0.04380091652274132 |
| Loss at step 650: 0.049652453511953354 |
| Loss at step 700: 0.031190911307930946 |
| Loss at step 750: 0.03930390998721123 |
| Loss at step 800: 0.03655650466680527 |
| Loss at step 850: 0.0601528100669384 |
| Loss at step 900: 0.07293304800987244 |
| Mean training loss after epoch 124: 0.043018826279542975 |
|
|
| EPOCH: 125 |
| Loss at step 0: 0.03883155435323715 |
| Loss at step 50: 0.06450485438108444 |
| Loss at step 100: 0.046584781259298325 |
| Loss at step 150: 0.050592802464962006 |
| Loss at step 200: 0.033204685896635056 |
| Loss at step 250: 0.03923625871539116 |
| Loss at step 300: 0.03889136016368866 |
| Loss at step 350: 0.03396601602435112 |
| Loss at step 400: 0.036183927208185196 |
| Loss at step 450: 0.03738848865032196 |
| Loss at step 500: 0.050185903906822205 |
| Loss at step 550: 0.05953901633620262 |
| Loss at step 600: 0.05247611179947853 |
| Loss at step 650: 0.04176788777112961 |
| Loss at step 700: 0.03954014182090759 |
| Loss at step 750: 0.051179856061935425 |
| Loss at step 800: 0.029204268008470535 |
| Loss at step 850: 0.03406751900911331 |
| Loss at step 900: 0.04037805274128914 |
| Mean training loss after epoch 125: 0.042219243697456714 |
|
|
| EPOCH: 126 |
| Loss at step 0: 0.05898353084921837 |
| Loss at step 50: 0.040512777864933014 |
| Loss at step 100: 0.0387885719537735 |
| Loss at step 150: 0.05783606320619583 |
| Loss at step 200: 0.0370011106133461 |
| Loss at step 250: 0.034667015075683594 |
| Loss at step 300: 0.07945484668016434 |
| Loss at step 350: 0.05463087558746338 |
| Loss at step 400: 0.05935561656951904 |
| Loss at step 450: 0.03419971093535423 |
| Loss at step 500: 0.0355793833732605 |
| Loss at step 550: 0.05645984038710594 |
| Loss at step 600: 0.049390655010938644 |
| Loss at step 650: 0.05601374804973602 |
| Loss at step 700: 0.03460899740457535 |
| Loss at step 750: 0.047203876078128815 |
| Loss at step 800: 0.043314892798662186 |
| Loss at step 850: 0.03529394418001175 |
| Loss at step 900: 0.03737284615635872 |
| Mean training loss after epoch 126: 0.0425018967548286 |
|
|
| EPOCH: 127 |
| Loss at step 0: 0.03094486892223358 |
| Loss at step 50: 0.033535122871398926 |
| Loss at step 100: 0.03193407505750656 |
| Loss at step 150: 0.0369366817176342 |
| Loss at step 200: 0.02800428494811058 |
| Loss at step 250: 0.061115555465221405 |
| Loss at step 300: 0.03858532756567001 |
| Loss at step 350: 0.053538210690021515 |
| Loss at step 400: 0.04869364574551582 |
| Loss at step 450: 0.055324409157037735 |
| Loss at step 500: 0.03902880474925041 |
| Loss at step 550: 0.030761603266000748 |
| Loss at step 600: 0.04287640005350113 |
| Loss at step 650: 0.03342721611261368 |
| Loss at step 700: 0.053739070892333984 |
| Loss at step 750: 0.03543848916888237 |
| Loss at step 800: 0.04534701257944107 |
| Loss at step 850: 0.04458793252706528 |
| Loss at step 900: 0.05748144909739494 |
| Mean training loss after epoch 127: 0.042397225750232935 |
|
|
| EPOCH: 128 |
| Loss at step 0: 0.04223093017935753 |
| Loss at step 50: 0.060702186077833176 |
| Loss at step 100: 0.05094020068645477 |
| Loss at step 150: 0.05862560123205185 |
| Loss at step 200: 0.05206675827503204 |
| Loss at step 250: 0.03660311549901962 |
| Loss at step 300: 0.033535201102495193 |
| Loss at step 350: 0.037460967898368835 |
| Loss at step 400: 0.053232479840517044 |
| Loss at step 450: 0.05418254807591438 |
| Loss at step 500: 0.03766205161809921 |
| Loss at step 550: 0.039794545620679855 |
| Loss at step 600: 0.04074326157569885 |
| Loss at step 650: 0.03707185387611389 |
| Loss at step 700: 0.04918089509010315 |
| Loss at step 750: 0.035559263080358505 |
| Loss at step 800: 0.038658492267131805 |
| Loss at step 850: 0.03465277701616287 |
| Loss at step 900: 0.055036671459674835 |
| Mean training loss after epoch 128: 0.04215331066812851 |
|
|
| EPOCH: 129 |
| Loss at step 0: 0.05584968626499176 |
| Loss at step 50: 0.044562358409166336 |
| Loss at step 100: 0.04858553409576416 |
| Loss at step 150: 0.035875383764505386 |
| Loss at step 200: 0.03444959968328476 |
| Loss at step 250: 0.049522798508405685 |
| Loss at step 300: 0.038845498114824295 |
| Loss at step 350: 0.0346490778028965 |
| Loss at step 400: 0.0353156253695488 |
| Loss at step 450: 0.04465166851878166 |
| Loss at step 500: 0.051781926304101944 |
| Loss at step 550: 0.04664802551269531 |
| Loss at step 600: 0.04371103271842003 |
| Loss at step 650: 0.03156234323978424 |
| Loss at step 700: 0.03441200777888298 |
| Loss at step 750: 0.04080637916922569 |
| Loss at step 800: 0.04525822773575783 |
| Loss at step 850: 0.038173094391822815 |
| Loss at step 900: 0.03883713483810425 |
| Mean training loss after epoch 129: 0.042144837785265975 |
|
|
| EPOCH: 130 |
| Loss at step 0: 0.03796138986945152 |
| Loss at step 50: 0.02935968153178692 |
| Loss at step 100: 0.0350450836122036 |
| Loss at step 150: 0.03702370822429657 |
| Loss at step 200: 0.04717714712023735 |
| Loss at step 250: 0.037151504307985306 |
| Loss at step 300: 0.030538195744156837 |
| Loss at step 350: 0.035160936415195465 |
| Loss at step 400: 0.056185707449913025 |
| Loss at step 450: 0.038169149309396744 |
| Loss at step 500: 0.03497578948736191 |
| Loss at step 550: 0.033725254237651825 |
| Loss at step 600: 0.03654317557811737 |
| Loss at step 650: 0.05485529825091362 |
| Loss at step 700: 0.035201311111450195 |
| Loss at step 750: 0.05112731456756592 |
| Loss at step 800: 0.0363975428044796 |
| Loss at step 850: 0.05157846212387085 |
| Loss at step 900: 0.053440894931554794 |
| Mean training loss after epoch 130: 0.04201112081135895 |
|
|
| EPOCH: 131 |
| Loss at step 0: 0.05593106895685196 |
| Loss at step 50: 0.04118822142481804 |
| Loss at step 100: 0.05003751814365387 |
| Loss at step 150: 0.0459161177277565 |
| Loss at step 200: 0.03542783111333847 |
| Loss at step 250: 0.03617458790540695 |
| Loss at step 300: 0.046611104160547256 |
| Loss at step 350: 0.05079322308301926 |
| Loss at step 400: 0.05026945844292641 |
| Loss at step 450: 0.03160965070128441 |
| Loss at step 500: 0.03939694166183472 |
| Loss at step 550: 0.042725078761577606 |
| Loss at step 600: 0.039472032338380814 |
| Loss at step 650: 0.043966952711343765 |
| Loss at step 700: 0.05131252482533455 |
| Loss at step 750: 0.05381901189684868 |
| Loss at step 800: 0.04540075361728668 |
| Loss at step 850: 0.03819124028086662 |
| Loss at step 900: 0.03499366715550423 |
| Mean training loss after epoch 131: 0.04209081147477698 |
|
|
| EPOCH: 132 |
| Loss at step 0: 0.03549477830529213 |
| Loss at step 50: 0.04868175461888313 |
| Loss at step 100: 0.06251970678567886 |
| Loss at step 150: 0.03984671086072922 |
| Loss at step 200: 0.03522738069295883 |
| Loss at step 250: 0.041131217032670975 |
| Loss at step 300: 0.03893343731760979 |
| Loss at step 350: 0.029867224395275116 |
| Loss at step 400: 0.03499307483434677 |
| Loss at step 450: 0.04396430775523186 |
| Loss at step 500: 0.039177216589450836 |
| Loss at step 550: 0.034836992621421814 |
| Loss at step 600: 0.0452614389359951 |
| Loss at step 650: 0.03610033541917801 |
| Loss at step 700: 0.03932995721697807 |
| Loss at step 750: 0.03963834419846535 |
| Loss at step 800: 0.03673923760652542 |
| Loss at step 850: 0.03914159908890724 |
| Loss at step 900: 0.0560661219060421 |
| Mean training loss after epoch 132: 0.042048414718351766 |
|
|
| EPOCH: 133 |
| Loss at step 0: 0.037345174700021744 |
| Loss at step 50: 0.035683225840330124 |
| Loss at step 100: 0.042083919048309326 |
| Loss at step 150: 0.036224111914634705 |
| Loss at step 200: 0.03736821562051773 |
| Loss at step 250: 0.04696378856897354 |
| Loss at step 300: 0.030614478513598442 |
| Loss at step 350: 0.04149696230888367 |
| Loss at step 400: 0.03438796103000641 |
| Loss at step 450: 0.029761843383312225 |
| Loss at step 500: 0.04799075052142143 |
| Loss at step 550: 0.03374233841896057 |
| Loss at step 600: 0.03222092241048813 |
| Loss at step 650: 0.03481902927160263 |
| Loss at step 700: 0.032734550535678864 |
| Loss at step 750: 0.03693768009543419 |
| Loss at step 800: 0.057545632123947144 |
| Loss at step 850: 0.03527749702334404 |
| Loss at step 900: 0.08574411273002625 |
| Mean training loss after epoch 133: 0.0421965813387368 |
|
|
| EPOCH: 134 |
| Loss at step 0: 0.033097926527261734 |
| Loss at step 50: 0.031262170523405075 |
| Loss at step 100: 0.03769355267286301 |
| Loss at step 150: 0.042281635105609894 |
| Loss at step 200: 0.04366563260555267 |
| Loss at step 250: 0.056263700127601624 |
| Loss at step 300: 0.0445694737136364 |
| Loss at step 350: 0.060232821851968765 |
| Loss at step 400: 0.043241508305072784 |
| Loss at step 450: 0.054095227271318436 |
| Loss at step 500: 0.03962637484073639 |
| Loss at step 550: 0.03208627551794052 |
| Loss at step 600: 0.036742426455020905 |
| Loss at step 650: 0.07225906848907471 |
| Loss at step 700: 0.031165501102805138 |
| Loss at step 750: 0.040376145392656326 |
| Loss at step 800: 0.031242134049534798 |
| Loss at step 850: 0.07587699592113495 |
| Loss at step 900: 0.03190525993704796 |
| Mean training loss after epoch 134: 0.04234294356829894 |
|
|
| EPOCH: 135 |
| Loss at step 0: 0.04674546420574188 |
| Loss at step 50: 0.03515247255563736 |
| Loss at step 100: 0.030674539506435394 |
| Loss at step 150: 0.03536396101117134 |
| Loss at step 200: 0.03699110075831413 |
| Loss at step 250: 0.05407152697443962 |
| Loss at step 300: 0.03668119013309479 |
| Loss at step 350: 0.05391205474734306 |
| Loss at step 400: 0.035997532308101654 |
| Loss at step 450: 0.0386599525809288 |
| Loss at step 500: 0.05625467747449875 |
| Loss at step 550: 0.040797557681798935 |
| Loss at step 600: 0.03620133921504021 |
| Loss at step 650: 0.041961271315813065 |
| Loss at step 700: 0.04411289095878601 |
| Loss at step 750: 0.04163749888539314 |
| Loss at step 800: 0.03395351395010948 |
| Loss at step 850: 0.035000745207071304 |
| Loss at step 900: 0.049789972603321075 |
| Mean training loss after epoch 135: 0.04251719271339206 |
|
|
| EPOCH: 136 |
| Loss at step 0: 0.03721001744270325 |
| Loss at step 50: 0.0380161888897419 |
| Loss at step 100: 0.03318065032362938 |
| Loss at step 150: 0.05025745928287506 |
| Loss at step 200: 0.02983713522553444 |
| Loss at step 250: 0.041543181985616684 |
| Loss at step 300: 0.03690667450428009 |
| Loss at step 350: 0.03452766686677933 |
| Loss at step 400: 0.06734222173690796 |
| Loss at step 450: 0.03532548248767853 |
| Loss at step 500: 0.0402277372777462 |
| Loss at step 550: 0.02923762984573841 |
| Loss at step 600: 0.03848644345998764 |
| Loss at step 650: 0.029214609414339066 |
| Loss at step 700: 0.03156350180506706 |
| Loss at step 750: 0.03646155819296837 |
| Loss at step 800: 0.030136052519083023 |
| Loss at step 850: 0.04044833034276962 |
| Loss at step 900: 0.03588998317718506 |
| Mean training loss after epoch 136: 0.042193242827299304 |
|
|
| EPOCH: 137 |
| Loss at step 0: 0.04032415151596069 |
| Loss at step 50: 0.04570217803120613 |
| Loss at step 100: 0.041807111352682114 |
| Loss at step 150: 0.03328615799546242 |
| Loss at step 200: 0.03072769194841385 |
| Loss at step 250: 0.033465202897787094 |
| Loss at step 300: 0.033506300300359726 |
| Loss at step 350: 0.03391355648636818 |
| Loss at step 400: 0.04394785314798355 |
| Loss at step 450: 0.05369193106889725 |
| Loss at step 500: 0.038760554045438766 |
| Loss at step 550: 0.052724115550518036 |
| Loss at step 600: 0.048280686140060425 |
| Loss at step 650: 0.05860856920480728 |
| Loss at step 700: 0.0367179699242115 |
| Loss at step 750: 0.037339795380830765 |
| Loss at step 800: 0.042971864342689514 |
| Loss at step 850: 0.03581308200955391 |
| Loss at step 900: 0.03864269703626633 |
| Mean training loss after epoch 137: 0.04252129275280275 |
|
|
| EPOCH: 138 |
| Loss at step 0: 0.03007676638662815 |
| Loss at step 50: 0.03923787921667099 |
| Loss at step 100: 0.03948834165930748 |
| Loss at step 150: 0.03789016231894493 |
| Loss at step 200: 0.046873949468135834 |
| Loss at step 250: 0.056466713547706604 |
| Loss at step 300: 0.03907657042145729 |
| Loss at step 350: 0.03223132714629173 |
| Loss at step 400: 0.04866752400994301 |
| Loss at step 450: 0.03630037605762482 |
| Loss at step 500: 0.029274195432662964 |
| Loss at step 550: 0.047705937176942825 |
| Loss at step 600: 0.05141275003552437 |
| Loss at step 650: 0.0336599238216877 |
| Loss at step 700: 0.03687726706266403 |
| Loss at step 750: 0.03609692305326462 |
| Loss at step 800: 0.03537045046687126 |
| Loss at step 850: 0.030984627082943916 |
| Loss at step 900: 0.032818686217069626 |
| Mean training loss after epoch 138: 0.04185507736051642 |
|
|
| EPOCH: 139 |
| Loss at step 0: 0.039597246795892715 |
| Loss at step 50: 0.03734810650348663 |
| Loss at step 100: 0.03580016270279884 |
| Loss at step 150: 0.03648393228650093 |
| Loss at step 200: 0.03841892629861832 |
| Loss at step 250: 0.03909728676080704 |
| Loss at step 300: 0.04612003266811371 |
| Loss at step 350: 0.0468924380838871 |
| Loss at step 400: 0.05363857373595238 |
| Loss at step 450: 0.05658219754695892 |
| Loss at step 500: 0.03320447728037834 |
| Loss at step 550: 0.054929088801145554 |
| Loss at step 600: 0.07501211762428284 |
| Loss at step 650: 0.041361112147569656 |
| Loss at step 700: 0.034978266805410385 |
| Loss at step 750: 0.0490257628262043 |
| Loss at step 800: 0.04843713343143463 |
| Loss at step 850: 0.052945345640182495 |
| Loss at step 900: 0.031348828226327896 |
| Mean training loss after epoch 139: 0.04195261146547571 |
|
|
| EPOCH: 140 |
| Loss at step 0: 0.038359805941581726 |
| Loss at step 50: 0.043496184051036835 |
| Loss at step 100: 0.03575534746050835 |
| Loss at step 150: 0.03715290129184723 |
| Loss at step 200: 0.05395453795790672 |
| Loss at step 250: 0.04312689229846001 |
| Loss at step 300: 0.039455071091651917 |
| Loss at step 350: 0.05315985530614853 |
| Loss at step 400: 0.04129849001765251 |
| Loss at step 450: 0.04127022251486778 |
| Loss at step 500: 0.04420750215649605 |
| Loss at step 550: 0.03587833046913147 |
| Loss at step 600: 0.038123976439237595 |
| Loss at step 650: 0.03822450339794159 |
| Loss at step 700: 0.040262363851070404 |
| Loss at step 750: 0.04293157905340195 |
| Loss at step 800: 0.03924191743135452 |
| Loss at step 850: 0.051242876797914505 |
| Loss at step 900: 0.030860869213938713 |
| Mean training loss after epoch 140: 0.041628152157849214 |
|
|
| EPOCH: 141 |
| Loss at step 0: 0.0357309952378273 |
| Loss at step 50: 0.0405811183154583 |
| Loss at step 100: 0.03416278213262558 |
| Loss at step 150: 0.03351762890815735 |
| Loss at step 200: 0.030533114448189735 |
| Loss at step 250: 0.03451740741729736 |
| Loss at step 300: 0.03665946051478386 |
| Loss at step 350: 0.05478139594197273 |
| Loss at step 400: 0.05036437511444092 |
| Loss at step 450: 0.05603531748056412 |
| Loss at step 500: 0.03402724489569664 |
| Loss at step 550: 0.03750381991267204 |
| Loss at step 600: 0.040972087532281876 |
| Loss at step 650: 0.042260900139808655 |
| Loss at step 700: 0.03811081871390343 |
| Loss at step 750: 0.036845553666353226 |
| Loss at step 800: 0.0415690578520298 |
| Loss at step 850: 0.03965042531490326 |
| Loss at step 900: 0.03870100900530815 |
| Mean training loss after epoch 141: 0.04202673006763082 |
|
|
| EPOCH: 142 |
| Loss at step 0: 0.03593452647328377 |
| Loss at step 50: 0.0351838618516922 |
| Loss at step 100: 0.06876881420612335 |
| Loss at step 150: 0.038944195955991745 |
| Loss at step 200: 0.040944647043943405 |
| Loss at step 250: 0.039801549166440964 |
| Loss at step 300: 0.03724919632077217 |
| Loss at step 350: 0.03012264519929886 |
| Loss at step 400: 0.028339020907878876 |
| Loss at step 450: 0.043429140001535416 |
| Loss at step 500: 0.039767246693372726 |
| Loss at step 550: 0.034304797649383545 |
| Loss at step 600: 0.0415533110499382 |
| Loss at step 650: 0.04240945726633072 |
| Loss at step 700: 0.03782442957162857 |
| Loss at step 750: 0.038175780326128006 |
| Loss at step 800: 0.0390368290245533 |
| Loss at step 850: 0.04891939461231232 |
| Loss at step 900: 0.04027196764945984 |
| Mean training loss after epoch 142: 0.042164087712542334 |
|
|
| EPOCH: 143 |
| Loss at step 0: 0.05464017391204834 |
| Loss at step 50: 0.04778584465384483 |
| Loss at step 100: 0.043277475982904434 |
| Loss at step 150: 0.030136309564113617 |
| Loss at step 200: 0.038401342928409576 |
| Loss at step 250: 0.03655426949262619 |
| Loss at step 300: 0.03619815781712532 |
| Loss at step 350: 0.03555699810385704 |
| Loss at step 400: 0.06355958431959152 |
| Loss at step 450: 0.03683250769972801 |
| Loss at step 500: 0.031722113490104675 |
| Loss at step 550: 0.051293518394231796 |
| Loss at step 600: 0.04472873732447624 |
| Loss at step 650: 0.04285021498799324 |
| Loss at step 700: 0.03676331043243408 |
| Loss at step 750: 0.053878091275691986 |
| Loss at step 800: 0.039428550750017166 |
| Loss at step 850: 0.029044492170214653 |
| Loss at step 900: 0.046056412160396576 |
| Mean training loss after epoch 143: 0.04196451393875486 |
|
|
| EPOCH: 144 |
| Loss at step 0: 0.03647203743457794 |
| Loss at step 50: 0.03546447679400444 |
| Loss at step 100: 0.028027458116412163 |
| Loss at step 150: 0.03706343099474907 |
| Loss at step 200: 0.02946298196911812 |
| Loss at step 250: 0.04093487188220024 |
| Loss at step 300: 0.03480026498436928 |
| Loss at step 350: 0.0481966957449913 |
| Loss at step 400: 0.04102826490998268 |
| Loss at step 450: 0.035431042313575745 |
| Loss at step 500: 0.05099305883049965 |
| Loss at step 550: 0.042340170592069626 |
| Loss at step 600: 0.05557699128985405 |
| Loss at step 650: 0.03197412192821503 |
| Loss at step 700: 0.035774651914834976 |
| Loss at step 750: 0.03557714819908142 |
| Loss at step 800: 0.042168546468019485 |
| Loss at step 850: 0.02944686822593212 |
| Loss at step 900: 0.034017935395240784 |
| Mean training loss after epoch 144: 0.04199131450323916 |
|
|
| EPOCH: 145 |
| Loss at step 0: 0.0357891246676445 |
| Loss at step 50: 0.05408082529902458 |
| Loss at step 100: 0.03127046301960945 |
| Loss at step 150: 0.035092175006866455 |
| Loss at step 200: 0.04589945450425148 |
| Loss at step 250: 0.03216953203082085 |
| Loss at step 300: 0.04434205964207649 |
| Loss at step 350: 0.06629910320043564 |
| Loss at step 400: 0.04920700937509537 |
| Loss at step 450: 0.042137082666158676 |
| Loss at step 500: 0.031183796003460884 |
| Loss at step 550: 0.03542826697230339 |
| Loss at step 600: 0.04166002199053764 |
| Loss at step 650: 0.037325046956539154 |
| Loss at step 700: 0.04343381151556969 |
| Loss at step 750: 0.03837285563349724 |
| Loss at step 800: 0.0291228536516428 |
| Loss at step 850: 0.03319094330072403 |
| Loss at step 900: 0.036391112953424454 |
| Mean training loss after epoch 145: 0.04157962997569077 |
|
|
| EPOCH: 146 |
| Loss at step 0: 0.03486700356006622 |
| Loss at step 50: 0.04799339547753334 |
| Loss at step 100: 0.053472988307476044 |
| Loss at step 150: 0.049768686294555664 |
| Loss at step 200: 0.04211004450917244 |
| Loss at step 250: 0.08333371579647064 |
| Loss at step 300: 0.03735140338540077 |
| Loss at step 350: 0.0437965951859951 |
| Loss at step 400: 0.04522951692342758 |
| Loss at step 450: 0.050311364233493805 |
| Loss at step 500: 0.05300811678171158 |
| Loss at step 550: 0.037035051733255386 |
| Loss at step 600: 0.03540425002574921 |
| Loss at step 650: 0.03272141516208649 |
| Loss at step 700: 0.04344155266880989 |
| Loss at step 750: 0.04171430692076683 |
| Loss at step 800: 0.03223220631480217 |
| Loss at step 850: 0.04631401598453522 |
| Loss at step 900: 0.034416262060403824 |
| Mean training loss after epoch 146: 0.04228796660383818 |
|
|
| EPOCH: 147 |
| Loss at step 0: 0.04061895236372948 |
| Loss at step 50: 0.06510394811630249 |
| Loss at step 100: 0.034833405166864395 |
| Loss at step 150: 0.04406037554144859 |
| Loss at step 200: 0.045396748930215836 |
| Loss at step 250: 0.04945330694317818 |
| Loss at step 300: 0.0697096735239029 |
| Loss at step 350: 0.040425848215818405 |
| Loss at step 400: 0.03286696970462799 |
| Loss at step 450: 0.030878452584147453 |
| Loss at step 500: 0.052155978977680206 |
| Loss at step 550: 0.050235211849212646 |
| Loss at step 600: 0.03293605148792267 |
| Loss at step 650: 0.05016927048563957 |
| Loss at step 700: 0.03877097740769386 |
| Loss at step 750: 0.051481712609529495 |
| Loss at step 800: 0.0331539511680603 |
| Loss at step 850: 0.0463896170258522 |
| Loss at step 900: 0.050032272934913635 |
| Mean training loss after epoch 147: 0.041787002301181174 |
|
|
| EPOCH: 148 |
| Loss at step 0: 0.0357489250600338 |
| Loss at step 50: 0.05856280401349068 |
| Loss at step 100: 0.0320919044315815 |
| Loss at step 150: 0.0821550190448761 |
| Loss at step 200: 0.04068780690431595 |
| Loss at step 250: 0.03332633152604103 |
| Loss at step 300: 0.0528477281332016 |
| Loss at step 350: 0.042828939855098724 |
| Loss at step 400: 0.04491664841771126 |
| Loss at step 450: 0.04121486842632294 |
| Loss at step 500: 0.03646986186504364 |
| Loss at step 550: 0.040154021233320236 |
| Loss at step 600: 0.03764384612441063 |
| Loss at step 650: 0.05622980371117592 |
| Loss at step 700: 0.0694388672709465 |
| Loss at step 750: 0.026673797518014908 |
| Loss at step 800: 0.033385686576366425 |
| Loss at step 850: 0.051487911492586136 |
| Loss at step 900: 0.03869228437542915 |
| Mean training loss after epoch 148: 0.0420387861078609 |
|
|
| EPOCH: 149 |
| Loss at step 0: 0.032291069626808167 |
| Loss at step 50: 0.04381278157234192 |
| Loss at step 100: 0.05175172910094261 |
| Loss at step 150: 0.039617761969566345 |
| Loss at step 200: 0.04136640951037407 |
| Loss at step 250: 0.06429476290941238 |
| Loss at step 300: 0.06880918890237808 |
| Loss at step 350: 0.0365937314927578 |
| Loss at step 400: 0.04158145561814308 |
| Loss at step 450: 0.04357963427901268 |
| Loss at step 500: 0.03778946399688721 |
| Loss at step 550: 0.02897937409579754 |
| Loss at step 600: 0.03430233523249626 |
| Loss at step 650: 0.04783837869763374 |
| Loss at step 700: 0.033608194440603256 |
| Loss at step 750: 0.04544506222009659 |
| Loss at step 800: 0.048203710466623306 |
| Loss at step 850: 0.04275382310152054 |
| Loss at step 900: 0.041548918932676315 |
| Mean training loss after epoch 149: 0.042700570968311355 |
|
|
| EPOCH: 150 |
| Loss at step 0: 0.049964435398578644 |
| Loss at step 50: 0.048390697687864304 |
| Loss at step 100: 0.05685277283191681 |
| Loss at step 150: 0.036489762365818024 |
| Loss at step 200: 0.05473584309220314 |
| Loss at step 250: 0.03579915314912796 |
| Loss at step 300: 0.03483206406235695 |
| Loss at step 350: 0.049429502338171005 |
| Loss at step 400: 0.050645049661397934 |
| Loss at step 450: 0.035665128380060196 |
| Loss at step 500: 0.055477336049079895 |
| Loss at step 550: 0.036538902670145035 |
| Loss at step 600: 0.034477416425943375 |
| Loss at step 650: 0.029415998607873917 |
| Loss at step 700: 0.036369092762470245 |
| Loss at step 750: 0.03708411753177643 |
| Loss at step 800: 0.04495839774608612 |
| Loss at step 850: 0.03702973574399948 |
| Loss at step 900: 0.03885873034596443 |
| Mean training loss after epoch 150: 0.041629522208815446 |
|
|
| EPOCH: 151 |
| Loss at step 0: 0.03872460499405861 |
| Loss at step 50: 0.035674791783094406 |
| Loss at step 100: 0.03539711609482765 |
| Loss at step 150: 0.035271402448415756 |
| Loss at step 200: 0.038769178092479706 |
| Loss at step 250: 0.05824952945113182 |
| Loss at step 300: 0.035957783460617065 |
| Loss at step 350: 0.03432445973157883 |
| Loss at step 400: 0.05303642526268959 |
| Loss at step 450: 0.030788464471697807 |
| Loss at step 500: 0.03501858562231064 |
| Loss at step 550: 0.04111636430025101 |
| Loss at step 600: 0.032229967415332794 |
| Loss at step 650: 0.03571541979908943 |
| Loss at step 700: 0.03328194469213486 |
| Loss at step 750: 0.03248866647481918 |
| Loss at step 800: 0.03424318879842758 |
| Loss at step 850: 0.039510175585746765 |
| Loss at step 900: 0.05264480784535408 |
| Mean training loss after epoch 151: 0.041698548992448396 |
|
|
| EPOCH: 152 |
| Loss at step 0: 0.043211840093135834 |
| Loss at step 50: 0.03340320661664009 |
| Loss at step 100: 0.048411767929792404 |
| Loss at step 150: 0.056272272020578384 |
| Loss at step 200: 0.051562827080488205 |
| Loss at step 250: 0.034891560673713684 |
| Loss at step 300: 0.05148777738213539 |
| Loss at step 350: 0.03595224395394325 |
| Loss at step 400: 0.028903206810355186 |
| Loss at step 450: 0.038183365017175674 |
| Loss at step 500: 0.03731566295027733 |
| Loss at step 550: 0.0356873981654644 |
| Loss at step 600: 0.0313677042722702 |
| Loss at step 650: 0.03932942450046539 |
| Loss at step 700: 0.0665639266371727 |
| Loss at step 750: 0.03317286819219589 |
| Loss at step 800: 0.03564203903079033 |
| Loss at step 850: 0.03477342799305916 |
| Loss at step 900: 0.04680752009153366 |
| Mean training loss after epoch 152: 0.041610624573664115 |
|
|
| EPOCH: 153 |
| Loss at step 0: 0.030752496793866158 |
| Loss at step 50: 0.03984297439455986 |
| Loss at step 100: 0.031056780368089676 |
| Loss at step 150: 0.03939105570316315 |
| Loss at step 200: 0.06974080204963684 |
| Loss at step 250: 0.03848470747470856 |
| Loss at step 300: 0.03297055512666702 |
| Loss at step 350: 0.04547300189733505 |
| Loss at step 400: 0.03682214766740799 |
| Loss at step 450: 0.03131253272294998 |
| Loss at step 500: 0.04345107078552246 |
| Loss at step 550: 0.04177214205265045 |
| Loss at step 600: 0.030688272789120674 |
| Loss at step 650: 0.048584792762994766 |
| Loss at step 700: 0.038229744881391525 |
| Loss at step 750: 0.031089046970009804 |
| Loss at step 800: 0.03985687717795372 |
| Loss at step 850: 0.041947998106479645 |
| Loss at step 900: 0.044597942382097244 |
| Mean training loss after epoch 153: 0.042380063657535674 |
|
|
| EPOCH: 154 |
| Loss at step 0: 0.05345641076564789 |
| Loss at step 50: 0.042899783700704575 |
| Loss at step 100: 0.03357723355293274 |
| Loss at step 150: 0.05140424147248268 |
| Loss at step 200: 0.02819659747183323 |
| Loss at step 250: 0.04493157938122749 |
| Loss at step 300: 0.03656427562236786 |
| Loss at step 350: 0.030955562368035316 |
| Loss at step 400: 0.0369301363825798 |
| Loss at step 450: 0.040071986615657806 |
| Loss at step 500: 0.05197841301560402 |
| Loss at step 550: 0.03778393194079399 |
| Loss at step 600: 0.03276905044913292 |
| Loss at step 650: 0.03182387724518776 |
| Loss at step 700: 0.03351379930973053 |
| Loss at step 750: 0.033173974603414536 |
| Loss at step 800: 0.0386836901307106 |
| Loss at step 850: 0.03227386251091957 |
| Loss at step 900: 0.035299576818943024 |
| Mean training loss after epoch 154: 0.04212207293539032 |
|
|
| EPOCH: 155 |
| Loss at step 0: 0.05649520829319954 |
| Loss at step 50: 0.03102577105164528 |
| Loss at step 100: 0.032175131142139435 |
| Loss at step 150: 0.0337371900677681 |
| Loss at step 200: 0.03588712960481644 |
| Loss at step 250: 0.047558512538671494 |
| Loss at step 300: 0.05539235100150108 |
| Loss at step 350: 0.0677049309015274 |
| Loss at step 400: 0.04687006399035454 |
| Loss at step 450: 0.042771343141794205 |
| Loss at step 500: 0.03482664376497269 |
| Loss at step 550: 0.044766876846551895 |
| Loss at step 600: 0.045993443578481674 |
| Loss at step 650: 0.041110921651124954 |
| Loss at step 700: 0.03490821272134781 |
| Loss at step 750: 0.04456814005970955 |
| Loss at step 800: 0.05052883177995682 |
| Loss at step 850: 0.06528449058532715 |
| Loss at step 900: 0.036868300288915634 |
| Mean training loss after epoch 155: 0.04227992473269449 |
|
|
| EPOCH: 156 |
| Loss at step 0: 0.05328263342380524 |
| Loss at step 50: 0.05558167025446892 |
| Loss at step 100: 0.035837870091199875 |
| Loss at step 150: 0.03872068226337433 |
| Loss at step 200: 0.062104836106300354 |
| Loss at step 250: 0.038871198892593384 |
| Loss at step 300: 0.035116370767354965 |
| Loss at step 350: 0.0337044931948185 |
| Loss at step 400: 0.035277001559734344 |
| Loss at step 450: 0.05462329462170601 |
| Loss at step 500: 0.03315630555152893 |
| Loss at step 550: 0.03951284661889076 |
| Loss at step 600: 0.04857729747891426 |
| Loss at step 650: 0.042670510709285736 |
| Loss at step 700: 0.030172094702720642 |
| Loss at step 750: 0.035838935524225235 |
| Loss at step 800: 0.05214604362845421 |
| Loss at step 850: 0.041880179196596146 |
| Loss at step 900: 0.03618021681904793 |
| Mean training loss after epoch 156: 0.04151801192469752 |
|
|
| EPOCH: 157 |
| Loss at step 0: 0.053754597902297974 |
| Loss at step 50: 0.04583209753036499 |
| Loss at step 100: 0.048923052847385406 |
| Loss at step 150: 0.037170544266700745 |
| Loss at step 200: 0.05567541718482971 |
| Loss at step 250: 0.04202282056212425 |
| Loss at step 300: 0.0530812032520771 |
| Loss at step 350: 0.036701470613479614 |
| Loss at step 400: 0.04105161875486374 |
| Loss at step 450: 0.035315290093421936 |
| Loss at step 500: 0.03592460975050926 |
| Loss at step 550: 0.05100354924798012 |
| Loss at step 600: 0.03835726156830788 |
| Loss at step 650: 0.040117476135492325 |
| Loss at step 700: 0.03319944441318512 |
| Loss at step 750: 0.04874331131577492 |
| Loss at step 800: 0.03211164474487305 |
| Loss at step 850: 0.03677653148770332 |
| Loss at step 900: 0.03921843320131302 |
| Mean training loss after epoch 157: 0.04208396263778019 |
|
|
| EPOCH: 158 |
| Loss at step 0: 0.053142696619033813 |
| Loss at step 50: 0.039625342935323715 |
| Loss at step 100: 0.041874464601278305 |
| Loss at step 150: 0.04172403737902641 |
| Loss at step 200: 0.03345603868365288 |
| Loss at step 250: 0.03492490202188492 |
| Loss at step 300: 0.038030609488487244 |
| Loss at step 350: 0.028495457023382187 |
| Loss at step 400: 0.04284537583589554 |
| Loss at step 450: 0.034388937056064606 |
| Loss at step 500: 0.055605459958314896 |
| Loss at step 550: 0.04129524528980255 |
| Loss at step 600: 0.03616258502006531 |
| Loss at step 650: 0.0285470113158226 |
| Loss at step 700: 0.0391588918864727 |
| Loss at step 750: 0.0373045951128006 |
| Loss at step 800: 0.036178652197122574 |
| Loss at step 850: 0.04029553756117821 |
| Loss at step 900: 0.030573036521673203 |
| Mean training loss after epoch 158: 0.041759737425728014 |
|
|
| EPOCH: 159 |
| Loss at step 0: 0.03646078705787659 |
| Loss at step 50: 0.03891812264919281 |
| Loss at step 100: 0.04230709373950958 |
| Loss at step 150: 0.038509808480739594 |
| Loss at step 200: 0.053831759840250015 |
| Loss at step 250: 0.045457784086465836 |
| Loss at step 300: 0.035048820078372955 |
| Loss at step 350: 0.0402105338871479 |
| Loss at step 400: 0.040246471762657166 |
| Loss at step 450: 0.03213101997971535 |
| Loss at step 500: 0.039840925484895706 |
| Loss at step 550: 0.034487396478652954 |
| Loss at step 600: 0.05263527110219002 |
| Loss at step 650: 0.03569900244474411 |
| Loss at step 700: 0.03732169046998024 |
| Loss at step 750: 0.03432917222380638 |
| Loss at step 800: 0.029399283230304718 |
| Loss at step 850: 0.03557441011071205 |
| Loss at step 900: 0.05276878550648689 |
| Mean training loss after epoch 159: 0.04198013888056408 |
|
|
| EPOCH: 160 |
| Loss at step 0: 0.03272593766450882 |
| Loss at step 50: 0.03808004409074783 |
| Loss at step 100: 0.04068906977772713 |
| Loss at step 150: 0.03765731677412987 |
| Loss at step 200: 0.044079214334487915 |
| Loss at step 250: 0.03938356041908264 |
| Loss at step 300: 0.05161362513899803 |
| Loss at step 350: 0.04128038510680199 |
| Loss at step 400: 0.05006210878491402 |
| Loss at step 450: 0.03323276713490486 |
| Loss at step 500: 0.04292380064725876 |
| Loss at step 550: 0.03615903854370117 |
| Loss at step 600: 0.04243318736553192 |
| Loss at step 650: 0.03709695115685463 |
| Loss at step 700: 0.05068572610616684 |
| Loss at step 750: 0.03722544386982918 |
| Loss at step 800: 0.056304723024368286 |
| Loss at step 850: 0.05459215119481087 |
| Loss at step 900: 0.035317618399858475 |
| Mean training loss after epoch 160: 0.04215756298970185 |
|
|
| EPOCH: 161 |
| Loss at step 0: 0.042033180594444275 |
| Loss at step 50: 0.04185538738965988 |
| Loss at step 100: 0.03522934764623642 |
| Loss at step 150: 0.05067601427435875 |
| Loss at step 200: 0.03864051774144173 |
| Loss at step 250: 0.04868760704994202 |
| Loss at step 300: 0.03932592645287514 |
| Loss at step 350: 0.04293591156601906 |
| Loss at step 400: 0.0446302592754364 |
| Loss at step 450: 0.03664654493331909 |
| Loss at step 500: 0.05163183808326721 |
| Loss at step 550: 0.044752635061740875 |
| Loss at step 600: 0.03847457468509674 |
| Loss at step 650: 0.05418030917644501 |
| Loss at step 700: 0.04585876315832138 |
| Loss at step 750: 0.03625411167740822 |
| Loss at step 800: 0.035573311150074005 |
| Loss at step 850: 0.04612034559249878 |
| Loss at step 900: 0.03266897052526474 |
| Mean training loss after epoch 161: 0.041682401608461254 |
|
|
| EPOCH: 162 |
| Loss at step 0: 0.03443225100636482 |
| Loss at step 50: 0.05295828357338905 |
| Loss at step 100: 0.044075947254896164 |
| Loss at step 150: 0.039766378700733185 |
| Loss at step 200: 0.03175616264343262 |
| Loss at step 250: 0.03700171783566475 |
| Loss at step 300: 0.06151432916522026 |
| Loss at step 350: 0.03539503365755081 |
| Loss at step 400: 0.035971418023109436 |
| Loss at step 450: 0.06043902039527893 |
| Loss at step 500: 0.026486312970519066 |
| Loss at step 550: 0.03172525390982628 |
| Loss at step 600: 0.05635107308626175 |
| Loss at step 650: 0.049782294780015945 |
| Loss at step 700: 0.03821375221014023 |
| Loss at step 750: 0.03226267918944359 |
| Loss at step 800: 0.04801420122385025 |
| Loss at step 850: 0.037035200744867325 |
| Loss at step 900: 0.0541158989071846 |
| Mean training loss after epoch 162: 0.041881437564312396 |
|
|
| EPOCH: 163 |
| Loss at step 0: 0.046113207936286926 |
| Loss at step 50: 0.0443924181163311 |
| Loss at step 100: 0.03236755356192589 |
| Loss at step 150: 0.03812683746218681 |
| Loss at step 200: 0.03635641187429428 |
| Loss at step 250: 0.062125999480485916 |
| Loss at step 300: 0.05177285149693489 |
| Loss at step 350: 0.04134644195437431 |
| Loss at step 400: 0.051762331277132034 |
| Loss at step 450: 0.038076095283031464 |
| Loss at step 500: 0.048686202615499496 |
| Loss at step 550: 0.04135816916823387 |
| Loss at step 600: 0.04114864766597748 |
| Loss at step 650: 0.036704305559396744 |
| Loss at step 700: 0.049851950258016586 |
| Loss at step 750: 0.03393217548727989 |
| Loss at step 800: 0.03302428126335144 |
| Loss at step 850: 0.036545637995004654 |
| Loss at step 900: 0.042946405708789825 |
| Mean training loss after epoch 163: 0.04166837653188881 |
|
|
| EPOCH: 164 |
| Loss at step 0: 0.05492968484759331 |
| Loss at step 50: 0.03143102303147316 |
| Loss at step 100: 0.03686508163809776 |
| Loss at step 150: 0.03780407831072807 |
| Loss at step 200: 0.03192293271422386 |
| Loss at step 250: 0.06838148087263107 |
| Loss at step 300: 0.05755019187927246 |
| Loss at step 350: 0.050360266119241714 |
| Loss at step 400: 0.036883942782878876 |
| Loss at step 450: 0.05594587326049805 |
| Loss at step 500: 0.0486307218670845 |
| Loss at step 550: 0.04816119745373726 |
| Loss at step 600: 0.05513868108391762 |
| Loss at step 650: 0.040407393127679825 |
| Loss at step 700: 0.035788264125585556 |
| Loss at step 750: 0.029059693217277527 |
| Loss at step 800: 0.03355870768427849 |
| Loss at step 850: 0.06275414675474167 |
| Loss at step 900: 0.04140639677643776 |
| Mean training loss after epoch 164: 0.0419985556732744 |
|
|
| EPOCH: 165 |
| Loss at step 0: 0.05084880441427231 |
| Loss at step 50: 0.0543023943901062 |
| Loss at step 100: 0.056270111352205276 |
| Loss at step 150: 0.034627776592969894 |
| Loss at step 200: 0.029606739059090614 |
| Loss at step 250: 0.04984544217586517 |
| Loss at step 300: 0.033280953764915466 |
| Loss at step 350: 0.04561571031808853 |
| Loss at step 400: 0.034940462559461594 |
| Loss at step 450: 0.04139413684606552 |
| Loss at step 500: 0.040452610701322556 |
| Loss at step 550: 0.04350801184773445 |
| Loss at step 600: 0.03656912222504616 |
| Loss at step 650: 0.04006418213248253 |
| Loss at step 700: 0.04635033383965492 |
| Loss at step 750: 0.03196395933628082 |
| Loss at step 800: 0.03807682916522026 |
| Loss at step 850: 0.030541956424713135 |
| Loss at step 900: 0.05490187183022499 |
| Mean training loss after epoch 165: 0.04234211670477062 |
|
|
| EPOCH: 166 |
| Loss at step 0: 0.0435408279299736 |
| Loss at step 50: 0.04618099704384804 |
| Loss at step 100: 0.05463716760277748 |
| Loss at step 150: 0.05916711688041687 |
| Loss at step 200: 0.05074099823832512 |
| Loss at step 250: 0.03297647461295128 |
| Loss at step 300: 0.03944069892168045 |
| Loss at step 350: 0.06627862900495529 |
| Loss at step 400: 0.03429859131574631 |
| Loss at step 450: 0.039032090455293655 |
| Loss at step 500: 0.0400698184967041 |
| Loss at step 550: 0.0376395508646965 |
| Loss at step 600: 0.03616385906934738 |
| Loss at step 650: 0.05470064654946327 |
| Loss at step 700: 0.04045157879590988 |
| Loss at step 750: 0.07541951537132263 |
| Loss at step 800: 0.05270006135106087 |
| Loss at step 850: 0.041193705052137375 |
| Loss at step 900: 0.03363768011331558 |
| Mean training loss after epoch 166: 0.041681498600872974 |
|
|
| EPOCH: 167 |
| Loss at step 0: 0.03260944038629532 |
| Loss at step 50: 0.031192278489470482 |
| Loss at step 100: 0.030275052413344383 |
| Loss at step 150: 0.0319354273378849 |
| Loss at step 200: 0.038708776235580444 |
| Loss at step 250: 0.05172756686806679 |
| Loss at step 300: 0.0345488116145134 |
| Loss at step 350: 0.04113924130797386 |
| Loss at step 400: 0.035771701484918594 |
| Loss at step 450: 0.03828283026814461 |
| Loss at step 500: 0.0383487194776535 |
| Loss at step 550: 0.06837842613458633 |
| Loss at step 600: 0.03420782834291458 |
| Loss at step 650: 0.03368475288152695 |
| Loss at step 700: 0.036766745150089264 |
| Loss at step 750: 0.03165431320667267 |
| Loss at step 800: 0.04517103731632233 |
| Loss at step 850: 0.051145922392606735 |
| Loss at step 900: 0.03939900919795036 |
| Mean training loss after epoch 167: 0.04159986691227727 |
|
|
| EPOCH: 168 |
| Loss at step 0: 0.04851570725440979 |
| Loss at step 50: 0.06281198561191559 |
| Loss at step 100: 0.04415833204984665 |
| Loss at step 150: 0.03673063963651657 |
| Loss at step 200: 0.03636414185166359 |
| Loss at step 250: 0.03222019225358963 |
| Loss at step 300: 0.0397556908428669 |
| Loss at step 350: 0.03364252671599388 |
| Loss at step 400: 0.03710572049021721 |
| Loss at step 450: 0.02969525195658207 |
| Loss at step 500: 0.08423370122909546 |
| Loss at step 550: 0.0415620356798172 |
| Loss at step 600: 0.027658242732286453 |
| Loss at step 650: 0.042391855269670486 |
| Loss at step 700: 0.03817259520292282 |
| Loss at step 750: 0.03259177878499031 |
| Loss at step 800: 0.03972531110048294 |
| Loss at step 850: 0.0303537305444479 |
| Loss at step 900: 0.03652387112379074 |
| Mean training loss after epoch 168: 0.04140357141579583 |
|
|
| EPOCH: 169 |
| Loss at step 0: 0.05354360118508339 |
| Loss at step 50: 0.03195581212639809 |
| Loss at step 100: 0.06500847637653351 |
| Loss at step 150: 0.06610570102930069 |
| Loss at step 200: 0.04113024100661278 |
| Loss at step 250: 0.027583038434386253 |
| Loss at step 300: 0.03656454011797905 |
| Loss at step 350: 0.052170515060424805 |
| Loss at step 400: 0.03852315619587898 |
| Loss at step 450: 0.03867664188146591 |
| Loss at step 500: 0.04367469251155853 |
| Loss at step 550: 0.03310701623558998 |
| Loss at step 600: 0.06611268222332001 |
| Loss at step 650: 0.03098241798579693 |
| Loss at step 700: 0.04891456663608551 |
| Loss at step 750: 0.03942761942744255 |
| Loss at step 800: 0.041297849267721176 |
| Loss at step 850: 0.052537478506565094 |
| Loss at step 900: 0.0340694934129715 |
| Mean training loss after epoch 169: 0.04182671864967801 |
|
|
| EPOCH: 170 |
| Loss at step 0: 0.0305585078895092 |
| Loss at step 50: 0.044676389545202255 |
| Loss at step 100: 0.03262167051434517 |
| Loss at step 150: 0.03465871140360832 |
| Loss at step 200: 0.053631119430065155 |
| Loss at step 250: 0.04847809672355652 |
| Loss at step 300: 0.03438197821378708 |
| Loss at step 350: 0.03563510254025459 |
| Loss at step 400: 0.041701238602399826 |
| Loss at step 450: 0.03720324486494064 |
| Loss at step 500: 0.03245695307850838 |
| Loss at step 550: 0.050643812865018845 |
| Loss at step 600: 0.0421716682612896 |
| Loss at step 650: 0.031076569110155106 |
| Loss at step 700: 0.03496023640036583 |
| Loss at step 750: 0.041589297354221344 |
| Loss at step 800: 0.0360647588968277 |
| Loss at step 850: 0.03295659273862839 |
| Loss at step 900: 0.03453189507126808 |
| Mean training loss after epoch 170: 0.04185354243169652 |
|
|
| EPOCH: 171 |
| Loss at step 0: 0.053923431783914566 |
| Loss at step 50: 0.05165528133511543 |
| Loss at step 100: 0.05359581485390663 |
| Loss at step 150: 0.03814880549907684 |
| Loss at step 200: 0.057905055582523346 |
| Loss at step 250: 0.05054771900177002 |
| Loss at step 300: 0.029955018311738968 |
| Loss at step 350: 0.04426371306180954 |
| Loss at step 400: 0.037701405584812164 |
| Loss at step 450: 0.04641666263341904 |
| Loss at step 500: 0.05583236366510391 |
| Loss at step 550: 0.03619731217622757 |
| Loss at step 600: 0.03259444236755371 |
| Loss at step 650: 0.060233090072870255 |
| Loss at step 700: 0.04138468950986862 |
| Loss at step 750: 0.03163965418934822 |
| Loss at step 800: 0.032825879752635956 |
| Loss at step 850: 0.05752290412783623 |
| Loss at step 900: 0.03443465754389763 |
| Mean training loss after epoch 171: 0.04144824135389282 |
|
|
| EPOCH: 172 |
| Loss at step 0: 0.03209567442536354 |
| Loss at step 50: 0.05818139761686325 |
| Loss at step 100: 0.05136827379465103 |
| Loss at step 150: 0.027739521116018295 |
| Loss at step 200: 0.042954180389642715 |
| Loss at step 250: 0.030690737068653107 |
| Loss at step 300: 0.03409494832158089 |
| Loss at step 350: 0.027100542560219765 |
| Loss at step 400: 0.03840995579957962 |
| Loss at step 450: 0.03903039172291756 |
| Loss at step 500: 0.04468401148915291 |
| Loss at step 550: 0.03910224139690399 |
| Loss at step 600: 0.0376056544482708 |
| Loss at step 650: 0.035212963819503784 |
| Loss at step 700: 0.03200206905603409 |
| Loss at step 750: 0.035134561359882355 |
| Loss at step 800: 0.03554246947169304 |
| Loss at step 850: 0.05527525022625923 |
| Loss at step 900: 0.03256995603442192 |
| Mean training loss after epoch 172: 0.04208877019441204 |
|
|
| EPOCH: 173 |
| Loss at step 0: 0.035287100821733475 |
| Loss at step 50: 0.039894670248031616 |
| Loss at step 100: 0.03790206089615822 |
| Loss at step 150: 0.03075706586241722 |
| Loss at step 200: 0.05236460268497467 |
| Loss at step 250: 0.04161772504448891 |
| Loss at step 300: 0.033940721303224564 |
| Loss at step 350: 0.06850095838308334 |
| Loss at step 400: 0.03070612996816635 |
| Loss at step 450: 0.045446161180734634 |
| Loss at step 500: 0.032258786261081696 |
| Loss at step 550: 0.03494522348046303 |
| Loss at step 600: 0.045330144464969635 |
| Loss at step 650: 0.036904022097587585 |
| Loss at step 700: 0.032146237790584564 |
| Loss at step 750: 0.034715086221694946 |
| Loss at step 800: 0.03468701243400574 |
| Loss at step 850: 0.033052023500204086 |
| Loss at step 900: 0.034720417112112045 |
| Mean training loss after epoch 173: 0.042048089476282406 |
|
|
| EPOCH: 174 |
| Loss at step 0: 0.03656395152211189 |
| Loss at step 50: 0.04568571224808693 |
| Loss at step 100: 0.03481351211667061 |
| Loss at step 150: 0.038224395364522934 |
| Loss at step 200: 0.058359503746032715 |
| Loss at step 250: 0.03833071514964104 |
| Loss at step 300: 0.043960943818092346 |
| Loss at step 350: 0.04375429451465607 |
| Loss at step 400: 0.03465664014220238 |
| Loss at step 450: 0.029857201501727104 |
| Loss at step 500: 0.044843051582574844 |
| Loss at step 550: 0.05128093063831329 |
| Loss at step 600: 0.06645429879426956 |
| Loss at step 650: 0.03523625433444977 |
| Loss at step 700: 0.03666084632277489 |
| Loss at step 750: 0.04340723156929016 |
| Loss at step 800: 0.03913688659667969 |
| Loss at step 850: 0.055996235460042953 |
| Loss at step 900: 0.032421406358480453 |
| Mean training loss after epoch 174: 0.04112288686655351 |
|
|
| EPOCH: 175 |
| Loss at step 0: 0.028204411268234253 |
| Loss at step 50: 0.030527090653777122 |
| Loss at step 100: 0.039564263075590134 |
| Loss at step 150: 0.03648807108402252 |
| Loss at step 200: 0.042801812291145325 |
| Loss at step 250: 0.07026997953653336 |
| Loss at step 300: 0.053135115653276443 |
| Loss at step 350: 0.0397661030292511 |
| Loss at step 400: 0.047537870705127716 |
| Loss at step 450: 0.03553953766822815 |
| Loss at step 500: 0.0405605211853981 |
| Loss at step 550: 0.04098905622959137 |
| Loss at step 600: 0.05999923497438431 |
| Loss at step 650: 0.05660012736916542 |
| Loss at step 700: 0.03235200420022011 |
| Loss at step 750: 0.03290760517120361 |
| Loss at step 800: 0.034900546073913574 |
| Loss at step 850: 0.05151025578379631 |
| Loss at step 900: 0.041141875088214874 |
| Mean training loss after epoch 175: 0.04161210253096021 |
|
|
| EPOCH: 176 |
| Loss at step 0: 0.03776821494102478 |
| Loss at step 50: 0.04021279513835907 |
| Loss at step 100: 0.041916023939847946 |
| Loss at step 150: 0.03895966708660126 |
| Loss at step 200: 0.032551463693380356 |
| Loss at step 250: 0.05833392217755318 |
| Loss at step 300: 0.048504963517189026 |
| Loss at step 350: 0.05007625371217728 |
| Loss at step 400: 0.03597918152809143 |
| Loss at step 450: 0.030169488862156868 |
| Loss at step 500: 0.03819960728287697 |
| Loss at step 550: 0.037356436252593994 |
| Loss at step 600: 0.03851252421736717 |
| Loss at step 650: 0.06501714140176773 |
| Loss at step 700: 0.05189223960042 |
| Loss at step 750: 0.03168368339538574 |
| Loss at step 800: 0.03951912373304367 |
| Loss at step 850: 0.03564509004354477 |
| Loss at step 900: 0.03276278078556061 |
| Mean training loss after epoch 176: 0.04125692002546749 |
|
|
| EPOCH: 177 |
| Loss at step 0: 0.03515172004699707 |
| Loss at step 50: 0.03367337957024574 |
| Loss at step 100: 0.0318077951669693 |
| Loss at step 150: 0.03582939878106117 |
| Loss at step 200: 0.05992460623383522 |
| Loss at step 250: 0.0508149079978466 |
| Loss at step 300: 0.041243456304073334 |
| Loss at step 350: 0.05429394170641899 |
| Loss at step 400: 0.048130929470062256 |
| Loss at step 450: 0.035247184336185455 |
| Loss at step 500: 0.03542075678706169 |
| Loss at step 550: 0.055204953998327255 |
| Loss at step 600: 0.052897412329912186 |
| Loss at step 650: 0.06927145272493362 |
| Loss at step 700: 0.04794507846236229 |
| Loss at step 750: 0.03416883945465088 |
| Loss at step 800: 0.04960576817393303 |
| Loss at step 850: 0.03425171971321106 |
| Loss at step 900: 0.03997402638196945 |
| Mean training loss after epoch 177: 0.04163434571707681 |
|
|
| EPOCH: 178 |
| Loss at step 0: 0.04990256577730179 |
| Loss at step 50: 0.0538577102124691 |
| Loss at step 100: 0.03529036417603493 |
| Loss at step 150: 0.03500866889953613 |
| Loss at step 200: 0.03199452906847 |
| Loss at step 250: 0.036851897835731506 |
| Loss at step 300: 0.029136331751942635 |
| Loss at step 350: 0.030358897522091866 |
| Loss at step 400: 0.051947467029094696 |
| Loss at step 450: 0.037148743867874146 |
| Loss at step 500: 0.03686618059873581 |
| Loss at step 550: 0.036732930690050125 |
| Loss at step 600: 0.04222632199525833 |
| Loss at step 650: 0.03982901945710182 |
| Loss at step 700: 0.03642851859331131 |
| Loss at step 750: 0.04014425352215767 |
| Loss at step 800: 0.0999036505818367 |
| Loss at step 850: 0.038242340087890625 |
| Loss at step 900: 0.035852327942848206 |
| Mean training loss after epoch 178: 0.04207657766875936 |
|
|
| EPOCH: 179 |
| Loss at step 0: 0.04087178036570549 |
| Loss at step 50: 0.04753424972295761 |
| Loss at step 100: 0.03290014714002609 |
| Loss at step 150: 0.03695341572165489 |
| Loss at step 200: 0.03138614073395729 |
| Loss at step 250: 0.040319520980119705 |
| Loss at step 300: 0.049994390457868576 |
| Loss at step 350: 0.05205334722995758 |
| Loss at step 400: 0.06807678192853928 |
| Loss at step 450: 0.051330793648958206 |
| Loss at step 500: 0.03462173044681549 |
| Loss at step 550: 0.03375495225191116 |
| Loss at step 600: 0.036820266395807266 |
| Loss at step 650: 0.03800901770591736 |
| Loss at step 700: 0.030175233259797096 |
| Loss at step 750: 0.05230529233813286 |
| Loss at step 800: 0.03545716404914856 |
| Loss at step 850: 0.052242521196603775 |
| Loss at step 900: 0.030589766800403595 |
| Mean training loss after epoch 179: 0.04223245648003972 |
|
|
| EPOCH: 180 |
| Loss at step 0: 0.03807156905531883 |
| Loss at step 50: 0.05775832384824753 |
| Loss at step 100: 0.03275652229785919 |
| Loss at step 150: 0.03259390965104103 |
| Loss at step 200: 0.06442693620920181 |
| Loss at step 250: 0.04907452315092087 |
| Loss at step 300: 0.033002205193042755 |
| Loss at step 350: 0.05629289895296097 |
| Loss at step 400: 0.039508040994405746 |
| Loss at step 450: 0.03669936582446098 |
| Loss at step 500: 0.0339655764400959 |
| Loss at step 550: 0.035679228603839874 |
| Loss at step 600: 0.039923977106809616 |
| Loss at step 650: 0.05427751690149307 |
| Loss at step 700: 0.037826668471097946 |
| Loss at step 750: 0.035374172031879425 |
| Loss at step 800: 0.06559602171182632 |
| Loss at step 850: 0.03814015910029411 |
| Loss at step 900: 0.03909355774521828 |
| Mean training loss after epoch 180: 0.04139347126218937 |
|
|
| EPOCH: 181 |
| Loss at step 0: 0.03391748294234276 |
| Loss at step 50: 0.04056994244456291 |
| Loss at step 100: 0.049423567950725555 |
| Loss at step 150: 0.046806685626506805 |
| Loss at step 200: 0.03752606734633446 |
| Loss at step 250: 0.03340506553649902 |
| Loss at step 300: 0.04189173877239227 |
| Loss at step 350: 0.031038936227560043 |
| Loss at step 400: 0.0705329179763794 |
| Loss at step 450: 0.03440277650952339 |
| Loss at step 500: 0.0336567685008049 |
| Loss at step 550: 0.03429238125681877 |
| Loss at step 600: 0.031661275774240494 |
| Loss at step 650: 0.05041368678212166 |
| Loss at step 700: 0.030251847580075264 |
| Loss at step 750: 0.042865362018346786 |
| Loss at step 800: 0.033039726316928864 |
| Loss at step 850: 0.058725807815790176 |
| Loss at step 900: 0.03608013689517975 |
| Mean training loss after epoch 181: 0.04149253477157751 |
|
|
| EPOCH: 182 |
| Loss at step 0: 0.03710400313138962 |
| Loss at step 50: 0.03371499851346016 |
| Loss at step 100: 0.0341544970870018 |
| Loss at step 150: 0.04096578434109688 |
| Loss at step 200: 0.05157192423939705 |
| Loss at step 250: 0.03875488415360451 |
| Loss at step 300: 0.03593416139483452 |
| Loss at step 350: 0.03340543434023857 |
| Loss at step 400: 0.05967196449637413 |
| Loss at step 450: 0.03931530937552452 |
| Loss at step 500: 0.05755196511745453 |
| Loss at step 550: 0.03740415349602699 |
| Loss at step 600: 0.03923661261796951 |
| Loss at step 650: 0.040235571563243866 |
| Loss at step 700: 0.04200943186879158 |
| Loss at step 750: 0.03610384464263916 |
| Loss at step 800: 0.031481098383665085 |
| Loss at step 850: 0.0516573041677475 |
| Loss at step 900: 0.047245364636182785 |
| Mean training loss after epoch 182: 0.041903669381939145 |
|
|
| EPOCH: 183 |
| Loss at step 0: 0.04019775986671448 |
| Loss at step 50: 0.04790247604250908 |
| Loss at step 100: 0.04489818215370178 |
| Loss at step 150: 0.03816897049546242 |
| Loss at step 200: 0.033736344426870346 |
| Loss at step 250: 0.03466064855456352 |
| Loss at step 300: 0.040574636310338974 |
| Loss at step 350: 0.034679412841796875 |
| Loss at step 400: 0.034291334450244904 |
| Loss at step 450: 0.05978235602378845 |
| Loss at step 500: 0.04911112040281296 |
| Loss at step 550: 0.03342486545443535 |
| Loss at step 600: 0.052635472267866135 |
| Loss at step 650: 0.03890468180179596 |
| Loss at step 700: 0.03596392273902893 |
| Loss at step 750: 0.03404191508889198 |
| Loss at step 800: 0.05040137469768524 |
| Loss at step 850: 0.034710999578237534 |
| Loss at step 900: 0.038780633360147476 |
| Mean training loss after epoch 183: 0.041807248219371096 |
|
|
| EPOCH: 184 |
| Loss at step 0: 0.05211412534117699 |
| Loss at step 50: 0.04960509017109871 |
| Loss at step 100: 0.03784497454762459 |
| Loss at step 150: 0.03517859801650047 |
| Loss at step 200: 0.03709854185581207 |
| Loss at step 250: 0.03380673751235008 |
| Loss at step 300: 0.03791601583361626 |
| Loss at step 350: 0.04009474813938141 |
| Loss at step 400: 0.03905599191784859 |
| Loss at step 450: 0.03254836052656174 |
| Loss at step 500: 0.05125289410352707 |
| Loss at step 550: 0.038431569933891296 |
| Loss at step 600: 0.03682177513837814 |
| Loss at step 650: 0.03914055600762367 |
| Loss at step 700: 0.03562482073903084 |
| Loss at step 750: 0.04414800554513931 |
| Loss at step 800: 0.03894803300499916 |
| Loss at step 850: 0.03770125284790993 |
| Loss at step 900: 0.030288131907582283 |
| Mean training loss after epoch 184: 0.04138687876527752 |
|
|
| EPOCH: 185 |
| Loss at step 0: 0.03609955310821533 |
| Loss at step 50: 0.05682943016290665 |
| Loss at step 100: 0.038018904626369476 |
| Loss at step 150: 0.030884547159075737 |
| Loss at step 200: 0.041269998997449875 |
| Loss at step 250: 0.04373003542423248 |
| Loss at step 300: 0.035877473652362823 |
| Loss at step 350: 0.037849120795726776 |
| Loss at step 400: 0.043793316930532455 |
| Loss at step 450: 0.03450731933116913 |
| Loss at step 500: 0.04273483157157898 |
| Loss at step 550: 0.03578299283981323 |
| Loss at step 600: 0.032646261155605316 |
| Loss at step 650: 0.0462360754609108 |
| Loss at step 700: 0.054228439927101135 |
| Loss at step 750: 0.05155330151319504 |
| Loss at step 800: 0.03744790330529213 |
| Loss at step 850: 0.06680920720100403 |
| Loss at step 900: 0.03223875164985657 |
| Mean training loss after epoch 185: 0.041660312590981595 |
|
|
| EPOCH: 186 |
| Loss at step 0: 0.033333949744701385 |
| Loss at step 50: 0.05429576709866524 |
| Loss at step 100: 0.03890854865312576 |
| Loss at step 150: 0.03497329354286194 |
| Loss at step 200: 0.034743137657642365 |
| Loss at step 250: 0.0356815941631794 |
| Loss at step 300: 0.05399752035737038 |
| Loss at step 350: 0.03176051005721092 |
| Loss at step 400: 0.03740697726607323 |
| Loss at step 450: 0.04082540050148964 |
| Loss at step 500: 0.05412736162543297 |
| Loss at step 550: 0.03972582891583443 |
| Loss at step 600: 0.038021307438611984 |
| Loss at step 650: 0.04753376543521881 |
| Loss at step 700: 0.05196103826165199 |
| Loss at step 750: 0.04077177867293358 |
| Loss at step 800: 0.03566981106996536 |
| Loss at step 850: 0.03721268102526665 |
| Loss at step 900: 0.05761560797691345 |
| Mean training loss after epoch 186: 0.04170741890269175 |
|
|
| EPOCH: 187 |
| Loss at step 0: 0.03566679731011391 |
| Loss at step 50: 0.11628812551498413 |
| Loss at step 100: 0.03040335141122341 |
| Loss at step 150: 0.03489764779806137 |
| Loss at step 200: 0.05575336515903473 |
| Loss at step 250: 0.046886786818504333 |
| Loss at step 300: 0.035252176225185394 |
| Loss at step 350: 0.043795567005872726 |
| Loss at step 400: 0.028728723526000977 |
| Loss at step 450: 0.036280758678913116 |
| Loss at step 500: 0.04253661632537842 |
| Loss at step 550: 0.05303025618195534 |
| Loss at step 600: 0.03752442076802254 |
| Loss at step 650: 0.031123116612434387 |
| Loss at step 700: 0.03173130005598068 |
| Loss at step 750: 0.04688844829797745 |
| Loss at step 800: 0.036983974277973175 |
| Loss at step 850: 0.04824747145175934 |
| Loss at step 900: 0.039069145917892456 |
| Mean training loss after epoch 187: 0.04183094795626491 |
|
|
| EPOCH: 188 |
| Loss at step 0: 0.031547438353300095 |
| Loss at step 50: 0.04061029478907585 |
| Loss at step 100: 0.04006023705005646 |
| Loss at step 150: 0.029827365651726723 |
| Loss at step 200: 0.040378764271736145 |
| Loss at step 250: 0.03698677942156792 |
| Loss at step 300: 0.03617756441235542 |
| Loss at step 350: 0.03405392915010452 |
| Loss at step 400: 0.048338402062654495 |
| Loss at step 450: 0.030941160395741463 |
| Loss at step 500: 0.04532836750149727 |
| Loss at step 550: 0.041155822575092316 |
| Loss at step 600: 0.045638855546712875 |
| Loss at step 650: 0.029842229560017586 |
| Loss at step 700: 0.03459912911057472 |
| Loss at step 750: 0.043139681220054626 |
| Loss at step 800: 0.033857397735118866 |
| Loss at step 850: 0.037075892090797424 |
| Loss at step 900: 0.0419020876288414 |
| Mean training loss after epoch 188: 0.04069976957201132 |
|
|
| EPOCH: 189 |
| Loss at step 0: 0.033498603850603104 |
| Loss at step 50: 0.029143836349248886 |
| Loss at step 100: 0.05419186130166054 |
| Loss at step 150: 0.04173846170306206 |
| Loss at step 200: 0.037460848689079285 |
| Loss at step 250: 0.03811207413673401 |
| Loss at step 300: 0.07409770786762238 |
| Loss at step 350: 0.05379877984523773 |
| Loss at step 400: 0.0445668064057827 |
| Loss at step 450: 0.04734111204743385 |
| Loss at step 500: 0.03445335105061531 |
| Loss at step 550: 0.07331730425357819 |
| Loss at step 600: 0.03149990737438202 |
| Loss at step 650: 0.04199940711259842 |
| Loss at step 700: 0.05004751309752464 |
| Loss at step 750: 0.04749492183327675 |
| Loss at step 800: 0.03820522874593735 |
| Loss at step 850: 0.03572584688663483 |
| Loss at step 900: 0.04117851331830025 |
| Mean training loss after epoch 189: 0.041296352620429196 |
|
|
| EPOCH: 190 |
| Loss at step 0: 0.03617745265364647 |
| Loss at step 50: 0.03318284451961517 |
| Loss at step 100: 0.03383420780301094 |
| Loss at step 150: 0.048314373940229416 |
| Loss at step 200: 0.03600124642252922 |
| Loss at step 250: 0.07072673738002777 |
| Loss at step 300: 0.04310063272714615 |
| Loss at step 350: 0.03930129110813141 |
| Loss at step 400: 0.042464740574359894 |
| Loss at step 450: 0.03321487456560135 |
| Loss at step 500: 0.0326659195125103 |
| Loss at step 550: 0.03674949333071709 |
| Loss at step 600: 0.04988384619355202 |
| Loss at step 650: 0.03373335674405098 |
| Loss at step 700: 0.04392938315868378 |
| Loss at step 750: 0.0330236591398716 |
| Loss at step 800: 0.041451502591371536 |
| Loss at step 850: 0.03775708004832268 |
| Loss at step 900: 0.027917273342609406 |
| Mean training loss after epoch 190: 0.04126379593039182 |
|
|
| EPOCH: 191 |
| Loss at step 0: 0.03522908687591553 |
| Loss at step 50: 0.056289736181497574 |
| Loss at step 100: 0.04285778850317001 |
| Loss at step 150: 0.036079779267311096 |
| Loss at step 200: 0.030695710331201553 |
| Loss at step 250: 0.03508748486638069 |
| Loss at step 300: 0.06639096885919571 |
| Loss at step 350: 0.050350844860076904 |
| Loss at step 400: 0.035781338810920715 |
| Loss at step 450: 0.045949894934892654 |
| Loss at step 500: 0.02796305902302265 |
| Loss at step 550: 0.03321106359362602 |
| Loss at step 600: 0.0372857004404068 |
| Loss at step 650: 0.06294380873441696 |
| Loss at step 700: 0.05089360848069191 |
| Loss at step 750: 0.038762085139751434 |
| Loss at step 800: 0.039276279509067535 |
| Loss at step 850: 0.060114022344350815 |
| Loss at step 900: 0.04756484180688858 |
| Mean training loss after epoch 191: 0.041512265670369426 |
|
|
| EPOCH: 192 |
| Loss at step 0: 0.034350521862506866 |
| Loss at step 50: 0.04084480553865433 |
| Loss at step 100: 0.0338444747030735 |
| Loss at step 150: 0.054738275706768036 |
| Loss at step 200: 0.03252284228801727 |
| Loss at step 250: 0.05226755887269974 |
| Loss at step 300: 0.03735841438174248 |
| Loss at step 350: 0.03251798450946808 |
| Loss at step 400: 0.051244113594293594 |
| Loss at step 450: 0.03884818032383919 |
| Loss at step 500: 0.04193374887108803 |
| Loss at step 550: 0.033448901027441025 |
| Loss at step 600: 0.033464811742305756 |
| Loss at step 650: 0.039219241589307785 |
| Loss at step 700: 0.03877369686961174 |
| Loss at step 750: 0.031385522335767746 |
| Loss at step 800: 0.03320163115859032 |
| Loss at step 850: 0.05189124494791031 |
| Loss at step 900: 0.04130826145410538 |
| Mean training loss after epoch 192: 0.041748421510923776 |
|
|
| EPOCH: 193 |
| Loss at step 0: 0.04113989695906639 |
| Loss at step 50: 0.03168807551264763 |
| Loss at step 100: 0.035098232328891754 |
| Loss at step 150: 0.034963954240083694 |
| Loss at step 200: 0.05798143520951271 |
| Loss at step 250: 0.03390754386782646 |
| Loss at step 300: 0.04076037555932999 |
| Loss at step 350: 0.0436348132789135 |
| Loss at step 400: 0.03736273944377899 |
| Loss at step 450: 0.038291774690151215 |
| Loss at step 500: 0.040943972766399384 |
| Loss at step 550: 0.04079652577638626 |
| Loss at step 600: 0.04327332228422165 |
| Loss at step 650: 0.03669068217277527 |
| Loss at step 700: 0.037559207528829575 |
| Loss at step 750: 0.032399732619524 |
| Loss at step 800: 0.034300658851861954 |
| Loss at step 850: 0.03065013885498047 |
| Loss at step 900: 0.040528304874897 |
| Mean training loss after epoch 193: 0.04117203234776314 |
|
|
| EPOCH: 194 |
| Loss at step 0: 0.06797874718904495 |
| Loss at step 50: 0.052924975752830505 |
| Loss at step 100: 0.05212656781077385 |
| Loss at step 150: 0.03927493467926979 |
| Loss at step 200: 0.043678004294633865 |
| Loss at step 250: 0.036535218358039856 |
| Loss at step 300: 0.03414019197225571 |
| Loss at step 350: 0.03480542451143265 |
| Loss at step 400: 0.04807673394680023 |
| Loss at step 450: 0.03477676957845688 |
| Loss at step 500: 0.05026653781533241 |
| Loss at step 550: 0.049459606409072876 |
| Loss at step 600: 0.05336325243115425 |
| Loss at step 650: 0.05143073946237564 |
| Loss at step 700: 0.029971925541758537 |
| Loss at step 750: 0.040802404284477234 |
| Loss at step 800: 0.037547655403614044 |
| Loss at step 850: 0.03940505161881447 |
| Loss at step 900: 0.029753219336271286 |
| Mean training loss after epoch 194: 0.04150237108090285 |
|
|
| EPOCH: 195 |
| Loss at step 0: 0.03520885854959488 |
| Loss at step 50: 0.03639410436153412 |
| Loss at step 100: 0.034854043275117874 |
| Loss at step 150: 0.03551192209124565 |
| Loss at step 200: 0.025638144463300705 |
| Loss at step 250: 0.07378534972667694 |
| Loss at step 300: 0.058501437306404114 |
| Loss at step 350: 0.06586998701095581 |
| Loss at step 400: 0.0365779846906662 |
| Loss at step 450: 0.036176346242427826 |
| Loss at step 500: 0.04393264651298523 |
| Loss at step 550: 0.03411954268813133 |
| Loss at step 600: 0.03807177394628525 |
| Loss at step 650: 0.04445924982428551 |
| Loss at step 700: 0.06824284046888351 |
| Loss at step 750: 0.039268508553504944 |
| Loss at step 800: 0.02649448812007904 |
| Loss at step 850: 0.05371391400694847 |
| Loss at step 900: 0.06416977196931839 |
| Mean training loss after epoch 195: 0.04103545524449999 |
|
|
| EPOCH: 196 |
| Loss at step 0: 0.038097675889730453 |
| Loss at step 50: 0.044847916811704636 |
| Loss at step 100: 0.03804577887058258 |
| Loss at step 150: 0.03618156909942627 |
| Loss at step 200: 0.04062632471323013 |
| Loss at step 250: 0.04594080522656441 |
| Loss at step 300: 0.03366316854953766 |
| Loss at step 350: 0.032435376197099686 |
| Loss at step 400: 0.0293735284358263 |
| Loss at step 450: 0.034268464893102646 |
| Loss at step 500: 0.03657921031117439 |
| Loss at step 550: 0.040299192070961 |
| Loss at step 600: 0.03601702302694321 |
| Loss at step 650: 0.04306451603770256 |
| Loss at step 700: 0.03470020368695259 |
| Loss at step 750: 0.0481594018638134 |
| Loss at step 800: 0.04332111403346062 |
| Loss at step 850: 0.048539843410253525 |
| Loss at step 900: 0.040765244513750076 |
| Mean training loss after epoch 196: 0.04173837025671689 |
|
|
| EPOCH: 197 |
| Loss at step 0: 0.05736205354332924 |
| Loss at step 50: 0.027131736278533936 |
| Loss at step 100: 0.03562883287668228 |
| Loss at step 150: 0.03762797266244888 |
| Loss at step 200: 0.07137754559516907 |
| Loss at step 250: 0.06394952535629272 |
| Loss at step 300: 0.04078604280948639 |
| Loss at step 350: 0.04103813320398331 |
| Loss at step 400: 0.040748171508312225 |
| Loss at step 450: 0.049523528665304184 |
| Loss at step 500: 0.04065272584557533 |
| Loss at step 550: 0.05001842975616455 |
| Loss at step 600: 0.05851582810282707 |
| Loss at step 650: 0.0581602081656456 |
| Loss at step 700: 0.02705790475010872 |
| Loss at step 750: 0.040607865899801254 |
| Loss at step 800: 0.03715815767645836 |
| Loss at step 850: 0.03969978168606758 |
| Loss at step 900: 0.043329522013664246 |
| Mean training loss after epoch 197: 0.04137408531614458 |
|
|
| EPOCH: 198 |
| Loss at step 0: 0.03971967101097107 |
| Loss at step 50: 0.03631794452667236 |
| Loss at step 100: 0.0338369719684124 |
| Loss at step 150: 0.04944315552711487 |
| Loss at step 200: 0.04070665314793587 |
| Loss at step 250: 0.051509302109479904 |
| Loss at step 300: 0.05552169308066368 |
| Loss at step 350: 0.03374495357275009 |
| Loss at step 400: 0.03621584549546242 |
| Loss at step 450: 0.03770939260721207 |
| Loss at step 500: 0.03463056683540344 |
| Loss at step 550: 0.02788911759853363 |
| Loss at step 600: 0.04447398707270622 |
| Loss at step 650: 0.05413061007857323 |
| Loss at step 700: 0.04044802114367485 |
| Loss at step 750: 0.07355301827192307 |
| Loss at step 800: 0.03673677518963814 |
| Loss at step 850: 0.037305835634469986 |
| Loss at step 900: 0.0418182909488678 |
| Mean training loss after epoch 198: 0.041858961478050453 |
|
|
| EPOCH: 199 |
| Loss at step 0: 0.04305747523903847 |
| Loss at step 50: 0.03307000920176506 |
| Loss at step 100: 0.033796682953834534 |
| Loss at step 150: 0.0348166823387146 |
| Loss at step 200: 0.045635029673576355 |
| Loss at step 250: 0.03464725613594055 |
| Loss at step 300: 0.07720781862735748 |
| Loss at step 350: 0.024645846337080002 |
| Loss at step 400: 0.03345625847578049 |
| Loss at step 450: 0.03154450282454491 |
| Loss at step 500: 0.045259904116392136 |
| Loss at step 550: 0.03695496916770935 |
| Loss at step 600: 0.0499909333884716 |
| Loss at step 650: 0.04106828570365906 |
| Loss at step 700: 0.032556742429733276 |
| Loss at step 750: 0.03082137182354927 |
| Loss at step 800: 0.03785950317978859 |
| Loss at step 850: 0.05331188812851906 |
| Loss at step 900: 0.027672991156578064 |
| Mean training loss after epoch 199: 0.041491717618427425 |
|
|
| EPOCH: 200 |
| Loss at step 0: 0.035790350288152695 |
| Loss at step 50: 0.041217826306819916 |
| Loss at step 100: 0.034450091421604156 |
| Loss at step 150: 0.03561009466648102 |
| Loss at step 200: 0.03485352545976639 |
| Loss at step 250: 0.029213659465312958 |
| Loss at step 300: 0.04888638108968735 |
| Loss at step 350: 0.049630194902420044 |
| Loss at step 400: 0.040246348828077316 |
| Loss at step 450: 0.051989808678627014 |
| Loss at step 500: 0.03532884642481804 |
| Loss at step 550: 0.05131925269961357 |
| Loss at step 600: 0.07050848752260208 |
| Loss at step 650: 0.051529932767152786 |
| Loss at step 700: 0.031611260026693344 |
| Loss at step 750: 0.05834163725376129 |
| Loss at step 800: 0.03199863061308861 |
| Loss at step 850: 0.050334375351667404 |
| Loss at step 900: 0.048048146069049835 |
| Mean training loss after epoch 200: 0.04144771575633842 |
|
|
| EPOCH: 201 |
| Loss at step 0: 0.049343012273311615 |
| Loss at step 50: 0.03183600679039955 |
| Loss at step 100: 0.048347923904657364 |
| Loss at step 150: 0.05308234319090843 |
| Loss at step 200: 0.05167638882994652 |
| Loss at step 250: 0.04280967637896538 |
| Loss at step 300: 0.0367121621966362 |
| Loss at step 350: 0.05468958988785744 |
| Loss at step 400: 0.03841932862997055 |
| Loss at step 450: 0.034620095044374466 |
| Loss at step 500: 0.057135775685310364 |
| Loss at step 550: 0.041183676570653915 |
| Loss at step 600: 0.040930718183517456 |
| Loss at step 650: 0.03261233866214752 |
| Loss at step 700: 0.03489133343100548 |
| Loss at step 750: 0.02763674221932888 |
| Loss at step 800: 0.03225076198577881 |
| Loss at step 850: 0.05398840829730034 |
| Loss at step 900: 0.049016717821359634 |
| Mean training loss after epoch 201: 0.041507416116848175 |
|
|
| EPOCH: 202 |
| Loss at step 0: 0.03353920951485634 |
| Loss at step 50: 0.03718918189406395 |
| Loss at step 100: 0.03364242613315582 |
| Loss at step 150: 0.03372986242175102 |
| Loss at step 200: 0.056242551654577255 |
| Loss at step 250: 0.04020671918988228 |
| Loss at step 300: 0.03130819648504257 |
| Loss at step 350: 0.03483444079756737 |
| Loss at step 400: 0.0482604056596756 |
| Loss at step 450: 0.03938969224691391 |
| Loss at step 500: 0.05265633016824722 |
| Loss at step 550: 0.03554932773113251 |
| Loss at step 600: 0.03541647270321846 |
| Loss at step 650: 0.029953882098197937 |
| Loss at step 700: 0.03256510570645332 |
| Loss at step 750: 0.04815101623535156 |
| Loss at step 800: 0.038196250796318054 |
| Loss at step 850: 0.03245275840163231 |
| Loss at step 900: 0.03632979467511177 |
| Mean training loss after epoch 202: 0.0410298561172953 |
|
|
| EPOCH: 203 |
| Loss at step 0: 0.030459241941571236 |
| Loss at step 50: 0.043799057602882385 |
| Loss at step 100: 0.047694768756628036 |
| Loss at step 150: 0.03889485448598862 |
| Loss at step 200: 0.06752520054578781 |
| Loss at step 250: 0.02685130387544632 |
| Loss at step 300: 0.036014627665281296 |
| Loss at step 350: 0.037845950573682785 |
| Loss at step 400: 0.033889371901750565 |
| Loss at step 450: 0.031847402453422546 |
| Loss at step 500: 0.038399528712034225 |
| Loss at step 550: 0.03463640436530113 |
| Loss at step 600: 0.039087750017642975 |
| Loss at step 650: 0.03970421105623245 |
| Loss at step 700: 0.03944718837738037 |
| Loss at step 750: 0.03693550452589989 |
| Loss at step 800: 0.037040844559669495 |
| Loss at step 850: 0.0331515371799469 |
| Loss at step 900: 0.04205691069364548 |
| Mean training loss after epoch 203: 0.0412048772275289 |
|
|
| EPOCH: 204 |
| Loss at step 0: 0.056759487837553024 |
| Loss at step 50: 0.0413278266787529 |
| Loss at step 100: 0.040270932018756866 |
| Loss at step 150: 0.0361940898001194 |
| Loss at step 200: 0.07127005606889725 |
| Loss at step 250: 0.05027209222316742 |
| Loss at step 300: 0.04283789172768593 |
| Loss at step 350: 0.05113120749592781 |
| Loss at step 400: 0.04532406106591225 |
| Loss at step 450: 0.04405355453491211 |
| Loss at step 500: 0.057780180126428604 |
| Loss at step 550: 0.037918925285339355 |
| Loss at step 600: 0.03008476458489895 |
| Loss at step 650: 0.04085720703005791 |
| Loss at step 700: 0.04501824453473091 |
| Loss at step 750: 0.03670662268996239 |
| Loss at step 800: 0.034831445664167404 |
| Loss at step 850: 0.035773713141679764 |
| Loss at step 900: 0.03600168228149414 |
| Mean training loss after epoch 204: 0.04139673354616488 |
|
|
| EPOCH: 205 |
| Loss at step 0: 0.040792644023895264 |
| Loss at step 50: 0.039271481335163116 |
| Loss at step 100: 0.04059242457151413 |
| Loss at step 150: 0.03818690404295921 |
| Loss at step 200: 0.04294337332248688 |
| Loss at step 250: 0.057097338140010834 |
| Loss at step 300: 0.03699008747935295 |
| Loss at step 350: 0.05602728947997093 |
| Loss at step 400: 0.036347582936286926 |
| Loss at step 450: 0.03644917160272598 |
| Loss at step 500: 0.03737808018922806 |
| Loss at step 550: 0.03688392788171768 |
| Loss at step 600: 0.03842123597860336 |
| Loss at step 650: 0.031956832855939865 |
| Loss at step 700: 0.0357840359210968 |
| Loss at step 750: 0.037642884999513626 |
| Loss at step 800: 0.031830888241529465 |
| Loss at step 850: 0.03746681660413742 |
| Loss at step 900: 0.04127761349081993 |
| Mean training loss after epoch 205: 0.04121694114329273 |
|
|
| EPOCH: 206 |
| Loss at step 0: 0.03936978057026863 |
| Loss at step 50: 0.03137960657477379 |
| Loss at step 100: 0.028490858152508736 |
| Loss at step 150: 0.038255635648965836 |
| Loss at step 200: 0.035686857998371124 |
| Loss at step 250: 0.04872118681669235 |
| Loss at step 300: 0.05316340923309326 |
| Loss at step 350: 0.037066906690597534 |
| Loss at step 400: 0.034964218735694885 |
| Loss at step 450: 0.034100595861673355 |
| Loss at step 500: 0.030275076627731323 |
| Loss at step 550: 0.05340017378330231 |
| Loss at step 600: 0.03532257676124573 |
| Loss at step 650: 0.039942771196365356 |
| Loss at step 700: 0.04474440962076187 |
| Loss at step 750: 0.0567655973136425 |
| Loss at step 800: 0.038071874529123306 |
| Loss at step 850: 0.03440165892243385 |
| Loss at step 900: 0.03704487904906273 |
| Mean training loss after epoch 206: 0.04125204876557723 |
|
|
| EPOCH: 207 |
| Loss at step 0: 0.038039278239011765 |
| Loss at step 50: 0.04785553738474846 |
| Loss at step 100: 0.042042892426252365 |
| Loss at step 150: 0.038614820688962936 |
| Loss at step 200: 0.030518142506480217 |
| Loss at step 250: 0.03909027948975563 |
| Loss at step 300: 0.03875470906496048 |
| Loss at step 350: 0.03967742994427681 |
| Loss at step 400: 0.03191816061735153 |
| Loss at step 450: 0.046760812401771545 |
| Loss at step 500: 0.048585861921310425 |
| Loss at step 550: 0.0373445563018322 |
| Loss at step 600: 0.04253246262669563 |
| Loss at step 650: 0.043145034462213516 |
| Loss at step 700: 0.03523382544517517 |
| Loss at step 750: 0.05494747310876846 |
| Loss at step 800: 0.030589915812015533 |
| Loss at step 850: 0.03915799781680107 |
| Loss at step 900: 0.04673401266336441 |
| Mean training loss after epoch 207: 0.04131905022841781 |
|
|
| EPOCH: 208 |
| Loss at step 0: 0.04305552691221237 |
| Loss at step 50: 0.034908805042505264 |
| Loss at step 100: 0.03438510000705719 |
| Loss at step 150: 0.05595008656382561 |
| Loss at step 200: 0.036673370748758316 |
| Loss at step 250: 0.061115823686122894 |
| Loss at step 300: 0.03532424941658974 |
| Loss at step 350: 0.03127359226346016 |
| Loss at step 400: 0.036795247346162796 |
| Loss at step 450: 0.03030930832028389 |
| Loss at step 500: 0.047763705253601074 |
| Loss at step 550: 0.03661363571882248 |
| Loss at step 600: 0.05410737171769142 |
| Loss at step 650: 0.06448393315076828 |
| Loss at step 700: 0.04093127325177193 |
| Loss at step 750: 0.05752434581518173 |
| Loss at step 800: 0.03656463697552681 |
| Loss at step 850: 0.03313460201025009 |
| Loss at step 900: 0.038126688450574875 |
| Mean training loss after epoch 208: 0.04117663974351466 |
|
|
| EPOCH: 209 |
| Loss at step 0: 0.03493553772568703 |
| Loss at step 50: 0.03722948580980301 |
| Loss at step 100: 0.05398183688521385 |
| Loss at step 150: 0.0524870790541172 |
| Loss at step 200: 0.037257131189107895 |
| Loss at step 250: 0.04640546068549156 |
| Loss at step 300: 0.056868430227041245 |
| Loss at step 350: 0.04795893654227257 |
| Loss at step 400: 0.03179486468434334 |
| Loss at step 450: 0.03765290603041649 |
| Loss at step 500: 0.03589199110865593 |
| Loss at step 550: 0.0380818247795105 |
| Loss at step 600: 0.04972294718027115 |
| Loss at step 650: 0.029078220948576927 |
| Loss at step 700: 0.029039815068244934 |
| Loss at step 750: 0.03712788224220276 |
| Loss at step 800: 0.059975765645504 |
| Loss at step 850: 0.046831708401441574 |
| Loss at step 900: 0.062213387340307236 |
| Mean training loss after epoch 209: 0.04130176584254196 |
|
|
| EPOCH: 210 |
| Loss at step 0: 0.03962980955839157 |
| Loss at step 50: 0.03791427984833717 |
| Loss at step 100: 0.03497166559100151 |
| Loss at step 150: 0.031263984739780426 |
| Loss at step 200: 0.051545023918151855 |
| Loss at step 250: 0.03977767378091812 |
| Loss at step 300: 0.04709793999791145 |
| Loss at step 350: 0.05070801451802254 |
| Loss at step 400: 0.05128360167145729 |
| Loss at step 450: 0.02927989326417446 |
| Loss at step 500: 0.04444617033004761 |
| Loss at step 550: 0.03739051893353462 |
| Loss at step 600: 0.05879747122526169 |
| Loss at step 650: 0.053738340735435486 |
| Loss at step 700: 0.033911995589733124 |
| Loss at step 750: 0.04100339114665985 |
| Loss at step 800: 0.04177099093794823 |
| Loss at step 850: 0.05663182586431503 |
| Loss at step 900: 0.06499599665403366 |
| Mean training loss after epoch 210: 0.04155924991527791 |
|
|
| EPOCH: 211 |
| Loss at step 0: 0.030548926442861557 |
| Loss at step 50: 0.03719782829284668 |
| Loss at step 100: 0.02642173133790493 |
| Loss at step 150: 0.035055577754974365 |
| Loss at step 200: 0.047963909804821014 |
| Loss at step 250: 0.030586957931518555 |
| Loss at step 300: 0.03406506031751633 |
| Loss at step 350: 0.034562744200229645 |
| Loss at step 400: 0.05038195103406906 |
| Loss at step 450: 0.03493376076221466 |
| Loss at step 500: 0.040189724415540695 |
| Loss at step 550: 0.03963726758956909 |
| Loss at step 600: 0.05514547601342201 |
| Loss at step 650: 0.03667889162898064 |
| Loss at step 700: 0.03222879767417908 |
| Loss at step 750: 0.035416096448898315 |
| Loss at step 800: 0.041518017649650574 |
| Loss at step 850: 0.038832809776067734 |
| Loss at step 900: 0.04651348665356636 |
| Mean training loss after epoch 211: 0.040945966876963814 |
|
|
| EPOCH: 212 |
| Loss at step 0: 0.03327246010303497 |
| Loss at step 50: 0.05043778568506241 |
| Loss at step 100: 0.04172353446483612 |
| Loss at step 150: 0.03507307171821594 |
| Loss at step 200: 0.029459193348884583 |
| Loss at step 250: 0.04598444327712059 |
| Loss at step 300: 0.03865790367126465 |
| Loss at step 350: 0.033216699957847595 |
| Loss at step 400: 0.03927202522754669 |
| Loss at step 450: 0.053678590804338455 |
| Loss at step 500: 0.0357988066971302 |
| Loss at step 550: 0.03844377398490906 |
| Loss at step 600: 0.040219828486442566 |
| Loss at step 650: 0.053833648562431335 |
| Loss at step 700: 0.05301403999328613 |
| Loss at step 750: 0.03721046820282936 |
| Loss at step 800: 0.030222177505493164 |
| Loss at step 850: 0.03824635595083237 |
| Loss at step 900: 0.03393498808145523 |
| Mean training loss after epoch 212: 0.04123707259816529 |
|
|
| EPOCH: 213 |
| Loss at step 0: 0.05674567446112633 |
| Loss at step 50: 0.03383493423461914 |
| Loss at step 100: 0.034547653049230576 |
| Loss at step 150: 0.042903732508420944 |
| Loss at step 200: 0.03742729872465134 |
| Loss at step 250: 0.04211706295609474 |
| Loss at step 300: 0.03560095280408859 |
| Loss at step 350: 0.03154923394322395 |
| Loss at step 400: 0.03430347889661789 |
| Loss at step 450: 0.035916153341531754 |
| Loss at step 500: 0.04978935420513153 |
| Loss at step 550: 0.02993885800242424 |
| Loss at step 600: 0.0533515140414238 |
| Loss at step 650: 0.04010763391852379 |
| Loss at step 700: 0.03576485812664032 |
| Loss at step 750: 0.03653561323881149 |
| Loss at step 800: 0.034997548907995224 |
| Loss at step 850: 0.056449100375175476 |
| Loss at step 900: 0.0658298134803772 |
| Mean training loss after epoch 213: 0.0411296719149041 |
|
|
| EPOCH: 214 |
| Loss at step 0: 0.03312665596604347 |
| Loss at step 50: 0.03246608376502991 |
| Loss at step 100: 0.051583290100097656 |
| Loss at step 150: 0.04974968358874321 |
| Loss at step 200: 0.05574967712163925 |
| Loss at step 250: 0.04247640445828438 |
| Loss at step 300: 0.0370447002351284 |
| Loss at step 350: 0.03886210918426514 |
| Loss at step 400: 0.05328872799873352 |
| Loss at step 450: 0.03179633617401123 |
| Loss at step 500: 0.05542878434062004 |
| Loss at step 550: 0.044399116188287735 |
| Loss at step 600: 0.034311443567276 |
| Loss at step 650: 0.03347017616033554 |
| Loss at step 700: 0.033897753804922104 |
| Loss at step 750: 0.04861360043287277 |
| Loss at step 800: 0.03839123249053955 |
| Loss at step 850: 0.06305564939975739 |
| Loss at step 900: 0.03083120472729206 |
| Mean training loss after epoch 214: 0.041287495432965664 |
|
|
| EPOCH: 215 |
| Loss at step 0: 0.03692357987165451 |
| Loss at step 50: 0.03876027837395668 |
| Loss at step 100: 0.038466617465019226 |
| Loss at step 150: 0.03568306565284729 |
| Loss at step 200: 0.032258838415145874 |
| Loss at step 250: 0.033888742327690125 |
| Loss at step 300: 0.031359050422906876 |
| Loss at step 350: 0.0355716198682785 |
| Loss at step 400: 0.036228060722351074 |
| Loss at step 450: 0.034554749727249146 |
| Loss at step 500: 0.052061960101127625 |
| Loss at step 550: 0.03502297401428223 |
| Loss at step 600: 0.03877376392483711 |
| Loss at step 650: 0.07062792778015137 |
| Loss at step 700: 0.03748231753706932 |
| Loss at step 750: 0.048514317721128464 |
| Loss at step 800: 0.03904194012284279 |
| Loss at step 850: 0.042537979781627655 |
| Loss at step 900: 0.03580579534173012 |
| Mean training loss after epoch 215: 0.04120782274665482 |
|
|
| EPOCH: 216 |
| Loss at step 0: 0.03698866814374924 |
| Loss at step 50: 0.04032306373119354 |
| Loss at step 100: 0.04292820021510124 |
| Loss at step 150: 0.043672092258930206 |
| Loss at step 200: 0.04129884019494057 |
| Loss at step 250: 0.06413669139146805 |
| Loss at step 300: 0.03333058953285217 |
| Loss at step 350: 0.036104969680309296 |
| Loss at step 400: 0.04874488711357117 |
| Loss at step 450: 0.039387352764606476 |
| Loss at step 500: 0.043956201523542404 |
| Loss at step 550: 0.02769467793405056 |
| Loss at step 600: 0.04720773547887802 |
| Loss at step 650: 0.05882936343550682 |
| Loss at step 700: 0.04028856381773949 |
| Loss at step 750: 0.06633619964122772 |
| Loss at step 800: 0.04738272726535797 |
| Loss at step 850: 0.03966781124472618 |
| Loss at step 900: 0.05900513380765915 |
| Mean training loss after epoch 216: 0.04053880623988569 |
|
|
| EPOCH: 217 |
| Loss at step 0: 0.03476103022694588 |
| Loss at step 50: 0.04007076099514961 |
| Loss at step 100: 0.03548238426446915 |
| Loss at step 150: 0.0420597568154335 |
| Loss at step 200: 0.05475492775440216 |
| Loss at step 250: 0.03957177326083183 |
| Loss at step 300: 0.046720318496227264 |
| Loss at step 350: 0.037851348519325256 |
| Loss at step 400: 0.035817135125398636 |
| Loss at step 450: 0.03365528956055641 |
| Loss at step 500: 0.041856925934553146 |
| Loss at step 550: 0.03881705924868584 |
| Loss at step 600: 0.04341663047671318 |
| Loss at step 650: 0.0302233025431633 |
| Loss at step 700: 0.04495518282055855 |
| Loss at step 750: 0.07632829248905182 |
| Loss at step 800: 0.033838625997304916 |
| Loss at step 850: 0.058509018272161484 |
| Loss at step 900: 0.0263651292771101 |
| Mean training loss after epoch 217: 0.04165038813961976 |
|
|
| EPOCH: 218 |
| Loss at step 0: 0.03161349147558212 |
| Loss at step 50: 0.03128485754132271 |
| Loss at step 100: 0.04100488871335983 |
| Loss at step 150: 0.029969152063131332 |
| Loss at step 200: 0.04331596568226814 |
| Loss at step 250: 0.03943278640508652 |
| Loss at step 300: 0.033166710287332535 |
| Loss at step 350: 0.029378816485404968 |
| Loss at step 400: 0.0342346727848053 |
| Loss at step 450: 0.036455534398555756 |
| Loss at step 500: 0.05356838181614876 |
| Loss at step 550: 0.03679025173187256 |
| Loss at step 600: 0.03293558210134506 |
| Loss at step 650: 0.03905247524380684 |
| Loss at step 700: 0.049808014184236526 |
| Loss at step 750: 0.05253978446125984 |
| Loss at step 800: 0.04768317937850952 |
| Loss at step 850: 0.03520479053258896 |
| Loss at step 900: 0.05699870362877846 |
| Mean training loss after epoch 218: 0.04153027399770741 |
|
|
| EPOCH: 219 |
| Loss at step 0: 0.033535972237586975 |
| Loss at step 50: 0.05234470218420029 |
| Loss at step 100: 0.02671230025589466 |
| Loss at step 150: 0.03986935690045357 |
| Loss at step 200: 0.04880509898066521 |
| Loss at step 250: 0.04045942798256874 |
| Loss at step 300: 0.029259618371725082 |
| Loss at step 350: 0.04278722032904625 |
| Loss at step 400: 0.04514151066541672 |
| Loss at step 450: 0.04280414804816246 |
| Loss at step 500: 0.03450876101851463 |
| Loss at step 550: 0.04778549447655678 |
| Loss at step 600: 0.051221564412117004 |
| Loss at step 650: 0.03971226140856743 |
| Loss at step 700: 0.02804519422352314 |
| Loss at step 750: 0.03532065451145172 |
| Loss at step 800: 0.05204001069068909 |
| Loss at step 850: 0.0395912267267704 |
| Loss at step 900: 0.033726051449775696 |
| Mean training loss after epoch 219: 0.041007235062433714 |
|
|
| EPOCH: 220 |
| Loss at step 0: 0.0667114332318306 |
| Loss at step 50: 0.047925014048814774 |
| Loss at step 100: 0.03692636638879776 |
| Loss at step 150: 0.030362550169229507 |
| Loss at step 200: 0.035843972116708755 |
| Loss at step 250: 0.0372978039085865 |
| Loss at step 300: 0.03436199575662613 |
| Loss at step 350: 0.04721400514245033 |
| Loss at step 400: 0.034291476011276245 |
| Loss at step 450: 0.03911425918340683 |
| Loss at step 500: 0.03957908973097801 |
| Loss at step 550: 0.03273768350481987 |
| Loss at step 600: 0.033866237848997116 |
| Loss at step 650: 0.035394810140132904 |
| Loss at step 700: 0.05452428758144379 |
| Loss at step 750: 0.04825645312666893 |
| Loss at step 800: 0.04440128430724144 |
| Loss at step 850: 0.04680107533931732 |
| Loss at step 900: 0.03324844688177109 |
| Mean training loss after epoch 220: 0.04129367575311521 |
|
|
| EPOCH: 221 |
| Loss at step 0: 0.032262157648801804 |
| Loss at step 50: 0.038194503635168076 |
| Loss at step 100: 0.07236742973327637 |
| Loss at step 150: 0.06887383759021759 |
| Loss at step 200: 0.04868302121758461 |
| Loss at step 250: 0.03233625367283821 |
| Loss at step 300: 0.04105150327086449 |
| Loss at step 350: 0.029542885720729828 |
| Loss at step 400: 0.03776249662041664 |
| Loss at step 450: 0.03501143306493759 |
| Loss at step 500: 0.03716370835900307 |
| Loss at step 550: 0.03228867053985596 |
| Loss at step 600: 0.034387730062007904 |
| Loss at step 650: 0.05523118004202843 |
| Loss at step 700: 0.034401074051856995 |
| Loss at step 750: 0.04115273803472519 |
| Loss at step 800: 0.03586626797914505 |
| Loss at step 850: 0.04088988155126572 |
| Loss at step 900: 0.03924999013543129 |
| Mean training loss after epoch 221: 0.04116944233174009 |
|
|
| EPOCH: 222 |
| Loss at step 0: 0.04622489586472511 |
| Loss at step 50: 0.034178245812654495 |
| Loss at step 100: 0.03427834063768387 |
| Loss at step 150: 0.038221172988414764 |
| Loss at step 200: 0.04681423306465149 |
| Loss at step 250: 0.038031622767448425 |
| Loss at step 300: 0.039281293749809265 |
| Loss at step 350: 0.034416165202856064 |
| Loss at step 400: 0.07188202440738678 |
| Loss at step 450: 0.038608696311712265 |
| Loss at step 500: 0.03562668710947037 |
| Loss at step 550: 0.04744146391749382 |
| Loss at step 600: 0.03219371289014816 |
| Loss at step 650: 0.029797524213790894 |
| Loss at step 700: 0.036422885954380035 |
| Loss at step 750: 0.038234759122133255 |
| Loss at step 800: 0.04142580181360245 |
| Loss at step 850: 0.03657116740942001 |
| Loss at step 900: 0.02977190725505352 |
| Mean training loss after epoch 222: 0.041052178505545996 |
|
|
| EPOCH: 223 |
| Loss at step 0: 0.034760016947984695 |
| Loss at step 50: 0.03509802743792534 |
| Loss at step 100: 0.04074249789118767 |
| Loss at step 150: 0.03870717063546181 |
| Loss at step 200: 0.04823608696460724 |
| Loss at step 250: 0.02629156783223152 |
| Loss at step 300: 0.034971971064805984 |
| Loss at step 350: 0.03723495826125145 |
| Loss at step 400: 0.03717128559947014 |
| Loss at step 450: 0.039042744785547256 |
| Loss at step 500: 0.04182994365692139 |
| Loss at step 550: 0.03563202545046806 |
| Loss at step 600: 0.0682126060128212 |
| Loss at step 650: 0.04124102368950844 |
| Loss at step 700: 0.033520348370075226 |
| Loss at step 750: 0.0550982840359211 |
| Loss at step 800: 0.05311400443315506 |
| Loss at step 850: 0.041202362626791 |
| Loss at step 900: 0.07201080769300461 |
| Mean training loss after epoch 223: 0.04102069054291383 |
|
|
| EPOCH: 224 |
| Loss at step 0: 0.037563711404800415 |
| Loss at step 50: 0.05556620657444 |
| Loss at step 100: 0.04725373536348343 |
| Loss at step 150: 0.029180817306041718 |
| Loss at step 200: 0.06255805492401123 |
| Loss at step 250: 0.05550690367817879 |
| Loss at step 300: 0.032185718417167664 |
| Loss at step 350: 0.03732905164361 |
| Loss at step 400: 0.03720862418413162 |
| Loss at step 450: 0.05634133890271187 |
| Loss at step 500: 0.037644654512405396 |
| Loss at step 550: 0.041919078677892685 |
| Loss at step 600: 0.04104840010404587 |
| Loss at step 650: 0.04214031249284744 |
| Loss at step 700: 0.03333045169711113 |
| Loss at step 750: 0.035872429609298706 |
| Loss at step 800: 0.055177945643663406 |
| Loss at step 850: 0.06531449407339096 |
| Loss at step 900: 0.04556897282600403 |
| Mean training loss after epoch 224: 0.041014108520898734 |
|
|
| EPOCH: 225 |
| Loss at step 0: 0.03936973586678505 |
| Loss at step 50: 0.03350040689110756 |
| Loss at step 100: 0.037846505641937256 |
| Loss at step 150: 0.03458798676729202 |
| Loss at step 200: 0.0477675199508667 |
| Loss at step 250: 0.03837289661169052 |
| Loss at step 300: 0.03477032855153084 |
| Loss at step 350: 0.03714022412896156 |
| Loss at step 400: 0.03415047377347946 |
| Loss at step 450: 0.03395478427410126 |
| Loss at step 500: 0.0354052372276783 |
| Loss at step 550: 0.030104510486125946 |
| Loss at step 600: 0.043730005621910095 |
| Loss at step 650: 0.05497589707374573 |
| Loss at step 700: 0.05407802760601044 |
| Loss at step 750: 0.03253479674458504 |
| Loss at step 800: 0.03486725315451622 |
| Loss at step 850: 0.03352895379066467 |
| Loss at step 900: 0.03306787833571434 |
| Mean training loss after epoch 225: 0.0410375071050071 |
|
|
| EPOCH: 226 |
| Loss at step 0: 0.03648871183395386 |
| Loss at step 50: 0.0366818942129612 |
| Loss at step 100: 0.037782009690999985 |
| Loss at step 150: 0.047991879284381866 |
| Loss at step 200: 0.04437065124511719 |
| Loss at step 250: 0.05026853084564209 |
| Loss at step 300: 0.048000041395425797 |
| Loss at step 350: 0.031056242063641548 |
| Loss at step 400: 0.035417258739471436 |
| Loss at step 450: 0.04013531655073166 |
| Loss at step 500: 0.05145081505179405 |
| Loss at step 550: 0.04228727146983147 |
| Loss at step 600: 0.05478085204958916 |
| Loss at step 650: 0.04012715443968773 |
| Loss at step 700: 0.05043533816933632 |
| Loss at step 750: 0.038202911615371704 |
| Loss at step 800: 0.036201126873493195 |
| Loss at step 850: 0.03652939572930336 |
| Loss at step 900: 0.026116954162716866 |
| Mean training loss after epoch 226: 0.04136124605547263 |
|
|
| EPOCH: 227 |
| Loss at step 0: 0.04205761104822159 |
| Loss at step 50: 0.04090822488069534 |
| Loss at step 100: 0.043332166969776154 |
| Loss at step 150: 0.05705716460943222 |
| Loss at step 200: 0.03839666768908501 |
| Loss at step 250: 0.029485201463103294 |
| Loss at step 300: 0.056229788810014725 |
| Loss at step 350: 0.03689929470419884 |
| Loss at step 400: 0.030563808977603912 |
| Loss at step 450: 0.04460404813289642 |
| Loss at step 500: 0.03188980743288994 |
| Loss at step 550: 0.029740963131189346 |
| Loss at step 600: 0.035941313952207565 |
| Loss at step 650: 0.05650283768773079 |
| Loss at step 700: 0.04136836528778076 |
| Loss at step 750: 0.05103546008467674 |
| Loss at step 800: 0.049101848155260086 |
| Loss at step 850: 0.04119361191987991 |
| Loss at step 900: 0.034654609858989716 |
| Mean training loss after epoch 227: 0.04193418883263811 |
|
|
| EPOCH: 228 |
| Loss at step 0: 0.038860369473695755 |
| Loss at step 50: 0.03692442551255226 |
| Loss at step 100: 0.030994482338428497 |
| Loss at step 150: 0.037316933274269104 |
| Loss at step 200: 0.03066915273666382 |
| Loss at step 250: 0.03554951027035713 |
| Loss at step 300: 0.05594007298350334 |
| Loss at step 350: 0.05369291454553604 |
| Loss at step 400: 0.05064915493130684 |
| Loss at step 450: 0.04152677580714226 |
| Loss at step 500: 0.056227780878543854 |
| Loss at step 550: 0.04867443069815636 |
| Loss at step 600: 0.03939269483089447 |
| Loss at step 650: 0.0381941944360733 |
| Loss at step 700: 0.046426430344581604 |
| Loss at step 750: 0.03176429122686386 |
| Loss at step 800: 0.041528403759002686 |
| Loss at step 850: 0.039497584104537964 |
| Loss at step 900: 0.037410248070955276 |
| Mean training loss after epoch 228: 0.04146860102648293 |
|
|
| EPOCH: 229 |
| Loss at step 0: 0.03637472540140152 |
| Loss at step 50: 0.034893423318862915 |
| Loss at step 100: 0.04737719148397446 |
| Loss at step 150: 0.033185504376888275 |
| Loss at step 200: 0.0315871462225914 |
| Loss at step 250: 0.03322930634021759 |
| Loss at step 300: 0.03275240957736969 |
| Loss at step 350: 0.034512899816036224 |
| Loss at step 400: 0.041698455810546875 |
| Loss at step 450: 0.03479326516389847 |
| Loss at step 500: 0.04883550852537155 |
| Loss at step 550: 0.0731632336974144 |
| Loss at step 600: 0.06841642409563065 |
| Loss at step 650: 0.03672165796160698 |
| Loss at step 700: 0.042949128895998 |
| Loss at step 750: 0.03123210184276104 |
| Loss at step 800: 0.03435734286904335 |
| Loss at step 850: 0.029435977339744568 |
| Loss at step 900: 0.04522766172885895 |
| Mean training loss after epoch 229: 0.04051139461063246 |
|
|
| EPOCH: 230 |
| Loss at step 0: 0.03913334757089615 |
| Loss at step 50: 0.055999431759119034 |
| Loss at step 100: 0.05603145435452461 |
| Loss at step 150: 0.03465709835290909 |
| Loss at step 200: 0.03714108467102051 |
| Loss at step 250: 0.049302857369184494 |
| Loss at step 300: 0.051496539264917374 |
| Loss at step 350: 0.03816322982311249 |
| Loss at step 400: 0.04686564579606056 |
| Loss at step 450: 0.05239836499094963 |
| Loss at step 500: 0.04643171653151512 |
| Loss at step 550: 0.03184344619512558 |
| Loss at step 600: 0.06275387108325958 |
| Loss at step 650: 0.0721859559416771 |
| Loss at step 700: 0.036288753151893616 |
| Loss at step 750: 0.029158808290958405 |
| Loss at step 800: 0.041492756456136703 |
| Loss at step 850: 0.039844442158937454 |
| Loss at step 900: 0.03264705464243889 |
| Mean training loss after epoch 230: 0.04095347651413509 |
|
|
| EPOCH: 231 |
| Loss at step 0: 0.03575587272644043 |
| Loss at step 50: 0.03215237334370613 |
| Loss at step 100: 0.05879320204257965 |
| Loss at step 150: 0.03561503812670708 |
| Loss at step 200: 0.04932316765189171 |
| Loss at step 250: 0.031335704028606415 |
| Loss at step 300: 0.03901561349630356 |
| Loss at step 350: 0.038512203842401505 |
| Loss at step 400: 0.03318621963262558 |
| Loss at step 450: 0.04640994220972061 |
| Loss at step 500: 0.032737333327531815 |
| Loss at step 550: 0.048904407769441605 |
| Loss at step 600: 0.03512667119503021 |
| Loss at step 650: 0.03650322183966637 |
| Loss at step 700: 0.038705259561538696 |
| Loss at step 750: 0.04180947691202164 |
| Loss at step 800: 0.04233972355723381 |
| Loss at step 850: 0.04108935967087746 |
| Loss at step 900: 0.03689875453710556 |
| Mean training loss after epoch 231: 0.04083216685388706 |
|
|
| EPOCH: 232 |
| Loss at step 0: 0.04886999726295471 |
| Loss at step 50: 0.05551153048872948 |
| Loss at step 100: 0.03532155975699425 |
| Loss at step 150: 0.038711488246917725 |
| Loss at step 200: 0.03422563523054123 |
| Loss at step 250: 0.0415818989276886 |
| Loss at step 300: 0.052041277289390564 |
| Loss at step 350: 0.031493090093135834 |
| Loss at step 400: 0.041716963052749634 |
| Loss at step 450: 0.045627281069755554 |
| Loss at step 500: 0.03781002387404442 |
| Loss at step 550: 0.04126285761594772 |
| Loss at step 600: 0.055617816746234894 |
| Loss at step 650: 0.030421976000070572 |
| Loss at step 700: 0.05503896623849869 |
| Loss at step 750: 0.04677780345082283 |
| Loss at step 800: 0.05122390016913414 |
| Loss at step 850: 0.05126957967877388 |
| Loss at step 900: 0.038854584097862244 |
| Mean training loss after epoch 232: 0.04083698726237329 |
|
|
| EPOCH: 233 |
| Loss at step 0: 0.03805645555257797 |
| Loss at step 50: 0.05763966590166092 |
| Loss at step 100: 0.04127464070916176 |
| Loss at step 150: 0.03656071051955223 |
| Loss at step 200: 0.04233010485768318 |
| Loss at step 250: 0.06715351343154907 |
| Loss at step 300: 0.056858159601688385 |
| Loss at step 350: 0.038870055228471756 |
| Loss at step 400: 0.04549760743975639 |
| Loss at step 450: 0.03370843455195427 |
| Loss at step 500: 0.036580368876457214 |
| Loss at step 550: 0.030094511806964874 |
| Loss at step 600: 0.027368923649191856 |
| Loss at step 650: 0.036164265125989914 |
| Loss at step 700: 0.03536178916692734 |
| Loss at step 750: 0.03123387135565281 |
| Loss at step 800: 0.03417902812361717 |
| Loss at step 850: 0.06495968252420425 |
| Loss at step 900: 0.033987708389759064 |
| Mean training loss after epoch 233: 0.04114598854343647 |
|
|
| EPOCH: 234 |
| Loss at step 0: 0.034131698310375214 |
| Loss at step 50: 0.03304879739880562 |
| Loss at step 100: 0.0375090166926384 |
| Loss at step 150: 0.04361898824572563 |
| Loss at step 200: 0.03519846498966217 |
| Loss at step 250: 0.03362388163805008 |
| Loss at step 300: 0.04630090668797493 |
| Loss at step 350: 0.03330394998192787 |
| Loss at step 400: 0.0410035215318203 |
| Loss at step 450: 0.03329356387257576 |
| Loss at step 500: 0.0334615595638752 |
| Loss at step 550: 0.0367317870259285 |
| Loss at step 600: 0.031206030398607254 |
| Loss at step 650: 0.0366726778447628 |
| Loss at step 700: 0.041628316044807434 |
| Loss at step 750: 0.03506317362189293 |
| Loss at step 800: 0.0323176383972168 |
| Loss at step 850: 0.03223805129528046 |
| Loss at step 900: 0.04771678149700165 |
| Mean training loss after epoch 234: 0.0408894696604532 |
|
|
| EPOCH: 235 |
| Loss at step 0: 0.036682069301605225 |
| Loss at step 50: 0.03544003888964653 |
| Loss at step 100: 0.03550875186920166 |
| Loss at step 150: 0.038712434470653534 |
| Loss at step 200: 0.0328836590051651 |
| Loss at step 250: 0.03970502316951752 |
| Loss at step 300: 0.033910173922777176 |
| Loss at step 350: 0.037062786519527435 |
| Loss at step 400: 0.07173572480678558 |
| Loss at step 450: 0.059028297662734985 |
| Loss at step 500: 0.033802054822444916 |
| Loss at step 550: 0.03551178053021431 |
| Loss at step 600: 0.050311341881752014 |
| Loss at step 650: 0.02960454858839512 |
| Loss at step 700: 0.044369637966156006 |
| Loss at step 750: 0.03446542099118233 |
| Loss at step 800: 0.026896147057414055 |
| Loss at step 850: 0.0351390466094017 |
| Loss at step 900: 0.03585878759622574 |
| Mean training loss after epoch 235: 0.0412483900519354 |
|
|
| EPOCH: 236 |
| Loss at step 0: 0.03225017711520195 |
| Loss at step 50: 0.033339180052280426 |
| Loss at step 100: 0.035093218088150024 |
| Loss at step 150: 0.041725076735019684 |
| Loss at step 200: 0.039611514657735825 |
| Loss at step 250: 0.03811977803707123 |
| Loss at step 300: 0.04116872325539589 |
| Loss at step 350: 0.04522959887981415 |
| Loss at step 400: 0.034504808485507965 |
| Loss at step 450: 0.03280076012015343 |
| Loss at step 500: 0.052272897213697433 |
| Loss at step 550: 0.035520099103450775 |
| Loss at step 600: 0.029213305562734604 |
| Loss at step 650: 0.03851242735981941 |
| Loss at step 700: 0.041083406656980515 |
| Loss at step 750: 0.047109778970479965 |
| Loss at step 800: 0.046369973570108414 |
| Loss at step 850: 0.03564689680933952 |
| Loss at step 900: 0.03542470932006836 |
| Mean training loss after epoch 236: 0.04041676042970818 |
|
|
| EPOCH: 237 |
| Loss at step 0: 0.03667047992348671 |
| Loss at step 50: 0.05486058071255684 |
| Loss at step 100: 0.03827936202287674 |
| Loss at step 150: 0.0426441989839077 |
| Loss at step 200: 0.03697700798511505 |
| Loss at step 250: 0.034738317131996155 |
| Loss at step 300: 0.03901257738471031 |
| Loss at step 350: 0.04829375445842743 |
| Loss at step 400: 0.0641462430357933 |
| Loss at step 450: 0.055503036826848984 |
| Loss at step 500: 0.0356808640062809 |
| Loss at step 550: 0.0351221039891243 |
| Loss at step 600: 0.04857305809855461 |
| Loss at step 650: 0.035922374576330185 |
| Loss at step 700: 0.0438312292098999 |
| Loss at step 750: 0.028933558613061905 |
| Loss at step 800: 0.049483731389045715 |
| Loss at step 850: 0.0362832210958004 |
| Loss at step 900: 0.03917879983782768 |
| Mean training loss after epoch 237: 0.04058157685580157 |
|
|
| EPOCH: 238 |
| Loss at step 0: 0.05662114545702934 |
| Loss at step 50: 0.07133477181196213 |
| Loss at step 100: 0.05366216599941254 |
| Loss at step 150: 0.05503072589635849 |
| Loss at step 200: 0.0364266000688076 |
| Loss at step 250: 0.03105452097952366 |
| Loss at step 300: 0.0429583378136158 |
| Loss at step 350: 0.03313998878002167 |
| Loss at step 400: 0.0366363525390625 |
| Loss at step 450: 0.02768549509346485 |
| Loss at step 500: 0.03483203426003456 |
| Loss at step 550: 0.04857843741774559 |
| Loss at step 600: 0.03954724594950676 |
| Loss at step 650: 0.035047147423028946 |
| Loss at step 700: 0.0341925173997879 |
| Loss at step 750: 0.03897126764059067 |
| Loss at step 800: 0.041156038641929626 |
| Loss at step 850: 0.042894527316093445 |
| Loss at step 900: 0.040707413107156754 |
| Mean training loss after epoch 238: 0.04040791804808925 |
|
|
| EPOCH: 239 |
| Loss at step 0: 0.032015059143304825 |
| Loss at step 50: 0.049798812717199326 |
| Loss at step 100: 0.029131358489394188 |
| Loss at step 150: 0.03209542855620384 |
| Loss at step 200: 0.039810456335544586 |
| Loss at step 250: 0.04038550332188606 |
| Loss at step 300: 0.040614526718854904 |
| Loss at step 350: 0.03396230190992355 |
| Loss at step 400: 0.05594230443239212 |
| Loss at step 450: 0.05109965801239014 |
| Loss at step 500: 0.03540639579296112 |
| Loss at step 550: 0.03735024482011795 |
| Loss at step 600: 0.03928496688604355 |
| Loss at step 650: 0.03327350690960884 |
| Loss at step 700: 0.04737665131688118 |
| Loss at step 750: 0.030339328572154045 |
| Loss at step 800: 0.03868836164474487 |
| Loss at step 850: 0.04888433218002319 |
| Loss at step 900: 0.054284047335386276 |
| Mean training loss after epoch 239: 0.040756947924889354 |
|
|
| EPOCH: 240 |
| Loss at step 0: 0.03621514514088631 |
| Loss at step 50: 0.037960972636938095 |
| Loss at step 100: 0.04036441445350647 |
| Loss at step 150: 0.04665497690439224 |
| Loss at step 200: 0.04599844664335251 |
| Loss at step 250: 0.03350859135389328 |
| Loss at step 300: 0.03978145867586136 |
| Loss at step 350: 0.04724019765853882 |
| Loss at step 400: 0.046361032873392105 |
| Loss at step 450: 0.04888570308685303 |
| Loss at step 500: 0.03416692838072777 |
| Loss at step 550: 0.0357830747961998 |
| Loss at step 600: 0.032979223877191544 |
| Loss at step 650: 0.04058676213026047 |
| Loss at step 700: 0.04173377901315689 |
| Loss at step 750: 0.0481884628534317 |
| Loss at step 800: 0.049226146191358566 |
| Loss at step 850: 0.03608490899205208 |
| Loss at step 900: 0.04782750457525253 |
| Mean training loss after epoch 240: 0.04125069013274491 |
|
|
| EPOCH: 241 |
| Loss at step 0: 0.02865777723491192 |
| Loss at step 50: 0.030689535662531853 |
| Loss at step 100: 0.03503900393843651 |
| Loss at step 150: 0.043451108038425446 |
| Loss at step 200: 0.03308131545782089 |
| Loss at step 250: 0.03468373790383339 |
| Loss at step 300: 0.04644292965531349 |
| Loss at step 350: 0.043308407068252563 |
| Loss at step 400: 0.05058228597044945 |
| Loss at step 450: 0.0519171804189682 |
| Loss at step 500: 0.04207949712872505 |
| Loss at step 550: 0.04765189439058304 |
| Loss at step 600: 0.0473044328391552 |
| Loss at step 650: 0.042676154524087906 |
| Loss at step 700: 0.032991547137498856 |
| Loss at step 750: 0.03649139404296875 |
| Loss at step 800: 0.06195422634482384 |
| Loss at step 850: 0.03607760742306709 |
| Loss at step 900: 0.03868535906076431 |
| Mean training loss after epoch 241: 0.04128943655345994 |
|
|
| EPOCH: 242 |
| Loss at step 0: 0.06794758886098862 |
| Loss at step 50: 0.03913401439785957 |
| Loss at step 100: 0.03793284669518471 |
| Loss at step 150: 0.04773392528295517 |
| Loss at step 200: 0.050653185695409775 |
| Loss at step 250: 0.025860438123345375 |
| Loss at step 300: 0.031997546553611755 |
| Loss at step 350: 0.03752926364541054 |
| Loss at step 400: 0.03538594767451286 |
| Loss at step 450: 0.029974274337291718 |
| Loss at step 500: 0.06059102341532707 |
| Loss at step 550: 0.039873912930488586 |
| Loss at step 600: 0.03409953415393829 |
| Loss at step 650: 0.03079247660934925 |
| Loss at step 700: 0.04495513439178467 |
| Loss at step 750: 0.03563304618000984 |
| Loss at step 800: 0.06306657195091248 |
| Loss at step 850: 0.03369851037859917 |
| Loss at step 900: 0.03801652416586876 |
| Mean training loss after epoch 242: 0.04038986093652591 |
|
|
| EPOCH: 243 |
| Loss at step 0: 0.04689265042543411 |
| Loss at step 50: 0.03451554477214813 |
| Loss at step 100: 0.03268691524863243 |
| Loss at step 150: 0.03130830079317093 |
| Loss at step 200: 0.03496095910668373 |
| Loss at step 250: 0.04636874422430992 |
| Loss at step 300: 0.037325453013181686 |
| Loss at step 350: 0.03737280145287514 |
| Loss at step 400: 0.0473397895693779 |
| Loss at step 450: 0.05233558267354965 |
| Loss at step 500: 0.03210145980119705 |
| Loss at step 550: 0.03945494815707207 |
| Loss at step 600: 0.03543688729405403 |
| Loss at step 650: 0.03888096660375595 |
| Loss at step 700: 0.04111270606517792 |
| Loss at step 750: 0.029099611565470695 |
| Loss at step 800: 0.03430042788386345 |
| Loss at step 850: 0.03829982504248619 |
| Loss at step 900: 0.039772406220436096 |
| Mean training loss after epoch 243: 0.040650974677156795 |
|
|
| EPOCH: 244 |
| Loss at step 0: 0.03601839020848274 |
| Loss at step 50: 0.03880251199007034 |
| Loss at step 100: 0.03493691235780716 |
| Loss at step 150: 0.02994871512055397 |
| Loss at step 200: 0.032999180257320404 |
| Loss at step 250: 0.03631100431084633 |
| Loss at step 300: 0.033189717680215836 |
| Loss at step 350: 0.03836946189403534 |
| Loss at step 400: 0.0370786115527153 |
| Loss at step 450: 0.033676642924547195 |
| Loss at step 500: 0.0403299480676651 |
| Loss at step 550: 0.03196725621819496 |
| Loss at step 600: 0.04806798696517944 |
| Loss at step 650: 0.04054030403494835 |
| Loss at step 700: 0.03655192628502846 |
| Loss at step 750: 0.042547885328531265 |
| Loss at step 800: 0.03825114667415619 |
| Loss at step 850: 0.03418620303273201 |
| Loss at step 900: 0.031493254005908966 |
| Mean training loss after epoch 244: 0.0404629212424858 |
|
|
| EPOCH: 245 |
| Loss at step 0: 0.037501245737075806 |
| Loss at step 50: 0.036022938787937164 |
| Loss at step 100: 0.06744180619716644 |
| Loss at step 150: 0.04214358702301979 |
| Loss at step 200: 0.034945204854011536 |
| Loss at step 250: 0.029832666739821434 |
| Loss at step 300: 0.054127588868141174 |
| Loss at step 350: 0.027778994292020798 |
| Loss at step 400: 0.03888265788555145 |
| Loss at step 450: 0.03413667902350426 |
| Loss at step 500: 0.039878036826848984 |
| Loss at step 550: 0.0383932963013649 |
| Loss at step 600: 0.03586817532777786 |
| Loss at step 650: 0.04729319363832474 |
| Loss at step 700: 0.060726214200258255 |
| Loss at step 750: 0.032855063676834106 |
| Loss at step 800: 0.03362588211894035 |
| Loss at step 850: 0.0399589017033577 |
| Loss at step 900: 0.03648059815168381 |
| Mean training loss after epoch 245: 0.040341950649185096 |
|
|
| EPOCH: 246 |
| Loss at step 0: 0.031979892402887344 |
| Loss at step 50: 0.06765910238027573 |
| Loss at step 100: 0.0613054521381855 |
| Loss at step 150: 0.03503995016217232 |
| Loss at step 200: 0.03813648223876953 |
| Loss at step 250: 0.03782317787408829 |
| Loss at step 300: 0.03368839621543884 |
| Loss at step 350: 0.03500468283891678 |
| Loss at step 400: 0.04959193989634514 |
| Loss at step 450: 0.0331171415746212 |
| Loss at step 500: 0.03166046738624573 |
| Loss at step 550: 0.037088554352521896 |
| Loss at step 600: 0.03945896402001381 |
| Loss at step 650: 0.03642991930246353 |
| Loss at step 700: 0.03628809377551079 |
| Loss at step 750: 0.05210085213184357 |
| Loss at step 800: 0.03868551924824715 |
| Loss at step 850: 0.041673678904771805 |
| Loss at step 900: 0.046727459877729416 |
| Mean training loss after epoch 246: 0.04055890298164539 |
|
|
| EPOCH: 247 |
| Loss at step 0: 0.050526224076747894 |
| Loss at step 50: 0.03471115231513977 |
| Loss at step 100: 0.02761066146194935 |
| Loss at step 150: 0.03715616837143898 |
| Loss at step 200: 0.06270107626914978 |
| Loss at step 250: 0.036140527576208115 |
| Loss at step 300: 0.03348053991794586 |
| Loss at step 350: 0.04763561859726906 |
| Loss at step 400: 0.04335961118340492 |
| Loss at step 450: 0.03691563382744789 |
| Loss at step 500: 0.03904525935649872 |
| Loss at step 550: 0.04260946810245514 |
| Loss at step 600: 0.049258239567279816 |
| Loss at step 650: 0.03189285472035408 |
| Loss at step 700: 0.0375247485935688 |
| Loss at step 750: 0.03473655879497528 |
| Loss at step 800: 0.03262423351407051 |
| Loss at step 850: 0.046003468334674835 |
| Loss at step 900: 0.0298653244972229 |
| Mean training loss after epoch 247: 0.04082909842401044 |
|
|
| EPOCH: 248 |
| Loss at step 0: 0.037177521735429764 |
| Loss at step 50: 0.03851611167192459 |
| Loss at step 100: 0.02362552471458912 |
| Loss at step 150: 0.04929950833320618 |
| Loss at step 200: 0.03481481969356537 |
| Loss at step 250: 0.030810419470071793 |
| Loss at step 300: 0.060464564710855484 |
| Loss at step 350: 0.03210290148854256 |
| Loss at step 400: 0.03180817514657974 |
| Loss at step 450: 0.03282691538333893 |
| Loss at step 500: 0.04492408409714699 |
| Loss at step 550: 0.04005492478609085 |
| Loss at step 600: 0.03255147859454155 |
| Loss at step 650: 0.039729081094264984 |
| Loss at step 700: 0.0332181416451931 |
| Loss at step 750: 0.03630407899618149 |
| Loss at step 800: 0.025422068312764168 |
| Loss at step 850: 0.028221523389220238 |
| Loss at step 900: 0.03473244234919548 |
| Mean training loss after epoch 248: 0.040842697498942614 |
|
|
| EPOCH: 249 |
| Loss at step 0: 0.04472397640347481 |
| Loss at step 50: 0.04787271097302437 |
| Loss at step 100: 0.04805320128798485 |
| Loss at step 150: 0.034147415310144424 |
| Loss at step 200: 0.03937199339270592 |
| Loss at step 250: 0.032839335501194 |
| Loss at step 300: 0.030625566840171814 |
| Loss at step 350: 0.06075826659798622 |
| Loss at step 400: 0.030696650967001915 |
| Loss at step 450: 0.043999478220939636 |
| Loss at step 500: 0.05251993238925934 |
| Loss at step 550: 0.03596007078886032 |
| Loss at step 600: 0.03571072220802307 |
| Loss at step 650: 0.03692999854683876 |
| Loss at step 700: 0.05128655582666397 |
| Loss at step 750: 0.03889640048146248 |
| Loss at step 800: 0.0332813523709774 |
| Loss at step 850: 0.05229688435792923 |
| Loss at step 900: 0.03806335851550102 |
| Mean training loss after epoch 249: 0.04073733212088725 |
|
|
| EPOCH: 250 |
| Loss at step 0: 0.03640764579176903 |
| Loss at step 50: 0.029529381543397903 |
| Loss at step 100: 0.03215811029076576 |
| Loss at step 150: 0.03508197143673897 |
| Loss at step 200: 0.06036895886063576 |
| Loss at step 250: 0.05026475340127945 |
| Loss at step 300: 0.03762553259730339 |
| Loss at step 350: 0.039728518575429916 |
| Loss at step 400: 0.051400668919086456 |
| Loss at step 450: 0.03622720018029213 |
| Loss at step 500: 0.04798509180545807 |
| Loss at step 550: 0.036011118441820145 |
| Loss at step 600: 0.03086777590215206 |
| Loss at step 650: 0.02947012148797512 |
| Loss at step 700: 0.033220358192920685 |
| Loss at step 750: 0.04179084300994873 |
| Loss at step 800: 0.03159976005554199 |
| Loss at step 850: 0.0499589778482914 |
| Loss at step 900: 0.05065244436264038 |
| Mean training loss after epoch 250: 0.04118522900396954 |
|
|
| EPOCH: 251 |
| Loss at step 0: 0.02755182981491089 |
| Loss at step 50: 0.049285098910331726 |
| Loss at step 100: 0.03654266893863678 |
| Loss at step 150: 0.04497296363115311 |
| Loss at step 200: 0.03556593507528305 |
| Loss at step 250: 0.03584232181310654 |
| Loss at step 300: 0.032394710928201675 |
| Loss at step 350: 0.03097252920269966 |
| Loss at step 400: 0.03149956464767456 |
| Loss at step 450: 0.034771617501974106 |
| Loss at step 500: 0.03608296439051628 |
| Loss at step 550: 0.04529433697462082 |
| Loss at step 600: 0.033899255096912384 |
| Loss at step 650: 0.041714031249284744 |
| Loss at step 700: 0.03436104208230972 |
| Loss at step 750: 0.03629880025982857 |
| Loss at step 800: 0.038352809846401215 |
| Loss at step 850: 0.048439860343933105 |
| Loss at step 900: 0.040245577692985535 |
| Mean training loss after epoch 251: 0.04107680348063837 |
|
|
| EPOCH: 252 |
| Loss at step 0: 0.04162813723087311 |
| Loss at step 50: 0.03678770735859871 |
| Loss at step 100: 0.036010533571243286 |
| Loss at step 150: 0.05083076283335686 |
| Loss at step 200: 0.04758510738611221 |
| Loss at step 250: 0.04673706740140915 |
| Loss at step 300: 0.03488384932279587 |
| Loss at step 350: 0.04749500751495361 |
| Loss at step 400: 0.03849106281995773 |
| Loss at step 450: 0.0354052409529686 |
| Loss at step 500: 0.032203562557697296 |
| Loss at step 550: 0.046269387006759644 |
| Loss at step 600: 0.03831486403942108 |
| Loss at step 650: 0.04683487117290497 |
| Loss at step 700: 0.03641248121857643 |
| Loss at step 750: 0.03526393696665764 |
| Loss at step 800: 0.039593227207660675 |
| Loss at step 850: 0.037872251123189926 |
| Loss at step 900: 0.03107110597193241 |
| Mean training loss after epoch 252: 0.040780991514417914 |
|
|
| EPOCH: 253 |
| Loss at step 0: 0.04510458558797836 |
| Loss at step 50: 0.053014062345027924 |
| Loss at step 100: 0.032169047743082047 |
| Loss at step 150: 0.040137842297554016 |
| Loss at step 200: 0.03011404722929001 |
| Loss at step 250: 0.049014877527952194 |
| Loss at step 300: 0.04769476503133774 |
| Loss at step 350: 0.06785585731267929 |
| Loss at step 400: 0.05150467902421951 |
| Loss at step 450: 0.037310872226953506 |
| Loss at step 500: 0.03545456752181053 |
| Loss at step 550: 0.0527246855199337 |
| Loss at step 600: 0.034180376678705215 |
| Loss at step 650: 0.03249421715736389 |
| Loss at step 700: 0.03964534029364586 |
| Loss at step 750: 0.04362183064222336 |
| Loss at step 800: 0.033075109124183655 |
| Loss at step 850: 0.037886276841163635 |
| Loss at step 900: 0.04572896659374237 |
| Mean training loss after epoch 253: 0.040567862515304004 |
|
|
| EPOCH: 254 |
| Loss at step 0: 0.03452959656715393 |
| Loss at step 50: 0.03886321932077408 |
| Loss at step 100: 0.04838445782661438 |
| Loss at step 150: 0.03837645798921585 |
| Loss at step 200: 0.03629623353481293 |
| Loss at step 250: 0.030103985220193863 |
| Loss at step 300: 0.03791821002960205 |
| Loss at step 350: 0.03373400866985321 |
| Loss at step 400: 0.0403708778321743 |
| Loss at step 450: 0.034411486238241196 |
| Loss at step 500: 0.03582341596484184 |
| Loss at step 550: 0.03185276687145233 |
| Loss at step 600: 0.05909956619143486 |
| Loss at step 650: 0.03586168587207794 |
| Loss at step 700: 0.03290408104658127 |
| Loss at step 750: 0.035262443125247955 |
| Loss at step 800: 0.038654476404190063 |
| Loss at step 850: 0.03391736373305321 |
| Loss at step 900: 0.03784594684839249 |
| Mean training loss after epoch 254: 0.04157071075698079 |
|
|
| EPOCH: 255 |
| Loss at step 0: 0.036882199347019196 |
| Loss at step 50: 0.03993494436144829 |
| Loss at step 100: 0.04192814603447914 |
| Loss at step 150: 0.05400025099515915 |
| Loss at step 200: 0.036772482097148895 |
| Loss at step 250: 0.03589187189936638 |
| Loss at step 300: 0.041940346360206604 |
| Loss at step 350: 0.03448191285133362 |
| Loss at step 400: 0.029735693708062172 |
| Loss at step 450: 0.033656343817710876 |
| Loss at step 500: 0.055729810148477554 |
| Loss at step 550: 0.03558502718806267 |
| Loss at step 600: 0.03889984264969826 |
| Loss at step 650: 0.049456886947155 |
| Loss at step 700: 0.036641478538513184 |
| Loss at step 750: 0.048241570591926575 |
| Loss at step 800: 0.0453965850174427 |
| Loss at step 850: 0.03907885402441025 |
| Loss at step 900: 0.06074392423033714 |
| Mean training loss after epoch 255: 0.041012863235385305 |
|
|
| EPOCH: 256 |
| Loss at step 0: 0.03528229892253876 |
| Loss at step 50: 0.03847229480743408 |
| Loss at step 100: 0.04104534536600113 |
| Loss at step 150: 0.03357851132750511 |
| Loss at step 200: 0.04962961748242378 |
| Loss at step 250: 0.054125647991895676 |
| Loss at step 300: 0.0447879284620285 |
| Loss at step 350: 0.04007836803793907 |
| Loss at step 400: 0.04048490524291992 |
| Loss at step 450: 0.03496857360005379 |
| Loss at step 500: 0.05139027535915375 |
| Loss at step 550: 0.044430509209632874 |
| Loss at step 600: 0.049299247562885284 |
| Loss at step 650: 0.04760231077671051 |
| Loss at step 700: 0.05970483645796776 |
| Loss at step 750: 0.030023494735360146 |
| Loss at step 800: 0.051494043320417404 |
| Loss at step 850: 0.047272972762584686 |
| Loss at step 900: 0.0323207825422287 |
| Mean training loss after epoch 256: 0.041227073287531765 |
| /athenahomes/gabrijel/miniconda3/envs/track-generator/lib/python3.11/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: '/athenahomes/gabrijel/miniconda3/envs/track-generator/lib/python3.11/site-packages/torchvision/image.so: undefined symbol: _ZN3c1017RegisterOperatorsD1Ev'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source? |
| warn( |
| Schedule: cosine |
| Cfg: False |
| Output path: /scratch/shared/beegfs/gabrijel/m2l/mini |
| Patch Size: 2 |
| Device: cuda:3 |
| ===================================================================================== |
| Layer (type:depth-idx) Param # |
| ===================================================================================== |
| DiT 75,264 |
| ├─PatchEmbed: 1-1 -- |
| │ └─Conv2d: 2-1 1,920 |
| ├─TimestepEmbedder: 1-2 -- |
| │ └─Mlp: 2-2 -- |
| │ │ └─Linear: 3-1 98,688 |
| │ │ └─SiLU: 3-2 -- |
| │ │ └─Linear: 3-3 147,840 |
| ├─ModuleList: 1-3 -- |
| │ └─DiTBlock: 2-3 -- |
| │ │ └─LayerNorm: 3-4 -- |
| │ │ └─MultiheadAttention: 3-5 591,360 |
| │ │ └─LayerNorm: 3-6 -- |
| │ │ └─Mlp: 3-7 1,181,568 |
| │ │ └─Sequential: 3-8 887,040 |
| │ └─DiTBlock: 2-4 -- |
| │ │ └─LayerNorm: 3-9 -- |
| │ │ └─MultiheadAttention: 3-10 591,360 |
| │ │ └─LayerNorm: 3-11 -- |
| │ │ └─Mlp: 3-12 1,181,568 |
| │ │ └─Sequential: 3-13 887,040 |
| │ └─DiTBlock: 2-5 -- |
| │ │ └─LayerNorm: 3-14 -- |
| │ │ └─MultiheadAttention: 3-15 591,360 |
| │ │ └─LayerNorm: 3-16 -- |
| │ │ └─Mlp: 3-17 1,181,568 |
| │ │ └─Sequential: 3-18 887,040 |
| │ └─DiTBlock: 2-6 -- |
| │ │ └─LayerNorm: 3-19 -- |
| │ │ └─MultiheadAttention: 3-20 591,360 |
| │ │ └─LayerNorm: 3-21 -- |
| │ │ └─Mlp: 3-22 1,181,568 |
| │ │ └─Sequential: 3-23 887,040 |
| │ └─DiTBlock: 2-7 -- |
| │ │ └─LayerNorm: 3-24 -- |
| │ │ └─MultiheadAttention: 3-25 591,360 |
| │ │ └─LayerNorm: 3-26 -- |
| │ │ └─Mlp: 3-27 1,181,568 |
| │ │ └─Sequential: 3-28 887,040 |
| │ └─DiTBlock: 2-8 -- |
| │ │ └─LayerNorm: 3-29 -- |
| │ │ └─MultiheadAttention: 3-30 591,360 |
| │ │ └─LayerNorm: 3-31 -- |
| │ │ └─Mlp: 3-32 1,181,568 |
| │ │ └─Sequential: 3-33 887,040 |
| ├─FinalLayer: 1-4 -- |
| │ └─LayerNorm: 2-9 -- |
| │ └─Linear: 2-10 1,540 |
| │ └─Sequential: 2-11 -- |
| │ │ └─SiLU: 3-34 -- |
| │ │ └─Linear: 3-35 295,680 |
| ├─Unpatchify: 1-5 -- |
| ===================================================================================== |
| Total params: 16,580,740 |
| Trainable params: 16,505,476 |
| Non-trainable params: 75,264 |
| ===================================================================================== |
|
|
| EPOCH: 1 |
| Loss at step 0: 0.9927976131439209 |
| Loss at step 50: 0.2647687792778015 |
| Loss at step 100: 0.1879013627767563 |
| Loss at step 150: 0.12602929770946503 |
| Loss at step 200: 0.11745952069759369 |
| Loss at step 250: 0.10707128793001175 |
| Loss at step 300: 0.1432727724313736 |
| Loss at step 350: 0.11516840010881424 |
| Loss at step 400: 0.10792354494333267 |
| Loss at step 450: 0.09377487003803253 |
| Loss at step 500: 0.10157567262649536 |
| Loss at step 550: 0.10117540508508682 |
| Loss at step 600: 0.11454605311155319 |
| Loss at step 650: 0.09202694892883301 |
| Loss at step 700: 0.08804260939359665 |
| Loss at step 750: 0.08932791650295258 |
| Loss at step 800: 0.08639156073331833 |
| Loss at step 850: 0.09178578108549118 |
| Loss at step 900: 0.09026865661144257 |
| Mean training loss after epoch 1: 0.14250233578783617 |
|
|
| EPOCH: 2 |
| Loss at step 0: 0.08553174138069153 |
| Loss at step 50: 0.0918794572353363 |
| Loss at step 100: 0.07219964265823364 |
| Loss at step 150: 0.07668644934892654 |
| Loss at step 200: 0.08180388808250427 |
| Loss at step 250: 0.07124952226877213 |
| Loss at step 300: 0.08224574476480484 |
| Loss at step 350: 0.08902806788682938 |
| Loss at step 400: 0.0827031284570694 |
| Loss at step 450: 0.07146904617547989 |
| Loss at step 500: 0.07812539488077164 |
| Loss at step 550: 0.06646260619163513 |
| Loss at step 600: 0.06209966167807579 |
| Loss at step 650: 0.07814020663499832 |
| Loss at step 700: 0.058781880885362625 |
| Loss at step 750: 0.051821518689394 |
| Loss at step 800: 0.07093804329633713 |
| Loss at step 850: 0.0550219789147377 |
| Loss at step 900: 0.058788903057575226 |
| Mean training loss after epoch 2: 0.07401431160472603 |
|
|
| EPOCH: 3 |
| Loss at step 0: 0.06717473268508911 |
| Loss at step 50: 0.07093988358974457 |
| Loss at step 100: 0.06632817536592484 |
| Loss at step 150: 0.07580439746379852 |
| Loss at step 200: 0.06400690972805023 |
| Loss at step 250: 0.07392463833093643 |
| Loss at step 300: 0.048493415117263794 |
| Loss at step 350: 0.0506267324090004 |
| Loss at step 400: 0.07013332098722458 |
| Loss at step 450: 0.05782090872526169 |
| Loss at step 500: 0.07724333554506302 |
| Loss at step 550: 0.05080138146877289 |
| Loss at step 600: 0.06723678857088089 |
| Loss at step 650: 0.057337675243616104 |
| Loss at step 700: 0.05216258764266968 |
| Loss at step 750: 0.05217692628502846 |
| Loss at step 800: 0.05593561381101608 |
| Loss at step 850: 0.07336627691984177 |
| Loss at step 900: 0.05595633387565613 |
| Mean training loss after epoch 3: 0.06080897097219663 |
|
|
| EPOCH: 4 |
| Loss at step 0: 0.04946884885430336 |
| Loss at step 50: 0.054812829941511154 |
| Loss at step 100: 0.06162804737687111 |
| Loss at step 150: 0.05907006561756134 |
| Loss at step 200: 0.04258119687438011 |
| Loss at step 250: 0.06596523523330688 |
| Loss at step 300: 0.049682050943374634 |
| Loss at step 350: 0.05187582969665527 |
| Loss at step 400: 0.045258037745952606 |
| Loss at step 450: 0.050041262060403824 |
| Loss at step 500: 0.06738606095314026 |
| Loss at step 550: 0.057629767805337906 |
| Loss at step 600: 0.05861373245716095 |
| Loss at step 650: 0.054706621915102005 |
| Loss at step 700: 0.046074602752923965 |
| Loss at step 750: 0.04815784841775894 |
| Loss at step 800: 0.054592981934547424 |
| Loss at step 850: 0.06745585799217224 |
| Loss at step 900: 0.04226946830749512 |
| Mean training loss after epoch 4: 0.05705311590198006 |
|
|
| EPOCH: 5 |
| Loss at step 0: 0.04679422453045845 |
| Loss at step 50: 0.05274326354265213 |
| Loss at step 100: 0.060322683304548264 |
| Loss at step 150: 0.047847017645835876 |
| Loss at step 200: 0.050057556480169296 |
| Loss at step 250: 0.04592166468501091 |
| Loss at step 300: 0.06508708745241165 |
| Loss at step 350: 0.06251583248376846 |
| Loss at step 400: 0.06060163676738739 |
| Loss at step 450: 0.04539049416780472 |
| Loss at step 500: 0.04933956637978554 |
| Loss at step 550: 0.044366367161273956 |
| Loss at step 600: 0.05056780204176903 |
| Loss at step 650: 0.05007366091012955 |
| Loss at step 700: 0.04736810550093651 |
| Loss at step 750: 0.0458783358335495 |
| Loss at step 800: 0.061146657913923264 |
| Loss at step 850: 0.04192593693733215 |
| Loss at step 900: 0.05574965849518776 |
| Mean training loss after epoch 5: 0.054514974121377666 |
|
|
| EPOCH: 6 |
| Loss at step 0: 0.05552354082465172 |
| Loss at step 50: 0.06919590383768082 |
| Loss at step 100: 0.051523979753255844 |
| Loss at step 150: 0.07512925565242767 |
| Loss at step 200: 0.0469990149140358 |
| Loss at step 250: 0.04151985049247742 |
| Loss at step 300: 0.054267555475234985 |
| Loss at step 350: 0.046799663454294205 |
| Loss at step 400: 0.043437469750642776 |
| Loss at step 450: 0.0556630864739418 |
| Loss at step 500: 0.07886315882205963 |
| Loss at step 550: 0.07932259142398834 |
| Loss at step 600: 0.06433942168951035 |
| Loss at step 650: 0.05824227258563042 |
| Loss at step 700: 0.04445641115307808 |
| Loss at step 750: 0.05480702966451645 |
| Loss at step 800: 0.05086228623986244 |
| Loss at step 850: 0.050280384719371796 |
| Loss at step 900: 0.048203449696302414 |
| Mean training loss after epoch 6: 0.05360761121995668 |
|
|
| EPOCH: 7 |
| Loss at step 0: 0.05120495334267616 |
| Loss at step 50: 0.041711125522851944 |
| Loss at step 100: 0.03893940523266792 |
| Loss at step 150: 0.043815772980451584 |
| Loss at step 200: 0.04661528393626213 |
| Loss at step 250: 0.04535030573606491 |
| Loss at step 300: 0.052326902747154236 |
| Loss at step 350: 0.04767979308962822 |
| Loss at step 400: 0.05384628102183342 |
| Loss at step 450: 0.03882434964179993 |
| Loss at step 500: 0.05370200797915459 |
| Loss at step 550: 0.05347057804465294 |
| Loss at step 600: 0.04494878277182579 |
| Loss at step 650: 0.04447983577847481 |
| Loss at step 700: 0.04920825734734535 |
| Loss at step 750: 0.057055290788412094 |
| Loss at step 800: 0.10477184504270554 |
| Loss at step 850: 0.043160244822502136 |
| Loss at step 900: 0.04887056723237038 |
| Mean training loss after epoch 7: 0.05228752429599066 |
|
|
| EPOCH: 8 |
| Loss at step 0: 0.05011899024248123 |
| Loss at step 50: 0.06245705857872963 |
| Loss at step 100: 0.06631136685609818 |
| Loss at step 150: 0.04895845800638199 |
| Loss at step 200: 0.050268225371837616 |
| Loss at step 250: 0.04723553732037544 |
| Loss at step 300: 0.07536876201629639 |
| Loss at step 350: 0.06294748932123184 |
| Loss at step 400: 0.048326920717954636 |
| Loss at step 450: 0.04759557917714119 |
| Loss at step 500: 0.05410340428352356 |
| Loss at step 550: 0.05210983008146286 |
| Loss at step 600: 0.04399741441011429 |
| Loss at step 650: 0.041283536702394485 |
| Loss at step 700: 0.04357845336198807 |
| Loss at step 750: 0.060434866696596146 |
| Loss at step 800: 0.04585731029510498 |
| Loss at step 850: 0.04889417067170143 |
| Loss at step 900: 0.04579473286867142 |
| Mean training loss after epoch 8: 0.05117087174993334 |
|
|
| EPOCH: 9 |
| Loss at step 0: 0.047287292778491974 |
| Loss at step 50: 0.04318024218082428 |
| Loss at step 100: 0.046293068677186966 |
| Loss at step 150: 0.03605443984270096 |
| Loss at step 200: 0.045547377318143845 |
| Loss at step 250: 0.0459781251847744 |
| Loss at step 300: 0.04060017317533493 |
| Loss at step 350: 0.045362140983343124 |
| Loss at step 400: 0.04601544514298439 |
| Loss at step 450: 0.05350184813141823 |
| Loss at step 500: 0.04330907389521599 |
| Loss at step 550: 0.043803706765174866 |
| Loss at step 600: 0.049809280782938004 |
| Loss at step 650: 0.04990199953317642 |
| Loss at step 700: 0.05597102642059326 |
| Loss at step 750: 0.04491954296827316 |
| Loss at step 800: 0.03813697397708893 |
| Loss at step 850: 0.043534938246011734 |
| Loss at step 900: 0.052817512303590775 |
| Mean training loss after epoch 9: 0.050545577774010995 |
|
|
| EPOCH: 10 |
| Loss at step 0: 0.06345701217651367 |
| Loss at step 50: 0.04983758181333542 |
| Loss at step 100: 0.058084748685359955 |
| Loss at step 150: 0.057511843740940094 |
| Loss at step 200: 0.04950873926281929 |
| Loss at step 250: 0.044199153780937195 |
| Loss at step 300: 0.049990322440862656 |
| Loss at step 350: 0.042405661195516586 |
| Loss at step 400: 0.0410495363175869 |
| Loss at step 450: 0.050903402268886566 |
| Loss at step 500: 0.03998948261141777 |
| Loss at step 550: 0.04161235690116882 |
| Loss at step 600: 0.04452834278345108 |
| Loss at step 650: 0.04524111747741699 |
| Loss at step 700: 0.05646144971251488 |
| Loss at step 750: 0.039474230259656906 |
| Loss at step 800: 0.042933832854032516 |
| Loss at step 850: 0.04933589696884155 |
| Loss at step 900: 0.036208804696798325 |
| Mean training loss after epoch 10: 0.049290040768047515 |
|
|
| EPOCH: 11 |
| Loss at step 0: 0.03956316038966179 |
| Loss at step 50: 0.047969572246074677 |
| Loss at step 100: 0.06317726522684097 |
| Loss at step 150: 0.04703586921095848 |
| Loss at step 200: 0.04931914806365967 |
| Loss at step 250: 0.047175485640764236 |
| Loss at step 300: 0.04173384979367256 |
| Loss at step 350: 0.04106055572628975 |
| Loss at step 400: 0.04758727177977562 |
| Loss at step 450: 0.04685395956039429 |
| Loss at step 500: 0.06105189397931099 |
| Loss at step 550: 0.06041964516043663 |
| Loss at step 600: 0.055400021374225616 |
| Loss at step 650: 0.04655126482248306 |
| Loss at step 700: 0.06059831380844116 |
| Loss at step 750: 0.06167222931981087 |
| Loss at step 800: 0.042501598596572876 |
| Loss at step 850: 0.042257603257894516 |
| Loss at step 900: 0.045965805649757385 |
| Mean training loss after epoch 11: 0.04964323368654259 |
|
|
| EPOCH: 12 |
| Loss at step 0: 0.034203123301267624 |
| Loss at step 50: 0.05352773517370224 |
| Loss at step 100: 0.043338775634765625 |
| Loss at step 150: 0.04150288924574852 |
| Loss at step 200: 0.05577273294329643 |
| Loss at step 250: 0.055770426988601685 |
| Loss at step 300: 0.045309942215681076 |
| Loss at step 350: 0.04029889404773712 |
| Loss at step 400: 0.03722796589136124 |
| Loss at step 450: 0.04935304448008537 |
| Loss at step 500: 0.05218170955777168 |
| Loss at step 550: 0.061936669051647186 |
| Loss at step 600: 0.07341843843460083 |
| Loss at step 650: 0.0434122271835804 |
| Loss at step 700: 0.050429198890924454 |
| Loss at step 750: 0.054148439317941666 |
| Loss at step 800: 0.04618339240550995 |
| Loss at step 850: 0.039194535464048386 |
| Loss at step 900: 0.05453008413314819 |
| Mean training loss after epoch 12: 0.04853169766586345 |
|
|
| EPOCH: 13 |
| Loss at step 0: 0.05767318978905678 |
| Loss at step 50: 0.04307478666305542 |
| Loss at step 100: 0.039039019495248795 |
| Loss at step 150: 0.050793424248695374 |
| Loss at step 200: 0.05458158627152443 |
| Loss at step 250: 0.0410580076277256 |
| Loss at step 300: 0.03630630671977997 |
| Loss at step 350: 0.04801056906580925 |
| Loss at step 400: 0.038546375930309296 |
| Loss at step 450: 0.04437912628054619 |
| Loss at step 500: 0.04366833716630936 |
| Loss at step 550: 0.07518304139375687 |
| Loss at step 600: 0.04526819288730621 |
| Loss at step 650: 0.07137490063905716 |
| Loss at step 700: 0.056305669248104095 |
| Loss at step 750: 0.05747681111097336 |
| Loss at step 800: 0.05244934558868408 |
| Loss at step 850: 0.04294973239302635 |
| Loss at step 900: 0.03596752509474754 |
| Mean training loss after epoch 13: 0.048368280024718505 |
|
|
| EPOCH: 14 |
| Loss at step 0: 0.059222716838121414 |
| Loss at step 50: 0.04363219439983368 |
| Loss at step 100: 0.04555397853255272 |
| Loss at step 150: 0.03750060126185417 |
| Loss at step 200: 0.044108372181653976 |
| Loss at step 250: 0.05813005939126015 |
| Loss at step 300: 0.03639712557196617 |
| Loss at step 350: 0.07586781680583954 |
| Loss at step 400: 0.050042495131492615 |
| Loss at step 450: 0.054030727595090866 |
| Loss at step 500: 0.08736606687307358 |
| Loss at step 550: 0.0347326397895813 |
| Loss at step 600: 0.03835294768214226 |
| Loss at step 650: 0.04757542535662651 |
| Loss at step 700: 0.044310227036476135 |
| Loss at step 750: 0.03899923712015152 |
| Loss at step 800: 0.0389823392033577 |
| Loss at step 850: 0.046352993696928024 |
| Loss at step 900: 0.04369539022445679 |
| Mean training loss after epoch 14: 0.047218455241989095 |
|
|
| EPOCH: 15 |
| Loss at step 0: 0.061945103108882904 |
| Loss at step 50: 0.040272314101457596 |
| Loss at step 100: 0.05934322997927666 |
| Loss at step 150: 0.061598729342222214 |
| Loss at step 200: 0.03942662104964256 |
| Loss at step 250: 0.07209756970405579 |
| Loss at step 300: 0.056602660566568375 |
| Loss at step 350: 0.04495527222752571 |
| Loss at step 400: 0.054804615676403046 |
| Loss at step 450: 0.060904551297426224 |
| Loss at step 500: 0.052522361278533936 |
| Loss at step 550: 0.04878659546375275 |
| Loss at step 600: 0.04031701013445854 |
| Loss at step 650: 0.04928752779960632 |
| Loss at step 700: 0.04112618789076805 |
| Loss at step 750: 0.05811849236488342 |
| Loss at step 800: 0.05275645852088928 |
| Loss at step 850: 0.05110972002148628 |
| Loss at step 900: 0.05189656466245651 |
| Mean training loss after epoch 15: 0.04713583987420683 |
|
|
| EPOCH: 16 |
| Loss at step 0: 0.05751290172338486 |
| Loss at step 50: 0.05349951237440109 |
| Loss at step 100: 0.061340536922216415 |
| Loss at step 150: 0.04630905017256737 |
| Loss at step 200: 0.040518227964639664 |
| Loss at step 250: 0.04509662091732025 |
| Loss at step 300: 0.054649755358695984 |
| Loss at step 350: 0.04015738144516945 |
| Loss at step 400: 0.05614438280463219 |
| Loss at step 450: 0.04284175485372543 |
| Loss at step 500: 0.0447845458984375 |
| Loss at step 550: 0.04490595683455467 |
| Loss at step 600: 0.045310527086257935 |
| Loss at step 650: 0.04410380497574806 |
| Loss at step 700: 0.038343433290719986 |
| Loss at step 750: 0.05487282946705818 |
| Loss at step 800: 0.050881609320640564 |
| Loss at step 850: 0.05533342808485031 |
| Loss at step 900: 0.07102000713348389 |
| Mean training loss after epoch 16: 0.04699640631524802 |
|
|
| EPOCH: 17 |
| Loss at step 0: 0.0580948069691658 |
| Loss at step 50: 0.03580525517463684 |
| Loss at step 100: 0.03325394168496132 |
| Loss at step 150: 0.0388118177652359 |
| Loss at step 200: 0.039655230939388275 |
| Loss at step 250: 0.04357254132628441 |
| Loss at step 300: 0.0388103649020195 |
| Loss at step 350: 0.062123239040374756 |
| Loss at step 400: 0.0393681563436985 |
| Loss at step 450: 0.045537207275629044 |
| Loss at step 500: 0.04149201512336731 |
| Loss at step 550: 0.056296877562999725 |
| Loss at step 600: 0.041269417852163315 |
| Loss at step 650: 0.047324035316705704 |
| Loss at step 700: 0.04247691109776497 |
| Loss at step 750: 0.033218417316675186 |
| Loss at step 800: 0.03631206229329109 |
| Loss at step 850: 0.03608255833387375 |
| Loss at step 900: 0.04036899283528328 |
| Mean training loss after epoch 17: 0.047017275433201014 |
|
|
| EPOCH: 18 |
| Loss at step 0: 0.037257082760334015 |
| Loss at step 50: 0.03367447480559349 |
| Loss at step 100: 0.042098648846149445 |
| Loss at step 150: 0.039509519934654236 |
| Loss at step 200: 0.05941949412226677 |
| Loss at step 250: 0.038624610751867294 |
| Loss at step 300: 0.06093718111515045 |
| Loss at step 350: 0.044842202216386795 |
| Loss at step 400: 0.04626288637518883 |
| Loss at step 450: 0.05908516049385071 |
| Loss at step 500: 0.04012873023748398 |
| Loss at step 550: 0.038510892540216446 |
| Loss at step 600: 0.03817109763622284 |
| Loss at step 650: 0.04379972070455551 |
| Loss at step 700: 0.04390494152903557 |
| Loss at step 750: 0.03912144526839256 |
| Loss at step 800: 0.037638112902641296 |
| Loss at step 850: 0.04086960479617119 |
| Loss at step 900: 0.045219577848911285 |
| Mean training loss after epoch 18: 0.04582779359485485 |
|
|
| EPOCH: 19 |
| Loss at step 0: 0.041718464344739914 |
| Loss at step 50: 0.041183002293109894 |
| Loss at step 100: 0.04069666936993599 |
| Loss at step 150: 0.037516038864851 |
| Loss at step 200: 0.047058697789907455 |
| Loss at step 250: 0.04376176744699478 |
| Loss at step 300: 0.042468659579753876 |
| Loss at step 350: 0.04458988457918167 |
| Loss at step 400: 0.04017651826143265 |
| Loss at step 450: 0.07378578186035156 |
| Loss at step 500: 0.0400574654340744 |
| Loss at step 550: 0.039062704890966415 |
| Loss at step 600: 0.038819320499897 |
| Loss at step 650: 0.033067211508750916 |
| Loss at step 700: 0.042712289839982986 |
| Loss at step 750: 0.03872883319854736 |
| Loss at step 800: 0.05610249936580658 |
| Loss at step 850: 0.03668671473860741 |
| Loss at step 900: 0.043801337480545044 |
| Mean training loss after epoch 19: 0.04611589505410652 |
|
|
| EPOCH: 20 |
| Loss at step 0: 0.04827193170785904 |
| Loss at step 50: 0.05359834060072899 |
| Loss at step 100: 0.04653705283999443 |
| Loss at step 150: 0.05003025382757187 |
| Loss at step 200: 0.04673366993665695 |
| Loss at step 250: 0.040243230760097504 |
| Loss at step 300: 0.03828419744968414 |
| Loss at step 350: 0.04028535634279251 |
| Loss at step 400: 0.04643638804554939 |
| Loss at step 450: 0.03311799094080925 |
| Loss at step 500: 0.04406367987394333 |
| Loss at step 550: 0.03698456659913063 |
| Loss at step 600: 0.058640990406274796 |
| Loss at step 650: 0.040676768869161606 |
| Loss at step 700: 0.0424061119556427 |
| Loss at step 750: 0.05697501823306084 |
| Loss at step 800: 0.03824044018983841 |
| Loss at step 850: 0.066072478890419 |
| Loss at step 900: 0.041484538465738297 |
| Mean training loss after epoch 20: 0.04569254003004479 |
|
|
| EPOCH: 21 |
| Loss at step 0: 0.052745524793863297 |
| Loss at step 50: 0.04679976403713226 |
| Loss at step 100: 0.06415614485740662 |
| Loss at step 150: 0.03858758881688118 |
| Loss at step 200: 0.03995379060506821 |
| Loss at step 250: 0.04070864990353584 |
| Loss at step 300: 0.047886040061712265 |
| Loss at step 350: 0.04501399025321007 |
| Loss at step 400: 0.03356078267097473 |
| Loss at step 450: 0.03801573067903519 |
| Loss at step 500: 0.037307173013687134 |
| Loss at step 550: 0.04218755289912224 |
| Loss at step 600: 0.05996202677488327 |
| Loss at step 650: 0.04204526171088219 |
| Loss at step 700: 0.07033933699131012 |
| Loss at step 750: 0.03862342983484268 |
| Loss at step 800: 0.038705822080373764 |
| Loss at step 850: 0.03514530509710312 |
| Loss at step 900: 0.03651607781648636 |
| Mean training loss after epoch 21: 0.045421224570016995 |
|
|
| EPOCH: 22 |
| Loss at step 0: 0.05221446231007576 |
| Loss at step 50: 0.04250719025731087 |
| Loss at step 100: 0.05661725997924805 |
| Loss at step 150: 0.03511292487382889 |
| Loss at step 200: 0.040635790675878525 |
| Loss at step 250: 0.04838637635111809 |
| Loss at step 300: 0.0390804298222065 |
| Loss at step 350: 0.0534442774951458 |
| Loss at step 400: 0.038561638444662094 |
| Loss at step 450: 0.06379273533821106 |
| Loss at step 500: 0.035927664488554 |
| Loss at step 550: 0.06136680021882057 |
| Loss at step 600: 0.031272273510694504 |
| Loss at step 650: 0.045360028743743896 |
| Loss at step 700: 0.04633624106645584 |
| Loss at step 750: 0.04046519100666046 |
| Loss at step 800: 0.04161432385444641 |
| Loss at step 850: 0.042055003345012665 |
| Loss at step 900: 0.05448092147707939 |
| Mean training loss after epoch 22: 0.04483226262358651 |
|
|
| EPOCH: 23 |
| Loss at step 0: 0.04911420866847038 |
| Loss at step 50: 0.0415814071893692 |
| Loss at step 100: 0.04444144666194916 |
| Loss at step 150: 0.03530682623386383 |
| Loss at step 200: 0.0463300421833992 |
| Loss at step 250: 0.038280509412288666 |
| Loss at step 300: 0.05271584540605545 |
| Loss at step 350: 0.05282475799322128 |
| Loss at step 400: 0.04728353023529053 |
| Loss at step 450: 0.03833460435271263 |
| Loss at step 500: 0.057132475078105927 |
| Loss at step 550: 0.0409892238676548 |
| Loss at step 600: 0.040609076619148254 |
| Loss at step 650: 0.040882013738155365 |
| Loss at step 700: 0.04064009711146355 |
| Loss at step 750: 0.04380909353494644 |
| Loss at step 800: 0.03657805919647217 |
| Loss at step 850: 0.03418336808681488 |
| Loss at step 900: 0.034976694732904434 |
| Mean training loss after epoch 23: 0.04488808057630367 |
|
|
| EPOCH: 24 |
| Loss at step 0: 0.048248641192913055 |
| Loss at step 50: 0.05135813355445862 |
| Loss at step 100: 0.050695184618234634 |
| Loss at step 150: 0.05454166978597641 |
| Loss at step 200: 0.040228258818387985 |
| Loss at step 250: 0.045158132910728455 |
| Loss at step 300: 0.03014983795583248 |
| Loss at step 350: 0.030783720314502716 |
| Loss at step 400: 0.05381038039922714 |
| Loss at step 450: 0.03955896943807602 |
| Loss at step 500: 0.057066842913627625 |
| Loss at step 550: 0.036507830023765564 |
| Loss at step 600: 0.07247436791658401 |
| Loss at step 650: 0.0378834530711174 |
| Loss at step 700: 0.0714469701051712 |
| Loss at step 750: 0.040530942380428314 |
| Loss at step 800: 0.04141752049326897 |
| Loss at step 850: 0.034526366740465164 |
| Loss at step 900: 0.03569827228784561 |
| Mean training loss after epoch 24: 0.04518484828600497 |
|
|
| EPOCH: 25 |
| Loss at step 0: 0.03326995298266411 |
| Loss at step 50: 0.033842310309410095 |
| Loss at step 100: 0.03736729547381401 |
| Loss at step 150: 0.0668163076043129 |
| Loss at step 200: 0.03536819666624069 |
| Loss at step 250: 0.03721887618303299 |
| Loss at step 300: 0.0470084510743618 |
| Loss at step 350: 0.04588761925697327 |
| Loss at step 400: 0.0489131323993206 |
| Loss at step 450: 0.04456213489174843 |
| Loss at step 500: 0.03591672703623772 |
| Loss at step 550: 0.05648083612322807 |
| Loss at step 600: 0.04321033135056496 |
| Loss at step 650: 0.04505500942468643 |
| Loss at step 700: 0.03846661001443863 |
| Loss at step 750: 0.03818375989794731 |
| Loss at step 800: 0.036762554198503494 |
| Loss at step 850: 0.05255861207842827 |
| Loss at step 900: 0.04515066742897034 |
| Mean training loss after epoch 25: 0.044807881285656874 |
|
|
| EPOCH: 26 |
| Loss at step 0: 0.0373607873916626 |
| Loss at step 50: 0.043030742555856705 |
| Loss at step 100: 0.045759208500385284 |
| Loss at step 150: 0.04089202731847763 |
| Loss at step 200: 0.036131978034973145 |
| Loss at step 250: 0.05533413961529732 |
| Loss at step 300: 0.03699621185660362 |
| Loss at step 350: 0.03552429378032684 |
| Loss at step 400: 0.042007893323898315 |
| Loss at step 450: 0.04304102063179016 |
| Loss at step 500: 0.05209362506866455 |
| Loss at step 550: 0.06375249475240707 |
| Loss at step 600: 0.038298461586236954 |
| Loss at step 650: 0.04088416323065758 |
| Loss at step 700: 0.05234269052743912 |
| Loss at step 750: 0.053741395473480225 |
| Loss at step 800: 0.04723315313458443 |
| Loss at step 850: 0.057126112282276154 |
| Loss at step 900: 0.042302802205085754 |
| Mean training loss after epoch 26: 0.04456539857171492 |
|
|
| EPOCH: 27 |
| Loss at step 0: 0.04674964025616646 |
| Loss at step 50: 0.04263751581311226 |
| Loss at step 100: 0.03667943552136421 |
| Loss at step 150: 0.040494099259376526 |
| Loss at step 200: 0.0386606901884079 |
| Loss at step 250: 0.040493160486221313 |
| Loss at step 300: 0.03662142530083656 |
| Loss at step 350: 0.03603795915842056 |
| Loss at step 400: 0.037907931953668594 |
| Loss at step 450: 0.05081484094262123 |
| Loss at step 500: 0.0563945472240448 |
| Loss at step 550: 0.0526045560836792 |
| Loss at step 600: 0.06008012220263481 |
| Loss at step 650: 0.051519934087991714 |
| Loss at step 700: 0.05095212161540985 |
| Loss at step 750: 0.038178347051143646 |
| Loss at step 800: 0.04066329076886177 |
| Loss at step 850: 0.038567233830690384 |
| Loss at step 900: 0.052257291972637177 |
| Mean training loss after epoch 27: 0.04509363583584965 |
|
|
| EPOCH: 28 |
| Loss at step 0: 0.029913390055298805 |
| Loss at step 50: 0.03645753860473633 |
| Loss at step 100: 0.034360963851213455 |
| Loss at step 150: 0.03566978871822357 |
| Loss at step 200: 0.03904988244175911 |
| Loss at step 250: 0.03659699112176895 |
| Loss at step 300: 0.038922570645809174 |
| Loss at step 350: 0.033944662660360336 |
| Loss at step 400: 0.0383438803255558 |
| Loss at step 450: 0.05149287357926369 |
| Loss at step 500: 0.03719630837440491 |
| Loss at step 550: 0.03484205901622772 |
| Loss at step 600: 0.03765057399868965 |
| Loss at step 650: 0.06854096800088882 |
| Loss at step 700: 0.0372895784676075 |
| Loss at step 750: 0.044852498918771744 |
| Loss at step 800: 0.033244166523218155 |
| Loss at step 850: 0.04573490098118782 |
| Loss at step 900: 0.03735152631998062 |
| Mean training loss after epoch 28: 0.044933726687405286 |
|
|
| EPOCH: 29 |
| Loss at step 0: 0.029447682201862335 |
| Loss at step 50: 0.0398627370595932 |
| Loss at step 100: 0.053118184208869934 |
| Loss at step 150: 0.03270377963781357 |
| Loss at step 200: 0.04028083756566048 |
| Loss at step 250: 0.057774025946855545 |
| Loss at step 300: 0.04933173581957817 |
| Loss at step 350: 0.054515670984983444 |
| Loss at step 400: 0.05123170465230942 |
| Loss at step 450: 0.042456742376089096 |
| Loss at step 500: 0.040116313844919205 |
| Loss at step 550: 0.04087410494685173 |
| Loss at step 600: 0.03468229994177818 |
| Loss at step 650: 0.039522793143987656 |
| Loss at step 700: 0.03545898199081421 |
| Loss at step 750: 0.04619523510336876 |
| Loss at step 800: 0.05729640647768974 |
| Loss at step 850: 0.051194336265325546 |
| Loss at step 900: 0.038833290338516235 |
| Mean training loss after epoch 29: 0.04402439670760367 |
|
|
| EPOCH: 30 |
| Loss at step 0: 0.04906666278839111 |
| Loss at step 50: 0.04361962527036667 |
| Loss at step 100: 0.04044247418642044 |
| Loss at step 150: 0.05557110905647278 |
| Loss at step 200: 0.03803514316678047 |
| Loss at step 250: 0.04414329677820206 |
| Loss at step 300: 0.04065090790390968 |
| Loss at step 350: 0.03852323070168495 |
| Loss at step 400: 0.03416415676474571 |
| Loss at step 450: 0.05052315816283226 |
| Loss at step 500: 0.056460876017808914 |
| Loss at step 550: 0.041875824332237244 |
| Loss at step 600: 0.039723336696624756 |
| Loss at step 650: 0.03889099508523941 |
| Loss at step 700: 0.0717279314994812 |
| Loss at step 750: 0.05256667360663414 |
| Loss at step 800: 0.04998709261417389 |
| Loss at step 850: 0.03653858229517937 |
| Loss at step 900: 0.06387999653816223 |
| Mean training loss after epoch 30: 0.0444086283079978 |
|
|
| EPOCH: 31 |
| Loss at step 0: 0.06570300459861755 |
| Loss at step 50: 0.03289508447051048 |
| Loss at step 100: 0.033931732177734375 |
| Loss at step 150: 0.05701424181461334 |
| Loss at step 200: 0.0524381622672081 |
| Loss at step 250: 0.04885663837194443 |
| Loss at step 300: 0.04105885699391365 |
| Loss at step 350: 0.03617656230926514 |
| Loss at step 400: 0.07422742247581482 |
| Loss at step 450: 0.04127587378025055 |
| Loss at step 500: 0.04956771805882454 |
| Loss at step 550: 0.032986026257276535 |
| Loss at step 600: 0.039893150329589844 |
| Loss at step 650: 0.03447552025318146 |
| Loss at step 700: 0.039409857243299484 |
| Loss at step 750: 0.051296427845954895 |
| Loss at step 800: 0.043854985386133194 |
| Loss at step 850: 0.058668699115514755 |
| Loss at step 900: 0.037739936262369156 |
| Mean training loss after epoch 31: 0.04468885555998412 |
|
|
| EPOCH: 32 |
| Loss at step 0: 0.04160969331860542 |
| Loss at step 50: 0.03463306650519371 |
| Loss at step 100: 0.035780616104602814 |
| Loss at step 150: 0.05247391015291214 |
| Loss at step 200: 0.043705083429813385 |
| Loss at step 250: 0.0600837767124176 |
| Loss at step 300: 0.037433646619319916 |
| Loss at step 350: 0.05155890807509422 |
| Loss at step 400: 0.07468236237764359 |
| Loss at step 450: 0.06288442015647888 |
| Loss at step 500: 0.04347878322005272 |
| Loss at step 550: 0.04345081374049187 |
| Loss at step 600: 0.04002153500914574 |
| Loss at step 650: 0.030901411548256874 |
| Loss at step 700: 0.03812825679779053 |
| Loss at step 750: 0.039223719388246536 |
| Loss at step 800: 0.03850940614938736 |
| Loss at step 850: 0.04800470918416977 |
| Loss at step 900: 0.03695946931838989 |
| Mean training loss after epoch 32: 0.04362768248152504 |
|
|
| EPOCH: 33 |
| Loss at step 0: 0.07225683331489563 |
| Loss at step 50: 0.039506975561380386 |
| Loss at step 100: 0.04103633388876915 |
| Loss at step 150: 0.036195915192365646 |
| Loss at step 200: 0.034182168543338776 |
| Loss at step 250: 0.052266839891672134 |
| Loss at step 300: 0.0427260585129261 |
| Loss at step 350: 0.039806365966796875 |
| Loss at step 400: 0.046187058091163635 |
| Loss at step 450: 0.035503558814525604 |
| Loss at step 500: 0.0356760174036026 |
| Loss at step 550: 0.03603162616491318 |
| Loss at step 600: 0.0453050471842289 |
| Loss at step 650: 0.04466850683093071 |
| Loss at step 700: 0.03671862930059433 |
| Loss at step 750: 0.032369568943977356 |
| Loss at step 800: 0.039522137492895126 |
| Loss at step 850: 0.05315655469894409 |
| Loss at step 900: 0.034537188708782196 |
| Mean training loss after epoch 33: 0.04295502561551612 |
|
|
| EPOCH: 34 |
| Loss at step 0: 0.029321929439902306 |
| Loss at step 50: 0.034134674817323685 |
| Loss at step 100: 0.04984445869922638 |
| Loss at step 150: 0.035927992314100266 |
| Loss at step 200: 0.04887424409389496 |
| Loss at step 250: 0.037316031754016876 |
| Loss at step 300: 0.04465991631150246 |
| Loss at step 350: 0.05477762967348099 |
| Loss at step 400: 0.053357016295194626 |
| Loss at step 450: 0.050971269607543945 |
| Loss at step 500: 0.0499236024916172 |
| Loss at step 550: 0.04749893397092819 |
| Loss at step 600: 0.04637804254889488 |
| Loss at step 650: 0.04684502258896828 |
| Loss at step 700: 0.06652657687664032 |
| Loss at step 750: 0.043038755655288696 |
| Loss at step 800: 0.03603661060333252 |
| Loss at step 850: 0.04426427185535431 |
| Loss at step 900: 0.05061193183064461 |
| Mean training loss after epoch 34: 0.04389402393037195 |
|
|
| EPOCH: 35 |
| Loss at step 0: 0.052734531462192535 |
| Loss at step 50: 0.07013734430074692 |
| Loss at step 100: 0.055571962147951126 |
| Loss at step 150: 0.03865116834640503 |
| Loss at step 200: 0.0372077152132988 |
| Loss at step 250: 0.04977886378765106 |
| Loss at step 300: 0.04531254991889 |
| Loss at step 350: 0.05584704130887985 |
| Loss at step 400: 0.0377432182431221 |
| Loss at step 450: 0.04084886610507965 |
| Loss at step 500: 0.07135910540819168 |
| Loss at step 550: 0.03363371267914772 |
| Loss at step 600: 0.03212135657668114 |
| Loss at step 650: 0.03952178731560707 |
| Loss at step 700: 0.041552383452653885 |
| Loss at step 750: 0.03404490277171135 |
| Loss at step 800: 0.03914336860179901 |
| Loss at step 850: 0.043152667582035065 |
| Loss at step 900: 0.03232409805059433 |
| Mean training loss after epoch 35: 0.04302508351819983 |
|
|
| EPOCH: 36 |
| Loss at step 0: 0.04851457104086876 |
| Loss at step 50: 0.04562768712639809 |
| Loss at step 100: 0.03531981259584427 |
| Loss at step 150: 0.043214645236730576 |
| Loss at step 200: 0.03744692727923393 |
| Loss at step 250: 0.042286891490221024 |
| Loss at step 300: 0.049430038779973984 |
| Loss at step 350: 0.036436405032873154 |
| Loss at step 400: 0.034777041524648666 |
| Loss at step 450: 0.04153888300061226 |
| Loss at step 500: 0.03553176298737526 |
| Loss at step 550: 0.031961582601070404 |
| Loss at step 600: 0.05224454030394554 |
| Loss at step 650: 0.04388105496764183 |
| Loss at step 700: 0.040361806750297546 |
| Loss at step 750: 0.040422216057777405 |
| Loss at step 800: 0.03775598108768463 |
| Loss at step 850: 0.05734693259000778 |
| Loss at step 900: 0.04077355936169624 |
| Mean training loss after epoch 36: 0.04312853543743142 |
|
|
| EPOCH: 37 |
| Loss at step 0: 0.03714748099446297 |
| Loss at step 50: 0.050660282373428345 |
| Loss at step 100: 0.03697645664215088 |
| Loss at step 150: 0.0827973410487175 |
| Loss at step 200: 0.04317590221762657 |
| Loss at step 250: 0.03618314117193222 |
| Loss at step 300: 0.037134528160095215 |
| Loss at step 350: 0.03909546509385109 |
| Loss at step 400: 0.038076251745224 |
| Loss at step 450: 0.04257143288850784 |
| Loss at step 500: 0.03321288153529167 |
| Loss at step 550: 0.03748180717229843 |
| Loss at step 600: 0.06636057794094086 |
| Loss at step 650: 0.03770602494478226 |
| Loss at step 700: 0.08009560406208038 |
| Loss at step 750: 0.03693768009543419 |
| Loss at step 800: 0.03650718182325363 |
| Loss at step 850: 0.04664753004908562 |
| Loss at step 900: 0.03936724364757538 |
| Mean training loss after epoch 37: 0.04336765655941928 |
|
|
| EPOCH: 38 |
| Loss at step 0: 0.03465455397963524 |
| Loss at step 50: 0.030911577865481377 |
| Loss at step 100: 0.05364152789115906 |
| Loss at step 150: 0.03758525103330612 |
| Loss at step 200: 0.03997034952044487 |
| Loss at step 250: 0.0362134613096714 |
| Loss at step 300: 0.03510035201907158 |
| Loss at step 350: 0.03568132966756821 |
| Loss at step 400: 0.03927675634622574 |
| Loss at step 450: 0.043984606862068176 |
| Loss at step 500: 0.044523004442453384 |
| Loss at step 550: 0.036992426961660385 |
| Loss at step 600: 0.0432291254401207 |
| Loss at step 650: 0.04972879961133003 |
| Loss at step 700: 0.040408678352832794 |
| Loss at step 750: 0.03769862279295921 |
| Loss at step 800: 0.03676699846982956 |
| Loss at step 850: 0.03363250568509102 |
| Loss at step 900: 0.050196655094623566 |
| Mean training loss after epoch 38: 0.043665299007395054 |
|
|
| EPOCH: 39 |
| Loss at step 0: 0.04297603666782379 |
| Loss at step 50: 0.03462286666035652 |
| Loss at step 100: 0.03346511349081993 |
| Loss at step 150: 0.04194847494363785 |
| Loss at step 200: 0.036040548235177994 |
| Loss at step 250: 0.03550048545002937 |
| Loss at step 300: 0.04542850703001022 |
| Loss at step 350: 0.05183808133006096 |
| Loss at step 400: 0.03676900267601013 |
| Loss at step 450: 0.049038663506507874 |
| Loss at step 500: 0.03395945951342583 |
| Loss at step 550: 0.0338163748383522 |
| Loss at step 600: 0.033632803708314896 |
| Loss at step 650: 0.03675498068332672 |
| Loss at step 700: 0.041949450969696045 |
| Loss at step 750: 0.04227084293961525 |
| Loss at step 800: 0.04024386778473854 |
| Loss at step 850: 0.03923187032341957 |
| Loss at step 900: 0.034244369715452194 |
| Mean training loss after epoch 39: 0.04297231583754772 |
|
|
| EPOCH: 40 |
| Loss at step 0: 0.030180353671312332 |
| Loss at step 50: 0.049267880618572235 |
| Loss at step 100: 0.03916803002357483 |
| Loss at step 150: 0.056771960109472275 |
| Loss at step 200: 0.039305031299591064 |
| Loss at step 250: 0.039471086114645004 |
| Loss at step 300: 0.03026297129690647 |
| Loss at step 350: 0.0378076434135437 |
| Loss at step 400: 0.055273085832595825 |
| Loss at step 450: 0.04198889806866646 |
| Loss at step 500: 0.05805264040827751 |
| Loss at step 550: 0.04977709427475929 |
| Loss at step 600: 0.0806681215763092 |
| Loss at step 650: 0.08839074522256851 |
| Loss at step 700: 0.041415825486183167 |
| Loss at step 750: 0.0352671816945076 |
| Loss at step 800: 0.03809533268213272 |
| Loss at step 850: 0.03589823096990585 |
| Loss at step 900: 0.03772462159395218 |
| Mean training loss after epoch 40: 0.04321250221781385 |
|
|
| EPOCH: 41 |
| Loss at step 0: 0.05648473650217056 |
| Loss at step 50: 0.03883674740791321 |
| Loss at step 100: 0.03668802231550217 |
| Loss at step 150: 0.03581535816192627 |
| Loss at step 200: 0.05020207539200783 |
| Loss at step 250: 0.03667880967259407 |
| Loss at step 300: 0.042587947100400925 |
| Loss at step 350: 0.03846283629536629 |
| Loss at step 400: 0.036278266459703445 |
| Loss at step 450: 0.039367277175188065 |
| Loss at step 500: 0.03347210958600044 |
| Loss at step 550: 0.042285192757844925 |
| Loss at step 600: 0.03368893265724182 |
| Loss at step 650: 0.03954387828707695 |
| Loss at step 700: 0.04458700865507126 |
| Loss at step 750: 0.04066156595945358 |
| Loss at step 800: 0.036489538848400116 |
| Loss at step 850: 0.03715793415904045 |
| Loss at step 900: 0.06632266938686371 |
| Mean training loss after epoch 41: 0.042843369641966784 |
|
|
| EPOCH: 42 |
| Loss at step 0: 0.049074843525886536 |
| Loss at step 50: 0.031039627268910408 |
| Loss at step 100: 0.0409608855843544 |
| Loss at step 150: 0.0379287414252758 |
| Loss at step 200: 0.03141274303197861 |
| Loss at step 250: 0.03718740865588188 |
| Loss at step 300: 0.04370876029133797 |
| Loss at step 350: 0.02973066456615925 |
| Loss at step 400: 0.03484489768743515 |
| Loss at step 450: 0.05998479947447777 |
| Loss at step 500: 0.04298393055796623 |
| Loss at step 550: 0.0414208248257637 |
| Loss at step 600: 0.05147707462310791 |
| Loss at step 650: 0.06301922351121902 |
| Loss at step 700: 0.04647999256849289 |
| Loss at step 750: 0.0380011722445488 |
| Loss at step 800: 0.03523740544915199 |
| Loss at step 850: 0.07007824629545212 |
| Loss at step 900: 0.04118829593062401 |
| Mean training loss after epoch 42: 0.042861232980847486 |
|
|
| EPOCH: 43 |
| Loss at step 0: 0.028600996360182762 |
| Loss at step 50: 0.03958134725689888 |
| Loss at step 100: 0.0379943922162056 |
| Loss at step 150: 0.0403524711728096 |
| Loss at step 200: 0.038039691746234894 |
| Loss at step 250: 0.05107349529862404 |
| Loss at step 300: 0.04073113203048706 |
| Loss at step 350: 0.043610718101263046 |
| Loss at step 400: 0.07114291191101074 |
| Loss at step 450: 0.04253702610731125 |
| Loss at step 500: 0.05893601104617119 |
| Loss at step 550: 0.0509442500770092 |
| Loss at step 600: 0.044168759137392044 |
| Loss at step 650: 0.054990626871585846 |
| Loss at step 700: 0.042283836752176285 |
| Loss at step 750: 0.03912976384162903 |
| Loss at step 800: 0.051697466522455215 |
| Loss at step 850: 0.05673801526427269 |
| Loss at step 900: 0.035775937139987946 |
| Mean training loss after epoch 43: 0.04222319512836524 |
|
|
| EPOCH: 44 |
| Loss at step 0: 0.037387020885944366 |
| Loss at step 50: 0.03670916706323624 |
| Loss at step 100: 0.03850468993186951 |
| Loss at step 150: 0.04527543485164642 |
| Loss at step 200: 0.04270525649189949 |
| Loss at step 250: 0.0406525693833828 |
| Loss at step 300: 0.0392395555973053 |
| Loss at step 350: 0.03703998774290085 |
| Loss at step 400: 0.04435901716351509 |
| Loss at step 450: 0.030216777697205544 |
| Loss at step 500: 0.03501291945576668 |
| Loss at step 550: 0.05483980476856232 |
| Loss at step 600: 0.03645576909184456 |
| Loss at step 650: 0.03959187492728233 |
| Loss at step 700: 0.035912465304136276 |
| Loss at step 750: 0.033680666238069534 |
| Loss at step 800: 0.032813236117362976 |
| Loss at step 850: 0.03100864589214325 |
| Loss at step 900: 0.038795165717601776 |
| Mean training loss after epoch 44: 0.04228815817431029 |
|
|
| EPOCH: 45 |
| Loss at step 0: 0.05061938613653183 |
| Loss at step 50: 0.06159723922610283 |
| Loss at step 100: 0.04928106442093849 |
| Loss at step 150: 0.04391185939311981 |
| Loss at step 200: 0.03841090202331543 |
| Loss at step 250: 0.03663986548781395 |
| Loss at step 300: 0.033025361597537994 |
| Loss at step 350: 0.05331796035170555 |
| Loss at step 400: 0.03757380694150925 |
| Loss at step 450: 0.0336463637650013 |
| Loss at step 500: 0.03897778317332268 |
| Loss at step 550: 0.03989659994840622 |
| Loss at step 600: 0.038776081055402756 |
| Loss at step 650: 0.05733334645628929 |
| Loss at step 700: 0.03490510582923889 |
| Loss at step 750: 0.039473410695791245 |
| Loss at step 800: 0.05830595642328262 |
| Loss at step 850: 0.05670051649212837 |
| Loss at step 900: 0.040327806025743484 |
| Mean training loss after epoch 45: 0.042787098425077094 |
|
|
| EPOCH: 46 |
| Loss at step 0: 0.036934878677129745 |
| Loss at step 50: 0.033364858478307724 |
| Loss at step 100: 0.03347126394510269 |
| Loss at step 150: 0.041521187871694565 |
| Loss at step 200: 0.0378275141119957 |
| Loss at step 250: 0.03971440717577934 |
| Loss at step 300: 0.03849451243877411 |
| Loss at step 350: 0.030927833169698715 |
| Loss at step 400: 0.03408275172114372 |
| Loss at step 450: 0.04135040193796158 |
| Loss at step 500: 0.03571862354874611 |
| Loss at step 550: 0.039906665682792664 |
| Loss at step 600: 0.03439576178789139 |
| Loss at step 650: 0.052114088088274 |
| Loss at step 700: 0.04131833836436272 |
| Loss at step 750: 0.03150162473320961 |
| Loss at step 800: 0.053609561175107956 |
| Loss at step 850: 0.07118377834558487 |
| Loss at step 900: 0.038397371768951416 |
| Mean training loss after epoch 46: 0.0424582164174617 |
|
|
| EPOCH: 47 |
| Loss at step 0: 0.0376482829451561 |
| Loss at step 50: 0.033311713486909866 |
| Loss at step 100: 0.046062029898166656 |
| Loss at step 150: 0.07413475960493088 |
| Loss at step 200: 0.06404237449169159 |
| Loss at step 250: 0.04188799113035202 |
| Loss at step 300: 0.044554203748703 |
| Loss at step 350: 0.03766021877527237 |
| Loss at step 400: 0.05552181974053383 |
| Loss at step 450: 0.037229984998703 |
| Loss at step 500: 0.03611749783158302 |
| Loss at step 550: 0.03990129381418228 |
| Loss at step 600: 0.04216839373111725 |
| Loss at step 650: 0.0565626285970211 |
| Loss at step 700: 0.05230112373828888 |
| Loss at step 750: 0.03784063085913658 |
| Loss at step 800: 0.03894263133406639 |
| Loss at step 850: 0.033128369599580765 |
| Loss at step 900: 0.030786553397774696 |
| Mean training loss after epoch 47: 0.042213420318102025 |
|
|
| EPOCH: 48 |
| Loss at step 0: 0.06113245338201523 |
| Loss at step 50: 0.038732144981622696 |
| Loss at step 100: 0.04146707057952881 |
| Loss at step 150: 0.05363892391324043 |
| Loss at step 200: 0.04317339509725571 |
| Loss at step 250: 0.04006329923868179 |
| Loss at step 300: 0.038389842957258224 |
| Loss at step 350: 0.04166923090815544 |
| Loss at step 400: 0.050251998007297516 |
| Loss at step 450: 0.03234206885099411 |
| Loss at step 500: 0.05430510640144348 |
| Loss at step 550: 0.05468246340751648 |
| Loss at step 600: 0.03641614317893982 |
| Loss at step 650: 0.0358758419752121 |
| Loss at step 700: 0.04045901075005531 |
| Loss at step 750: 0.02902292087674141 |
| Loss at step 800: 0.037241075187921524 |
| Loss at step 850: 0.0750131905078888 |
| Loss at step 900: 0.0332866832613945 |
| Mean training loss after epoch 48: 0.04252630450538417 |
|
|
| EPOCH: 49 |
| Loss at step 0: 0.04021422937512398 |
| Loss at step 50: 0.03641494736075401 |
| Loss at step 100: 0.038768261671066284 |
| Loss at step 150: 0.04296419024467468 |
| Loss at step 200: 0.036955609917640686 |
| Loss at step 250: 0.040370337665081024 |
| Loss at step 300: 0.03228134289383888 |
| Loss at step 350: 0.048863768577575684 |
| Loss at step 400: 0.03823704645037651 |
| Loss at step 450: 0.05663494020700455 |
| Loss at step 500: 0.03468574583530426 |
| Loss at step 550: 0.04946057125926018 |
| Loss at step 600: 0.04328424111008644 |
| Loss at step 650: 0.0349988117814064 |
| Loss at step 700: 0.04406754672527313 |
| Loss at step 750: 0.037463266402482986 |
| Loss at step 800: 0.04004407674074173 |
| Loss at step 850: 0.03687194362282753 |
| Loss at step 900: 0.03602185472846031 |
| Mean training loss after epoch 49: 0.042539564672230024 |
|
|
| EPOCH: 50 |
| Loss at step 0: 0.03601893410086632 |
| Loss at step 50: 0.03488951548933983 |
| Loss at step 100: 0.04303381219506264 |
| Loss at step 150: 0.03584350645542145 |
| Loss at step 200: 0.03999720513820648 |
| Loss at step 250: 0.037075553089380264 |
| Loss at step 300: 0.04014355316758156 |
| Loss at step 350: 0.03799928352236748 |
| Loss at step 400: 0.03307428956031799 |
| Loss at step 450: 0.03780652582645416 |
| Loss at step 500: 0.04645475372672081 |
| Loss at step 550: 0.035657599568367004 |
| Loss at step 600: 0.0703330934047699 |
| Loss at step 650: 0.0441756397485733 |
| Loss at step 700: 0.03868475556373596 |
| Loss at step 750: 0.036150313913822174 |
| Loss at step 800: 0.05835910141468048 |
| Loss at step 850: 0.035950593650341034 |
| Loss at step 900: 0.033544495701789856 |
| Mean training loss after epoch 50: 0.04233735294412893 |
|
|
| EPOCH: 51 |
| Loss at step 0: 0.03536150977015495 |
| Loss at step 50: 0.05265863612294197 |
| Loss at step 100: 0.053060974925756454 |
| Loss at step 150: 0.03569833189249039 |
| Loss at step 200: 0.03769461810588837 |
| Loss at step 250: 0.03149081766605377 |
| Loss at step 300: 0.03774293139576912 |
| Loss at step 350: 0.03972206637263298 |
| Loss at step 400: 0.03773874789476395 |
| Loss at step 450: 0.040503572672605515 |
| Loss at step 500: 0.04576042294502258 |
| Loss at step 550: 0.0350419245660305 |
| Loss at step 600: 0.04863131046295166 |
| Loss at step 650: 0.05377759039402008 |
| Loss at step 700: 0.03654267638921738 |
| Loss at step 750: 0.032249435782432556 |
| Loss at step 800: 0.04548071324825287 |
| Loss at step 850: 0.05506904795765877 |
| Loss at step 900: 0.03483176603913307 |
| Mean training loss after epoch 51: 0.04182767730031504 |
|
|
| EPOCH: 52 |
| Loss at step 0: 0.035643722862005234 |
| Loss at step 50: 0.035993482917547226 |
| Loss at step 100: 0.039494823664426804 |
| Loss at step 150: 0.03567099943757057 |
| Loss at step 200: 0.0390964038670063 |
| Loss at step 250: 0.035363804548978806 |
| Loss at step 300: 0.04085114598274231 |
| Loss at step 350: 0.03391866385936737 |
| Loss at step 400: 0.048653654754161835 |
| Loss at step 450: 0.03442519158124924 |
| Loss at step 500: 0.035034891217947006 |
| Loss at step 550: 0.05182870849967003 |
| Loss at step 600: 0.038956109434366226 |
| Loss at step 650: 0.04642537236213684 |
| Loss at step 700: 0.03629075363278389 |
| Loss at step 750: 0.03320210427045822 |
| Loss at step 800: 0.03982256352901459 |
| Loss at step 850: 0.05380026251077652 |
| Loss at step 900: 0.04151134565472603 |
| Mean training loss after epoch 52: 0.041800402694626024 |
|
|
| EPOCH: 53 |
| Loss at step 0: 0.05759120360016823 |
| Loss at step 50: 0.0414610281586647 |
| Loss at step 100: 0.03528021648526192 |
| Loss at step 150: 0.058663200587034225 |
| Loss at step 200: 0.03417353704571724 |
| Loss at step 250: 0.05214337259531021 |
| Loss at step 300: 0.030978785827755928 |
| Loss at step 350: 0.03184428811073303 |
| Loss at step 400: 0.04949423298239708 |
| Loss at step 450: 0.046923208981752396 |
| Loss at step 500: 0.04200230911374092 |
| Loss at step 550: 0.05963723734021187 |
| Loss at step 600: 0.048732444643974304 |
| Loss at step 650: 0.03525372967123985 |
| Loss at step 700: 0.042062558233737946 |
| Loss at step 750: 0.0699642226099968 |
| Loss at step 800: 0.03157135844230652 |
| Loss at step 850: 0.03461334854364395 |
| Loss at step 900: 0.03986768424510956 |
| Mean training loss after epoch 53: 0.04206207239710446 |
|
|
| EPOCH: 54 |
| Loss at step 0: 0.043047960847616196 |
| Loss at step 50: 0.050478626042604446 |
| Loss at step 100: 0.03833431750535965 |
| Loss at step 150: 0.037333905696868896 |
| Loss at step 200: 0.07576791942119598 |
| Loss at step 250: 0.041819144040346146 |
| Loss at step 300: 0.05899069085717201 |
| Loss at step 350: 0.038387831300497055 |
| Loss at step 400: 0.04553873464465141 |
| Loss at step 450: 0.03330902010202408 |
| Loss at step 500: 0.03797665983438492 |
| Loss at step 550: 0.05493336543440819 |
| Loss at step 600: 0.038320817053318024 |
| Loss at step 650: 0.03387630730867386 |
| Loss at step 700: 0.03648149594664574 |
| Loss at step 750: 0.05657254159450531 |
| Loss at step 800: 0.0527743361890316 |
| Loss at step 850: 0.031990937888622284 |
| Loss at step 900: 0.05744639039039612 |
| Mean training loss after epoch 54: 0.042270664306385305 |
|
|
| EPOCH: 55 |
| Loss at step 0: 0.04634654521942139 |
| Loss at step 50: 0.038860127329826355 |
| Loss at step 100: 0.0369553305208683 |
| Loss at step 150: 0.04601672664284706 |
| Loss at step 200: 0.026475802063941956 |
| Loss at step 250: 0.040818262845277786 |
| Loss at step 300: 0.050957221537828445 |
| Loss at step 350: 0.05213375389575958 |
| Loss at step 400: 0.03895549848675728 |
| Loss at step 450: 0.039853379130363464 |
| Loss at step 500: 0.0456225723028183 |
| Loss at step 550: 0.03201737627387047 |
| Loss at step 600: 0.04900722950696945 |
| Loss at step 650: 0.033515963703393936 |
| Loss at step 700: 0.045205309987068176 |
| Loss at step 750: 0.05664574354887009 |
| Loss at step 800: 0.05195131152868271 |
| Loss at step 850: 0.03943021968007088 |
| Loss at step 900: 0.051934707909822464 |
| Mean training loss after epoch 55: 0.041760075203915524 |
|
|
| EPOCH: 56 |
| Loss at step 0: 0.035669680684804916 |
| Loss at step 50: 0.03731066361069679 |
| Loss at step 100: 0.03839431330561638 |
| Loss at step 150: 0.04437437653541565 |
| Loss at step 200: 0.03604341298341751 |
| Loss at step 250: 0.03787315636873245 |
| Loss at step 300: 0.04129105433821678 |
| Loss at step 350: 0.0681724026799202 |
| Loss at step 400: 0.037422459572553635 |
| Loss at step 450: 0.04147394374012947 |
| Loss at step 500: 0.030557721853256226 |
| Loss at step 550: 0.05729537457227707 |
| Loss at step 600: 0.044497594237327576 |
| Loss at step 650: 0.04281434789299965 |
| Loss at step 700: 0.03463006392121315 |
| Loss at step 750: 0.03795117139816284 |
| Loss at step 800: 0.03903064504265785 |
| Loss at step 850: 0.05226927250623703 |
| Loss at step 900: 0.044425107538700104 |
| Mean training loss after epoch 56: 0.04198346188518284 |
|
|
| EPOCH: 57 |
| Loss at step 0: 0.040313441306352615 |
| Loss at step 50: 0.04098132625222206 |
| Loss at step 100: 0.037053175270557404 |
| Loss at step 150: 0.03720230981707573 |
| Loss at step 200: 0.03237070143222809 |
| Loss at step 250: 0.042227283120155334 |
| Loss at step 300: 0.05192343145608902 |
| Loss at step 350: 0.03451365604996681 |
| Loss at step 400: 0.03476078435778618 |
| Loss at step 450: 0.03759030997753143 |
| Loss at step 500: 0.03982206806540489 |
| Loss at step 550: 0.04919624701142311 |
| Loss at step 600: 0.028733564540743828 |
| Loss at step 650: 0.036303646862506866 |
| Loss at step 700: 0.040104515850543976 |
| Loss at step 750: 0.040087904781103134 |
| Loss at step 800: 0.0365641824901104 |
| Loss at step 850: 0.03147405758500099 |
| Loss at step 900: 0.04074550420045853 |
| Mean training loss after epoch 57: 0.0418718279198384 |
|
|
| EPOCH: 58 |
| Loss at step 0: 0.027126234024763107 |
| Loss at step 50: 0.03955811262130737 |
| Loss at step 100: 0.03852409869432449 |
| Loss at step 150: 0.04588547348976135 |
| Loss at step 200: 0.0327930673956871 |
| Loss at step 250: 0.03407374769449234 |
| Loss at step 300: 0.0346791073679924 |
| Loss at step 350: 0.03419278562068939 |
| Loss at step 400: 0.04033670201897621 |
| Loss at step 450: 0.037295371294021606 |
| Loss at step 500: 0.036423176527023315 |
| Loss at step 550: 0.03983192518353462 |
| Loss at step 600: 0.03562218323349953 |
| Loss at step 650: 0.057547878473997116 |
| Loss at step 700: 0.04399918019771576 |
| Loss at step 750: 0.06756962835788727 |
| Loss at step 800: 0.0535319484770298 |
| Loss at step 850: 0.040021199733018875 |
| Loss at step 900: 0.05363428220152855 |
| Mean training loss after epoch 58: 0.04227230368035117 |
|
|
| EPOCH: 59 |
| Loss at step 0: 0.054008278995752335 |
| Loss at step 50: 0.029929330572485924 |
| Loss at step 100: 0.043072547763586044 |
| Loss at step 150: 0.04712403565645218 |
| Loss at step 200: 0.04172234237194061 |
| Loss at step 250: 0.0380144938826561 |
| Loss at step 300: 0.06645963340997696 |
| Loss at step 350: 0.03476366773247719 |
| Loss at step 400: 0.044849783182144165 |
| Loss at step 450: 0.03969757631421089 |
| Loss at step 500: 0.041757795959711075 |
| Loss at step 550: 0.03557545319199562 |
| Loss at step 600: 0.03668488934636116 |
| Loss at step 650: 0.034634947776794434 |
| Loss at step 700: 0.036369070410728455 |
| Loss at step 750: 0.027953891083598137 |
| Loss at step 800: 0.05794013664126396 |
| Loss at step 850: 0.0434873141348362 |
| Loss at step 900: 0.04062079265713692 |
| Mean training loss after epoch 59: 0.04175872626557533 |
|
|
| EPOCH: 60 |
| Loss at step 0: 0.05230647698044777 |
| Loss at step 50: 0.034911252558231354 |
| Loss at step 100: 0.038966234773397446 |
| Loss at step 150: 0.04930701106786728 |
| Loss at step 200: 0.039878372102975845 |
| Loss at step 250: 0.032641004770994186 |
| Loss at step 300: 0.05420990288257599 |
| Loss at step 350: 0.035432182252407074 |
| Loss at step 400: 0.047906797379255295 |
| Loss at step 450: 0.03237815201282501 |
| Loss at step 500: 0.05809643119573593 |
| Loss at step 550: 0.03982365503907204 |
| Loss at step 600: 0.03338718041777611 |
| Loss at step 650: 0.06101324036717415 |
| Loss at step 700: 0.05832071974873543 |
| Loss at step 750: 0.05672917887568474 |
| Loss at step 800: 0.035312261432409286 |
| Loss at step 850: 0.03269599750638008 |
| Loss at step 900: 0.03432564437389374 |
| Mean training loss after epoch 60: 0.04145304879336469 |
|
|
| EPOCH: 61 |
| Loss at step 0: 0.05218600109219551 |
| Loss at step 50: 0.028676575049757957 |
| Loss at step 100: 0.05183856189250946 |
| Loss at step 150: 0.04828957840800285 |
| Loss at step 200: 0.03797846660017967 |
| Loss at step 250: 0.04018789157271385 |
| Loss at step 300: 0.029013371095061302 |
| Loss at step 350: 0.046787358820438385 |
| Loss at step 400: 0.03374910354614258 |
| Loss at step 450: 0.04779788479208946 |
| Loss at step 500: 0.03813735023140907 |
| Loss at step 550: 0.06082974001765251 |
| Loss at step 600: 0.031995780766010284 |
| Loss at step 650: 0.036440473049879074 |
| Loss at step 700: 0.03242507576942444 |
| Loss at step 750: 0.047881174832582474 |
| Loss at step 800: 0.03955503925681114 |
| Loss at step 850: 0.039465587586164474 |
| Loss at step 900: 0.032889991998672485 |
| Mean training loss after epoch 61: 0.04225412909505464 |
|
|
| EPOCH: 62 |
| Loss at step 0: 0.07850071787834167 |
| Loss at step 50: 0.03590470924973488 |
| Loss at step 100: 0.03401835262775421 |
| Loss at step 150: 0.03575047105550766 |
| Loss at step 200: 0.03225778415799141 |
| Loss at step 250: 0.03880147635936737 |
| Loss at step 300: 0.03870739787817001 |
| Loss at step 350: 0.042368412017822266 |
| Loss at step 400: 0.03864947706460953 |
| Loss at step 450: 0.03193790838122368 |
| Loss at step 500: 0.043149128556251526 |
| Loss at step 550: 0.03357163816690445 |
| Loss at step 600: 0.043366439640522 |
| Loss at step 650: 0.04025929793715477 |
| Loss at step 700: 0.035577934235334396 |
| Loss at step 750: 0.03176751732826233 |
| Loss at step 800: 0.030962109565734863 |
| Loss at step 850: 0.03757867589592934 |
| Loss at step 900: 0.03229587897658348 |
| Mean training loss after epoch 62: 0.04198516900144787 |
|
|
| EPOCH: 63 |
| Loss at step 0: 0.057671867311000824 |
| Loss at step 50: 0.04842713847756386 |
| Loss at step 100: 0.030468273907899857 |
| Loss at step 150: 0.05220678821206093 |
| Loss at step 200: 0.04170433431863785 |
| Loss at step 250: 0.05136094242334366 |
| Loss at step 300: 0.04169003665447235 |
| Loss at step 350: 0.031596481800079346 |
| Loss at step 400: 0.04273003712296486 |
| Loss at step 450: 0.061871133744716644 |
| Loss at step 500: 0.04741637408733368 |
| Loss at step 550: 0.03889886289834976 |
| Loss at step 600: 0.0323534719645977 |
| Loss at step 650: 0.03431367874145508 |
| Loss at step 700: 0.03811480104923248 |
| Loss at step 750: 0.03434113413095474 |
| Loss at step 800: 0.049015481024980545 |
| Loss at step 850: 0.04040595889091492 |
| Loss at step 900: 0.03862114995718002 |
| Mean training loss after epoch 63: 0.04131053804334547 |
|
|
| EPOCH: 64 |
| Loss at step 0: 0.030019661411643028 |
| Loss at step 50: 0.03455144912004471 |
| Loss at step 100: 0.04964865371584892 |
| Loss at step 150: 0.03217058628797531 |
| Loss at step 200: 0.03193427622318268 |
| Loss at step 250: 0.03496089577674866 |
| Loss at step 300: 0.036626748740673065 |
| Loss at step 350: 0.0424736887216568 |
| Loss at step 400: 0.03943726047873497 |
| Loss at step 450: 0.040541987866163254 |
| Loss at step 500: 0.03838001936674118 |
| Loss at step 550: 0.03629530221223831 |
| Loss at step 600: 0.033054422587156296 |
| Loss at step 650: 0.0398416668176651 |
| Loss at step 700: 0.050053808838129044 |
| Loss at step 750: 0.035665612667798996 |
| Loss at step 800: 0.03748640790581703 |
| Loss at step 850: 0.04931091517210007 |
| Loss at step 900: 0.035654835402965546 |
| Mean training loss after epoch 64: 0.041265397145947035 |
|
|
| EPOCH: 65 |
| Loss at step 0: 0.03760325163602829 |
| Loss at step 50: 0.03528781980276108 |
| Loss at step 100: 0.0328512042760849 |
| Loss at step 150: 0.03513569384813309 |
| Loss at step 200: 0.032193057239055634 |
| Loss at step 250: 0.038936797529459 |
| Loss at step 300: 0.03411776199936867 |
| Loss at step 350: 0.050534818321466446 |
| Loss at step 400: 0.05266318842768669 |
| Loss at step 450: 0.03857654333114624 |
| Loss at step 500: 0.03783871978521347 |
| Loss at step 550: 0.03197762742638588 |
| Loss at step 600: 0.03530753776431084 |
| Loss at step 650: 0.038086168467998505 |
| Loss at step 700: 0.03254755958914757 |
| Loss at step 750: 0.033772069960832596 |
| Loss at step 800: 0.03325922414660454 |
| Loss at step 850: 0.037996433675289154 |
| Loss at step 900: 0.044859953224658966 |
| Mean training loss after epoch 65: 0.041470230835031215 |
|
|
| EPOCH: 66 |
| Loss at step 0: 0.039558980613946915 |
| Loss at step 50: 0.03798419609665871 |
| Loss at step 100: 0.0470040962100029 |
| Loss at step 150: 0.03549093380570412 |
| Loss at step 200: 0.06594263017177582 |
| Loss at step 250: 0.04572553560137749 |
| Loss at step 300: 0.05195481330156326 |
| Loss at step 350: 0.03912719711661339 |
| Loss at step 400: 0.033564552664756775 |
| Loss at step 450: 0.06788386404514313 |
| Loss at step 500: 0.04538029804825783 |
| Loss at step 550: 0.0429832860827446 |
| Loss at step 600: 0.044070933014154434 |
| Loss at step 650: 0.03674892336130142 |
| Loss at step 700: 0.04859331250190735 |
| Loss at step 750: 0.04238688573241234 |
| Loss at step 800: 0.05128232762217522 |
| Loss at step 850: 0.026888985186815262 |
| Loss at step 900: 0.054580822587013245 |
| Mean training loss after epoch 66: 0.04168821637159281 |
|
|
| EPOCH: 67 |
| Loss at step 0: 0.048112139105796814 |
| Loss at step 50: 0.03597401827573776 |
| Loss at step 100: 0.040139153599739075 |
| Loss at step 150: 0.03657975420355797 |
| Loss at step 200: 0.03313577175140381 |
| Loss at step 250: 0.03261137753725052 |
| Loss at step 300: 0.032174672931432724 |
| Loss at step 350: 0.03411630168557167 |
| Loss at step 400: 0.05067075043916702 |
| Loss at step 450: 0.03755392134189606 |
| Loss at step 500: 0.049025941640138626 |
| Loss at step 550: 0.053921084851026535 |
| Loss at step 600: 0.03517812862992287 |
| Loss at step 650: 0.03399652615189552 |
| Loss at step 700: 0.04195648804306984 |
| Loss at step 750: 0.05345770716667175 |
| Loss at step 800: 0.03349097818136215 |
| Loss at step 850: 0.0440969280898571 |
| Loss at step 900: 0.03051835112273693 |
| Mean training loss after epoch 67: 0.04172010423698977 |
|
|
| EPOCH: 68 |
| Loss at step 0: 0.029953550547361374 |
| Loss at step 50: 0.04670592024922371 |
| Loss at step 100: 0.035323210060596466 |
| Loss at step 150: 0.046807896345853806 |
| Loss at step 200: 0.062253061681985855 |
| Loss at step 250: 0.03276786580681801 |
| Loss at step 300: 0.033763352781534195 |
| Loss at step 350: 0.02836601622402668 |
| Loss at step 400: 0.03322148323059082 |
| Loss at step 450: 0.02729177102446556 |
| Loss at step 500: 0.04363562911748886 |
| Loss at step 550: 0.04207039624452591 |
| Loss at step 600: 0.04303210973739624 |
| Loss at step 650: 0.04108305275440216 |
| Loss at step 700: 0.033733610063791275 |
| Loss at step 750: 0.03041158616542816 |
| Loss at step 800: 0.03126723691821098 |
| Loss at step 850: 0.049652907997369766 |
| Loss at step 900: 0.04426473379135132 |
| Mean training loss after epoch 68: 0.04149038618855448 |
|
|
| EPOCH: 69 |
| Loss at step 0: 0.05051280930638313 |
| Loss at step 50: 0.032365623861551285 |
| Loss at step 100: 0.035237330943346024 |
| Loss at step 150: 0.03612693399190903 |
| Loss at step 200: 0.03822920098900795 |
| Loss at step 250: 0.029838262125849724 |
| Loss at step 300: 0.038925401866436005 |
| Loss at step 350: 0.03142131119966507 |
| Loss at step 400: 0.0644800141453743 |
| Loss at step 450: 0.03479154035449028 |
| Loss at step 500: 0.05621056631207466 |
| Loss at step 550: 0.035777267068624496 |
| Loss at step 600: 0.03171798586845398 |
| Loss at step 650: 0.03767917677760124 |
| Loss at step 700: 0.03505399078130722 |
| Loss at step 750: 0.035311244428157806 |
| Loss at step 800: 0.03133586049079895 |
| Loss at step 850: 0.03778800740838051 |
| Loss at step 900: 0.03941011801362038 |
| Mean training loss after epoch 69: 0.04125585565879655 |
|
|
| EPOCH: 70 |
| Loss at step 0: 0.033629123121500015 |
| Loss at step 50: 0.047444071620702744 |
| Loss at step 100: 0.05690629407763481 |
| Loss at step 150: 0.036379892379045486 |
| Loss at step 200: 0.03223971277475357 |
| Loss at step 250: 0.03425416722893715 |
| Loss at step 300: 0.05382617935538292 |
| Loss at step 350: 0.03412552550435066 |
| Loss at step 400: 0.03555070981383324 |
| Loss at step 450: 0.03165304288268089 |
| Loss at step 500: 0.056805454194545746 |
| Loss at step 550: 0.04862965643405914 |
| Loss at step 600: 0.045455142855644226 |
| Loss at step 650: 0.04065563902258873 |
| Loss at step 700: 0.034020159393548965 |
| Loss at step 750: 0.03897823393344879 |
| Loss at step 800: 0.04187474772334099 |
| Loss at step 850: 0.03641607239842415 |
| Loss at step 900: 0.04018361493945122 |
| Mean training loss after epoch 70: 0.04117715101379321 |
|
|
| EPOCH: 71 |
| Loss at step 0: 0.031739939004182816 |
| Loss at step 50: 0.03575426712632179 |
| Loss at step 100: 0.041028402745723724 |
| Loss at step 150: 0.03342348709702492 |
| Loss at step 200: 0.03431496396660805 |
| Loss at step 250: 0.03221401944756508 |
| Loss at step 300: 0.048587020486593246 |
| Loss at step 350: 0.03845015913248062 |
| Loss at step 400: 0.04822957515716553 |
| Loss at step 450: 0.031151946634054184 |
| Loss at step 500: 0.033823247998952866 |
| Loss at step 550: 0.032001834362745285 |
| Loss at step 600: 0.04520208016037941 |
| Loss at step 650: 0.03050629422068596 |
| Loss at step 700: 0.05655837431550026 |
| Loss at step 750: 0.04253324121236801 |
| Loss at step 800: 0.05259395018219948 |
| Loss at step 850: 0.039520565420389175 |
| Loss at step 900: 0.052242618054151535 |
| Mean training loss after epoch 71: 0.04180061946243747 |
|
|
| EPOCH: 72 |
| Loss at step 0: 0.03948841243982315 |
| Loss at step 50: 0.053918808698654175 |
| Loss at step 100: 0.03918600454926491 |
| Loss at step 150: 0.05136851593852043 |
| Loss at step 200: 0.03424490615725517 |
| Loss at step 250: 0.05601086467504501 |
| Loss at step 300: 0.03227918967604637 |
| Loss at step 350: 0.05336814001202583 |
| Loss at step 400: 0.050874777138233185 |
| Loss at step 450: 0.037637416273355484 |
| Loss at step 500: 0.02950294502079487 |
| Loss at step 550: 0.04061894118785858 |
| Loss at step 600: 0.030604930594563484 |
| Loss at step 650: 0.051947418600320816 |
| Loss at step 700: 0.03727438300848007 |
| Loss at step 750: 0.03555244579911232 |
| Loss at step 800: 0.043268777430057526 |
| Loss at step 850: 0.03692298382520676 |
| Loss at step 900: 0.048946987837553024 |
| Mean training loss after epoch 72: 0.04081341141521105 |
|
|
| EPOCH: 73 |
| Loss at step 0: 0.02806745283305645 |
| Loss at step 50: 0.029986565932631493 |
| Loss at step 100: 0.030133534222841263 |
| Loss at step 150: 0.04718175157904625 |
| Loss at step 200: 0.07402471452951431 |
| Loss at step 250: 0.038135360926389694 |
| Loss at step 300: 0.037517957389354706 |
| Loss at step 350: 0.039786506444215775 |
| Loss at step 400: 0.04700249806046486 |
| Loss at step 450: 0.03678031638264656 |
| Loss at step 500: 0.03160851448774338 |
| Loss at step 550: 0.033725056797266006 |
| Loss at step 600: 0.0418279692530632 |
| Loss at step 650: 0.040318895131349564 |
| Loss at step 700: 0.03881711885333061 |
| Loss at step 750: 0.035916246473789215 |
| Loss at step 800: 0.04123677313327789 |
| Loss at step 850: 0.04058227315545082 |
| Loss at step 900: 0.03485511615872383 |
| Mean training loss after epoch 73: 0.04174466865824293 |
|
|
| EPOCH: 74 |
| Loss at step 0: 0.02899175137281418 |
| Loss at step 50: 0.03824091702699661 |
| Loss at step 100: 0.038165558129549026 |
| Loss at step 150: 0.03265064209699631 |
| Loss at step 200: 0.032838623970746994 |
| Loss at step 250: 0.03530019149184227 |
| Loss at step 300: 0.03534851595759392 |
| Loss at step 350: 0.04351358860731125 |
| Loss at step 400: 0.04078420624136925 |
| Loss at step 450: 0.0450337752699852 |
| Loss at step 500: 0.03650393709540367 |
| Loss at step 550: 0.05287676304578781 |
| Loss at step 600: 0.0422467403113842 |
| Loss at step 650: 0.037526313215494156 |
| Loss at step 700: 0.03549263998866081 |
| Loss at step 750: 0.032682713121175766 |
| Loss at step 800: 0.03373374044895172 |
| Loss at step 850: 0.03157420828938484 |
| Loss at step 900: 0.03660622984170914 |
| Mean training loss after epoch 74: 0.04113919925548311 |
|
|
| EPOCH: 75 |
| Loss at step 0: 0.030831068754196167 |
| Loss at step 50: 0.03200463950634003 |
| Loss at step 100: 0.03574752062559128 |
| Loss at step 150: 0.051985789090394974 |
| Loss at step 200: 0.02577095478773117 |
| Loss at step 250: 0.03551045060157776 |
| Loss at step 300: 0.036199092864990234 |
| Loss at step 350: 0.031102705746889114 |
| Loss at step 400: 0.03810339793562889 |
| Loss at step 450: 0.05419360101222992 |
| Loss at step 500: 0.05693572759628296 |
| Loss at step 550: 0.03602773696184158 |
| Loss at step 600: 0.03415317088365555 |
| Loss at step 650: 0.04872572049498558 |
| Loss at step 700: 0.039838407188653946 |
| Loss at step 750: 0.03620055317878723 |
| Loss at step 800: 0.04124224931001663 |
| Loss at step 850: 0.03605523332953453 |
| Loss at step 900: 0.05318422615528107 |
| Mean training loss after epoch 75: 0.04112100785872194 |
|
|
| EPOCH: 76 |
| Loss at step 0: 0.05268241837620735 |
| Loss at step 50: 0.038748666644096375 |
| Loss at step 100: 0.05225672572851181 |
| Loss at step 150: 0.02982509881258011 |
| Loss at step 200: 0.037055786699056625 |
| Loss at step 250: 0.036166444420814514 |
| Loss at step 300: 0.042469609528779984 |
| Loss at step 350: 0.03820810094475746 |
| Loss at step 400: 0.036943428218364716 |
| Loss at step 450: 0.04263583943247795 |
| Loss at step 500: 0.042788028717041016 |
| Loss at step 550: 0.052858270704746246 |
| Loss at step 600: 0.030450038611888885 |
| Loss at step 650: 0.04485831782221794 |
| Loss at step 700: 0.06034138798713684 |
| Loss at step 750: 0.032158125191926956 |
| Loss at step 800: 0.03894932195544243 |
| Loss at step 850: 0.05779428780078888 |
| Loss at step 900: 0.0441526398062706 |
| Mean training loss after epoch 76: 0.04154504369944334 |
|
|
| EPOCH: 77 |
| Loss at step 0: 0.039356961846351624 |
| Loss at step 50: 0.028416398912668228 |
| Loss at step 100: 0.03396177664399147 |
| Loss at step 150: 0.04099806770682335 |
| Loss at step 200: 0.06283032149076462 |
| Loss at step 250: 0.04138881340622902 |
| Loss at step 300: 0.061035893857479095 |
| Loss at step 350: 0.04593011736869812 |
| Loss at step 400: 0.05210472643375397 |
| Loss at step 450: 0.03505970537662506 |
| Loss at step 500: 0.032422494143247604 |
| Loss at step 550: 0.05127855762839317 |
| Loss at step 600: 0.036935120820999146 |
| Loss at step 650: 0.03441636264324188 |
| Loss at step 700: 0.03484315797686577 |
| Loss at step 750: 0.03388001397252083 |
| Loss at step 800: 0.037996839731931686 |
| Loss at step 850: 0.03551885113120079 |
| Loss at step 900: 0.033726323395967484 |
| Mean training loss after epoch 77: 0.04113060351747122 |
|
|
| EPOCH: 78 |
| Loss at step 0: 0.032344914972782135 |
| Loss at step 50: 0.035145945847034454 |
| Loss at step 100: 0.03198443725705147 |
| Loss at step 150: 0.05172690376639366 |
| Loss at step 200: 0.04912606626749039 |
| Loss at step 250: 0.03330213204026222 |
| Loss at step 300: 0.04923785477876663 |
| Loss at step 350: 0.04205572232604027 |
| Loss at step 400: 0.03545050695538521 |
| Loss at step 450: 0.04081381857395172 |
| Loss at step 500: 0.04776203632354736 |
| Loss at step 550: 0.03815259784460068 |
| Loss at step 600: 0.05272816866636276 |
| Loss at step 650: 0.032447852194309235 |
| Loss at step 700: 0.0346088781952858 |
| Loss at step 750: 0.03316102921962738 |
| Loss at step 800: 0.03658919408917427 |
| Loss at step 850: 0.03636911138892174 |
| Loss at step 900: 0.04945993423461914 |
| Mean training loss after epoch 78: 0.041068576987205285 |
|
|
| EPOCH: 79 |
| Loss at step 0: 0.03452589362859726 |
| Loss at step 50: 0.034600645303726196 |
| Loss at step 100: 0.06494107842445374 |
| Loss at step 150: 0.036988429725170135 |
| Loss at step 200: 0.027170222252607346 |
| Loss at step 250: 0.052600547671318054 |
| Loss at step 300: 0.0349196158349514 |
| Loss at step 350: 0.031728651374578476 |
| Loss at step 400: 0.032518353313207626 |
| Loss at step 450: 0.03552037850022316 |
| Loss at step 500: 0.03851528465747833 |
| Loss at step 550: 0.03267049416899681 |
| Loss at step 600: 0.02810988575220108 |
| Loss at step 650: 0.0534193180501461 |
| Loss at step 700: 0.02989153005182743 |
| Loss at step 750: 0.04602940008044243 |
| Loss at step 800: 0.03656817600131035 |
| Loss at step 850: 0.045206040143966675 |
| Loss at step 900: 0.03326031565666199 |
| Mean training loss after epoch 79: 0.0412763856264002 |
|
|
| EPOCH: 80 |
| Loss at step 0: 0.03640799969434738 |
| Loss at step 50: 0.03698626905679703 |
| Loss at step 100: 0.036901723593473434 |
| Loss at step 150: 0.02941797859966755 |
| Loss at step 200: 0.05399346724152565 |
| Loss at step 250: 0.038527294993400574 |
| Loss at step 300: 0.03790619969367981 |
| Loss at step 350: 0.03662220761179924 |
| Loss at step 400: 0.034307632595300674 |
| Loss at step 450: 0.04109351709485054 |
| Loss at step 500: 0.03656264394521713 |
| Loss at step 550: 0.033020373433828354 |
| Loss at step 600: 0.031626660376787186 |
| Loss at step 650: 0.037163589149713516 |
| Loss at step 700: 0.03343282267451286 |
| Loss at step 750: 0.03530639782547951 |
| Loss at step 800: 0.07811196148395538 |
| Loss at step 850: 0.037117309868335724 |
| Loss at step 900: 0.036749448627233505 |
| Mean training loss after epoch 80: 0.04114491054252076 |
|
|
| EPOCH: 81 |
| Loss at step 0: 0.05479753762483597 |
| Loss at step 50: 0.03132958337664604 |
| Loss at step 100: 0.037393227219581604 |
| Loss at step 150: 0.03205867111682892 |
| Loss at step 200: 0.07083668559789658 |
| Loss at step 250: 0.04586269333958626 |
| Loss at step 300: 0.060223255306482315 |
| Loss at step 350: 0.03582262247800827 |
| Loss at step 400: 0.033127203583717346 |
| Loss at step 450: 0.02675713784992695 |
| Loss at step 500: 0.035183269530534744 |
| Loss at step 550: 0.034161537885665894 |
| Loss at step 600: 0.03505062684416771 |
| Loss at step 650: 0.05540237948298454 |
| Loss at step 700: 0.03116615302860737 |
| Loss at step 750: 0.028590409085154533 |
| Loss at step 800: 0.03489331901073456 |
| Loss at step 850: 0.03400343284010887 |
| Loss at step 900: 0.06202857568860054 |
| Mean training loss after epoch 81: 0.04102237092684517 |
|
|
| EPOCH: 82 |
| Loss at step 0: 0.03725624457001686 |
| Loss at step 50: 0.03783930093050003 |
| Loss at step 100: 0.03537292778491974 |
| Loss at step 150: 0.04294422268867493 |
| Loss at step 200: 0.03996248543262482 |
| Loss at step 250: 0.039540305733680725 |
| Loss at step 300: 0.033119600266218185 |
| Loss at step 350: 0.048681121319532394 |
| Loss at step 400: 0.05702071264386177 |
| Loss at step 450: 0.0402117483317852 |
| Loss at step 500: 0.040380872786045074 |
| Loss at step 550: 0.05610261484980583 |
| Loss at step 600: 0.035563670098781586 |
| Loss at step 650: 0.03173322603106499 |
| Loss at step 700: 0.03283904865384102 |
| Loss at step 750: 0.04451049864292145 |
| Loss at step 800: 0.03426874428987503 |
| Loss at step 850: 0.04587911069393158 |
| Loss at step 900: 0.039216335862874985 |
| Mean training loss after epoch 82: 0.041730244616185554 |
|
|
| EPOCH: 83 |
| Loss at step 0: 0.04744863510131836 |
| Loss at step 50: 0.03610475733876228 |
| Loss at step 100: 0.035557087510824203 |
| Loss at step 150: 0.03861255571246147 |
| Loss at step 200: 0.030603276565670967 |
| Loss at step 250: 0.04871303215622902 |
| Loss at step 300: 0.07224379479885101 |
| Loss at step 350: 0.03971085324883461 |
| Loss at step 400: 0.03391997143626213 |
| Loss at step 450: 0.044155556708574295 |
| Loss at step 500: 0.034668054431676865 |
| Loss at step 550: 0.05175238102674484 |
| Loss at step 600: 0.049534719437360764 |
| Loss at step 650: 0.03303040564060211 |
| Loss at step 700: 0.029844136908650398 |
| Loss at step 750: 0.04341820254921913 |
| Loss at step 800: 0.03215205669403076 |
| Loss at step 850: 0.03988838940858841 |
| Loss at step 900: 0.05346183106303215 |
| Mean training loss after epoch 83: 0.041038075031073235 |
|
|
| EPOCH: 84 |
| Loss at step 0: 0.0528486967086792 |
| Loss at step 50: 0.051950205117464066 |
| Loss at step 100: 0.040812257677316666 |
| Loss at step 150: 0.033961568027734756 |
| Loss at step 200: 0.048738449811935425 |
| Loss at step 250: 0.0713721364736557 |
| Loss at step 300: 0.031749922782182693 |
| Loss at step 350: 0.0486324168741703 |
| Loss at step 400: 0.04092682898044586 |
| Loss at step 450: 0.045225370675325394 |
| Loss at step 500: 0.03676753491163254 |
| Loss at step 550: 0.033339500427246094 |
| Loss at step 600: 0.04189130291342735 |
| Loss at step 650: 0.045919470489025116 |
| Loss at step 700: 0.03852545842528343 |
| Loss at step 750: 0.05126319080591202 |
| Loss at step 800: 0.052123866975307465 |
| Loss at step 850: 0.03563039004802704 |
| Loss at step 900: 0.032997358590364456 |
| Mean training loss after epoch 84: 0.041086105345837724 |
|
|
| EPOCH: 85 |
| Loss at step 0: 0.058152444660663605 |
| Loss at step 50: 0.038068030029535294 |
| Loss at step 100: 0.03797232359647751 |
| Loss at step 150: 0.04052959382534027 |
| Loss at step 200: 0.050743792206048965 |
| Loss at step 250: 0.03145845606923103 |
| Loss at step 300: 0.03338661044836044 |
| Loss at step 350: 0.03677058219909668 |
| Loss at step 400: 0.03495321795344353 |
| Loss at step 450: 0.039931200444698334 |
| Loss at step 500: 0.03238952159881592 |
| Loss at step 550: 0.02847830392420292 |
| Loss at step 600: 0.05000064894556999 |
| Loss at step 650: 0.0393374003469944 |
| Loss at step 700: 0.03686242923140526 |
| Loss at step 750: 0.04062030091881752 |
| Loss at step 800: 0.0347592793405056 |
| Loss at step 850: 0.031245408579707146 |
| Loss at step 900: 0.038717370480298996 |
| Mean training loss after epoch 85: 0.04107632083909662 |
|
|
| EPOCH: 86 |
| Loss at step 0: 0.03973394259810448 |
| Loss at step 50: 0.030584797263145447 |
| Loss at step 100: 0.06259417533874512 |
| Loss at step 150: 0.03153851255774498 |
| Loss at step 200: 0.055176712572574615 |
| Loss at step 250: 0.038454148918390274 |
| Loss at step 300: 0.04394076392054558 |
| Loss at step 350: 0.03507055342197418 |
| Loss at step 400: 0.03341430053114891 |
| Loss at step 450: 0.04914618283510208 |
| Loss at step 500: 0.04736623913049698 |
| Loss at step 550: 0.03985471650958061 |
| Loss at step 600: 0.03954179957509041 |
| Loss at step 650: 0.03194885700941086 |
| Loss at step 700: 0.035082027316093445 |
| Loss at step 750: 0.0482264868915081 |
| Loss at step 800: 0.03496174141764641 |
| Loss at step 850: 0.03850436955690384 |
| Loss at step 900: 0.057177163660526276 |
| Mean training loss after epoch 86: 0.04068803318194362 |
|
|
| EPOCH: 87 |
| Loss at step 0: 0.04399581253528595 |
| Loss at step 50: 0.03488341346383095 |
| Loss at step 100: 0.03428840637207031 |
| Loss at step 150: 0.04180179163813591 |
| Loss at step 200: 0.03338738903403282 |
| Loss at step 250: 0.03731721267104149 |
| Loss at step 300: 0.03757598251104355 |
| Loss at step 350: 0.034594327211380005 |
| Loss at step 400: 0.036474548280239105 |
| Loss at step 450: 0.04442448914051056 |
| Loss at step 500: 0.027858780696988106 |
| Loss at step 550: 0.05683402344584465 |
| Loss at step 600: 0.03499235212802887 |
| Loss at step 650: 0.05560460314154625 |
| Loss at step 700: 0.036094002425670624 |
| Loss at step 750: 0.046759847551584244 |
| Loss at step 800: 0.04085097834467888 |
| Loss at step 850: 0.027367902919650078 |
| Loss at step 900: 0.034118689596652985 |
| Mean training loss after epoch 87: 0.04076247038975009 |
|
|
| EPOCH: 88 |
| Loss at step 0: 0.05531320720911026 |
| Loss at step 50: 0.03014991246163845 |
| Loss at step 100: 0.05170894414186478 |
| Loss at step 150: 0.04032305255532265 |
| Loss at step 200: 0.03658030927181244 |
| Loss at step 250: 0.030780110508203506 |
| Loss at step 300: 0.034393101930618286 |
| Loss at step 350: 0.03057672642171383 |
| Loss at step 400: 0.04050949215888977 |
| Loss at step 450: 0.0360785648226738 |
| Loss at step 500: 0.03719230368733406 |
| Loss at step 550: 0.03799004480242729 |
| Loss at step 600: 0.03628097102046013 |
| Loss at step 650: 0.03763250261545181 |
| Loss at step 700: 0.036424919962882996 |
| Loss at step 750: 0.04473429173231125 |
| Loss at step 800: 0.04693128913640976 |
| Loss at step 850: 0.04265372082591057 |
| Loss at step 900: 0.031387049704790115 |
| Mean training loss after epoch 88: 0.04099947016146074 |
|
|
| EPOCH: 89 |
| Loss at step 0: 0.034374941140413284 |
| Loss at step 50: 0.0402178056538105 |
| Loss at step 100: 0.03843294084072113 |
| Loss at step 150: 0.036413002759218216 |
| Loss at step 200: 0.039501357823610306 |
| Loss at step 250: 0.029510509222745895 |
| Loss at step 300: 0.036525655537843704 |
| Loss at step 350: 0.04213511198759079 |
| Loss at step 400: 0.037092261016368866 |
| Loss at step 450: 0.03379799425601959 |
| Loss at step 500: 0.03295033797621727 |
| Loss at step 550: 0.031381383538246155 |
| Loss at step 600: 0.0407857745885849 |
| Loss at step 650: 0.03752834349870682 |
| Loss at step 700: 0.03810695558786392 |
| Loss at step 750: 0.03579120337963104 |
| Loss at step 800: 0.0520140565931797 |
| Loss at step 850: 0.038909491151571274 |
| Loss at step 900: 0.03678249195218086 |
| Mean training loss after epoch 89: 0.040811247470329944 |
|
|
| EPOCH: 90 |
| Loss at step 0: 0.04095301404595375 |
| Loss at step 50: 0.05082303658127785 |
| Loss at step 100: 0.04138334468007088 |
| Loss at step 150: 0.03449038416147232 |
| Loss at step 200: 0.038363419473171234 |
| Loss at step 250: 0.04477785900235176 |
| Loss at step 300: 0.03956672176718712 |
| Loss at step 350: 0.04098595306277275 |
| Loss at step 400: 0.03981344401836395 |
| Loss at step 450: 0.06300228089094162 |
| Loss at step 500: 0.051460955291986465 |
| Loss at step 550: 0.03568987548351288 |
| Loss at step 600: 0.039645079523324966 |
| Loss at step 650: 0.037908464670181274 |
| Loss at step 700: 0.03197165206074715 |
| Loss at step 750: 0.031707290560007095 |
| Loss at step 800: 0.049262162297964096 |
| Loss at step 850: 0.0344785712659359 |
| Loss at step 900: 0.034114595502614975 |
| Mean training loss after epoch 90: 0.040405405840195065 |
|
|
| EPOCH: 91 |
| Loss at step 0: 0.034299690276384354 |
| Loss at step 50: 0.049692295491695404 |
| Loss at step 100: 0.04133080318570137 |
| Loss at step 150: 0.03282630443572998 |
| Loss at step 200: 0.03252938389778137 |
| Loss at step 250: 0.05053752288222313 |
| Loss at step 300: 0.0312495119869709 |
| Loss at step 350: 0.03907566890120506 |
| Loss at step 400: 0.0547638013958931 |
| Loss at step 450: 0.036120086908340454 |
| Loss at step 500: 0.04640704020857811 |
| Loss at step 550: 0.038280077278614044 |
| Loss at step 600: 0.031406719237565994 |
| Loss at step 650: 0.034109827131032944 |
| Loss at step 700: 0.037395015358924866 |
| Loss at step 750: 0.03358292579650879 |
| Loss at step 800: 0.04851951450109482 |
| Loss at step 850: 0.03339794650673866 |
| Loss at step 900: 0.035246942192316055 |
| Mean training loss after epoch 91: 0.04033133036085665 |
|
|
| EPOCH: 92 |
| Loss at step 0: 0.032269254326820374 |
| Loss at step 50: 0.031161149963736534 |
| Loss at step 100: 0.060024768114089966 |
| Loss at step 150: 0.05082898959517479 |
| Loss at step 200: 0.050425995141267776 |
| Loss at step 250: 0.04740491509437561 |
| Loss at step 300: 0.054153922945261 |
| Loss at step 350: 0.03820345178246498 |
| Loss at step 400: 0.03505898267030716 |
| Loss at step 450: 0.03790952265262604 |
| Loss at step 500: 0.033213060349226 |
| Loss at step 550: 0.03766785189509392 |
| Loss at step 600: 0.03966356813907623 |
| Loss at step 650: 0.03682432696223259 |
| Loss at step 700: 0.03281429782509804 |
| Loss at step 750: 0.04580431431531906 |
| Loss at step 800: 0.04174647107720375 |
| Loss at step 850: 0.052905965596437454 |
| Loss at step 900: 0.03535788506269455 |
| Mean training loss after epoch 92: 0.04086645406438534 |
|
|
| EPOCH: 93 |
| Loss at step 0: 0.031763963401317596 |
| Loss at step 50: 0.04610699787735939 |
| Loss at step 100: 0.03144437074661255 |
| Loss at step 150: 0.036727018654346466 |
| Loss at step 200: 0.029585840180516243 |
| Loss at step 250: 0.042516931891441345 |
| Loss at step 300: 0.04100366309285164 |
| Loss at step 350: 0.0635860487818718 |
| Loss at step 400: 0.03528447076678276 |
| Loss at step 450: 0.04262498766183853 |
| Loss at step 500: 0.04326464980840683 |
| Loss at step 550: 0.06723412871360779 |
| Loss at step 600: 0.03085038810968399 |
| Loss at step 650: 0.03192776069045067 |
| Loss at step 700: 0.03921186178922653 |
| Loss at step 750: 0.03242463245987892 |
| Loss at step 800: 0.054762352257966995 |
| Loss at step 850: 0.035244882106781006 |
| Loss at step 900: 0.03946463763713837 |
| Mean training loss after epoch 93: 0.04065965882329735 |
|
|
| EPOCH: 94 |
| Loss at step 0: 0.04889944940805435 |
| Loss at step 50: 0.03690377622842789 |
| Loss at step 100: 0.03591502830386162 |
| Loss at step 150: 0.052436813712120056 |
| Loss at step 200: 0.034925758838653564 |
| Loss at step 250: 0.0353705957531929 |
| Loss at step 300: 0.061986085027456284 |
| Loss at step 350: 0.04875928536057472 |
| Loss at step 400: 0.040921784937381744 |
| Loss at step 450: 0.03871002793312073 |
| Loss at step 500: 0.03833552077412605 |
| Loss at step 550: 0.036446262151002884 |
| Loss at step 600: 0.039174746721982956 |
| Loss at step 650: 0.03723873943090439 |
| Loss at step 700: 0.03249559924006462 |
| Loss at step 750: 0.033896006643772125 |
| Loss at step 800: 0.03666757792234421 |
| Loss at step 850: 0.03137712925672531 |
| Loss at step 900: 0.051495201885700226 |
| Mean training loss after epoch 94: 0.04074181405418336 |
|
|
| EPOCH: 95 |
| Loss at step 0: 0.0352119505405426 |
| Loss at step 50: 0.03640684485435486 |
| Loss at step 100: 0.032651618123054504 |
| Loss at step 150: 0.04401347041130066 |
| Loss at step 200: 0.040212228894233704 |
| Loss at step 250: 0.07343073934316635 |
| Loss at step 300: 0.03445030748844147 |
| Loss at step 350: 0.05565089359879494 |
| Loss at step 400: 0.03517284616827965 |
| Loss at step 450: 0.03394188731908798 |
| Loss at step 500: 0.032530587166547775 |
| Loss at step 550: 0.05683201923966408 |
| Loss at step 600: 0.03782414272427559 |
| Loss at step 650: 0.03312551975250244 |
| Loss at step 700: 0.03550421819090843 |
| Loss at step 750: 0.051708586513996124 |
| Loss at step 800: 0.04553850367665291 |
| Loss at step 850: 0.05390772596001625 |
| Loss at step 900: 0.029858194291591644 |
| Mean training loss after epoch 95: 0.040436128432403746 |
|
|
| EPOCH: 96 |
| Loss at step 0: 0.03860674425959587 |
| Loss at step 50: 0.04928284138441086 |
| Loss at step 100: 0.030112681910395622 |
| Loss at step 150: 0.030009303241968155 |
| Loss at step 200: 0.03227289393544197 |
| Loss at step 250: 0.035471826791763306 |
| Loss at step 300: 0.045503050088882446 |
| Loss at step 350: 0.03439325466752052 |
| Loss at step 400: 0.03252046927809715 |
| Loss at step 450: 0.041864316910505295 |
| Loss at step 500: 0.06249577924609184 |
| Loss at step 550: 0.032730989158153534 |
| Loss at step 600: 0.050127070397138596 |
| Loss at step 650: 0.05265422537922859 |
| Loss at step 700: 0.033094629645347595 |
| Loss at step 750: 0.03569718822836876 |
| Loss at step 800: 0.03472054749727249 |
| Loss at step 850: 0.048938579857349396 |
| Loss at step 900: 0.04076386243104935 |
| Mean training loss after epoch 96: 0.0409784188537773 |
|
|
| EPOCH: 97 |
| Loss at step 0: 0.039326950907707214 |
| Loss at step 50: 0.032597895711660385 |
| Loss at step 100: 0.06816505640745163 |
| Loss at step 150: 0.03475397452712059 |
| Loss at step 200: 0.05419183522462845 |
| Loss at step 250: 0.0331057570874691 |
| Loss at step 300: 0.04599350318312645 |
| Loss at step 350: 0.031561993062496185 |
| Loss at step 400: 0.03588414564728737 |
| Loss at step 450: 0.044904645532369614 |
| Loss at step 500: 0.03464886173605919 |
| Loss at step 550: 0.041914213448762894 |
| Loss at step 600: 0.03834737092256546 |
| Loss at step 650: 0.051248062402009964 |
| Loss at step 700: 0.03469805046916008 |
| Loss at step 750: 0.03618532046675682 |
| Loss at step 800: 0.0414600595831871 |
| Loss at step 850: 0.03126781806349754 |
| Loss at step 900: 0.04621324688196182 |
| Mean training loss after epoch 97: 0.04076281934778001 |
|
|
| EPOCH: 98 |
| Loss at step 0: 0.03838682174682617 |
| Loss at step 50: 0.05533002316951752 |
| Loss at step 100: 0.030290279537439346 |
| Loss at step 150: 0.05067134276032448 |
| Loss at step 200: 0.04985364153981209 |
| Loss at step 250: 0.06334389001131058 |
| Loss at step 300: 0.05746433138847351 |
| Loss at step 350: 0.031397782266139984 |
| Loss at step 400: 0.03370814770460129 |
| Loss at step 450: 0.034410182386636734 |
| Loss at step 500: 0.05077733099460602 |
| Loss at step 550: 0.03374994918704033 |
| Loss at step 600: 0.042185988277196884 |
| Loss at step 650: 0.03396277129650116 |
| Loss at step 700: 0.036716412752866745 |
| Loss at step 750: 0.03274477645754814 |
| Loss at step 800: 0.033424075692892075 |
| Loss at step 850: 0.04032216593623161 |
| Loss at step 900: 0.03501669317483902 |
| Mean training loss after epoch 98: 0.040670416941409555 |
|
|
| EPOCH: 99 |
| Loss at step 0: 0.03827624395489693 |
| Loss at step 50: 0.048487599939107895 |
| Loss at step 100: 0.049376241862773895 |
| Loss at step 150: 0.0495765246450901 |
| Loss at step 200: 0.031661536544561386 |
| Loss at step 250: 0.03776783123612404 |
| Loss at step 300: 0.0347079262137413 |
| Loss at step 350: 0.05254741013050079 |
| Loss at step 400: 0.05152810364961624 |
| Loss at step 450: 0.035800751298666 |
| Loss at step 500: 0.06741048395633698 |
| Loss at step 550: 0.03415104001760483 |
| Loss at step 600: 0.03682698681950569 |
| Loss at step 650: 0.03967992216348648 |
| Loss at step 700: 0.027164705097675323 |
| Loss at step 750: 0.050535522401332855 |
| Loss at step 800: 0.02985253557562828 |
| Loss at step 850: 0.038912225514650345 |
| Loss at step 900: 0.035812925547361374 |
| Mean training loss after epoch 99: 0.04073020558891647 |
|
|
| EPOCH: 100 |
| Loss at step 0: 0.032607365399599075 |
| Loss at step 50: 0.06099787354469299 |
| Loss at step 100: 0.05005550757050514 |
| Loss at step 150: 0.03425469622015953 |
| Loss at step 200: 0.04151573032140732 |
| Loss at step 250: 0.048218220472335815 |
| Loss at step 300: 0.03169530630111694 |
| Loss at step 350: 0.028787700459361076 |
| Loss at step 400: 0.029403982684016228 |
| Loss at step 450: 0.03593473508954048 |
| Loss at step 500: 0.046153709292411804 |
| Loss at step 550: 0.03828473761677742 |
| Loss at step 600: 0.03990305960178375 |
| Loss at step 650: 0.062365055084228516 |
| Loss at step 700: 0.038013067096471786 |
| Loss at step 750: 0.033276695758104324 |
| Loss at step 800: 0.03590114414691925 |
| Loss at step 850: 0.040111057460308075 |
| Loss at step 900: 0.04882941395044327 |
| Mean training loss after epoch 100: 0.04036760972792914 |
|
|
| EPOCH: 101 |
| Loss at step 0: 0.03448909521102905 |
| Loss at step 50: 0.03724827989935875 |
| Loss at step 100: 0.06898139417171478 |
| Loss at step 150: 0.02924741432070732 |
| Loss at step 200: 0.03393682464957237 |
| Loss at step 250: 0.03785952553153038 |
| Loss at step 300: 0.03147255256772041 |
| Loss at step 350: 0.0666293352842331 |
| Loss at step 400: 0.05227957293391228 |
| Loss at step 450: 0.032412849366664886 |
| Loss at step 500: 0.028628254309296608 |
| Loss at step 550: 0.03578814119100571 |
| Loss at step 600: 0.052394889295101166 |
| Loss at step 650: 0.0314219631254673 |
| Loss at step 700: 0.03494124859571457 |
| Loss at step 750: 0.03220677748322487 |
| Loss at step 800: 0.0704156830906868 |
| Loss at step 850: 0.038153473287820816 |
| Loss at step 900: 0.04836264252662659 |
| Mean training loss after epoch 101: 0.04046514103494918 |
|
|
| EPOCH: 102 |
| Loss at step 0: 0.05540713295340538 |
| Loss at step 50: 0.06708371639251709 |
| Loss at step 100: 0.038608841598033905 |
| Loss at step 150: 0.03573644906282425 |
| Loss at step 200: 0.039247605949640274 |
| Loss at step 250: 0.032899416983127594 |
| Loss at step 300: 0.03535171225667 |
| Loss at step 350: 0.03774742782115936 |
| Loss at step 400: 0.03619568049907684 |
| Loss at step 450: 0.032232049852609634 |
| Loss at step 500: 0.05181793496012688 |
| Loss at step 550: 0.03434407338500023 |
| Loss at step 600: 0.03519069403409958 |
| Loss at step 650: 0.036724913865327835 |
| Loss at step 700: 0.038438115268945694 |
| Loss at step 750: 0.044145528227090836 |
| Loss at step 800: 0.046652257442474365 |
| Loss at step 850: 0.0551580972969532 |
| Loss at step 900: 0.03391639143228531 |
| Mean training loss after epoch 102: 0.04041277822742521 |
|
|
| EPOCH: 103 |
| Loss at step 0: 0.03112451732158661 |
| Loss at step 50: 0.04157055541872978 |
| Loss at step 100: 0.047702889889478683 |
| Loss at step 150: 0.03303459286689758 |
| Loss at step 200: 0.03123403526842594 |
| Loss at step 250: 0.046796660870313644 |
| Loss at step 300: 0.06682164967060089 |
| Loss at step 350: 0.030688513070344925 |
| Loss at step 400: 0.04190076142549515 |
| Loss at step 450: 0.05001092329621315 |
| Loss at step 500: 0.03804301470518112 |
| Loss at step 550: 0.058865346014499664 |
| Loss at step 600: 0.05053896829485893 |
| Loss at step 650: 0.03335892781615257 |
| Loss at step 700: 0.05287832021713257 |
| Loss at step 750: 0.05739011988043785 |
| Loss at step 800: 0.03445238992571831 |
| Loss at step 850: 0.061511073261499405 |
| Loss at step 900: 0.029992185533046722 |
| Mean training loss after epoch 103: 0.04018082712759087 |
|
|
| EPOCH: 104 |
| Loss at step 0: 0.05760432034730911 |
| Loss at step 50: 0.05041252821683884 |
| Loss at step 100: 0.027177143841981888 |
| Loss at step 150: 0.039961639791727066 |
| Loss at step 200: 0.0325581319630146 |
| Loss at step 250: 0.05665874481201172 |
| Loss at step 300: 0.03727357089519501 |
| Loss at step 350: 0.028202036395668983 |
| Loss at step 400: 0.03974505886435509 |
| Loss at step 450: 0.03842462599277496 |
| Loss at step 500: 0.03134937211871147 |
| Loss at step 550: 0.04020616412162781 |
| Loss at step 600: 0.04058777168393135 |
| Loss at step 650: 0.03416196629405022 |
| Loss at step 700: 0.06070096045732498 |
| Loss at step 750: 0.03729524463415146 |
| Loss at step 800: 0.04380563274025917 |
| Loss at step 850: 0.03157120570540428 |
| Loss at step 900: 0.03815029188990593 |
| Mean training loss after epoch 104: 0.040242427686002974 |
|
|
| EPOCH: 105 |
| Loss at step 0: 0.0319080725312233 |
| Loss at step 50: 0.03261766955256462 |
| Loss at step 100: 0.04951752722263336 |
| Loss at step 150: 0.031716614961624146 |
| Loss at step 200: 0.046612419188022614 |
| Loss at step 250: 0.0327620655298233 |
| Loss at step 300: 0.042399682104587555 |
| Loss at step 350: 0.04419102892279625 |
| Loss at step 400: 0.032193947583436966 |
| Loss at step 450: 0.03805965930223465 |
| Loss at step 500: 0.049167387187480927 |
| Loss at step 550: 0.03526442497968674 |
| Loss at step 600: 0.034376244992017746 |
| Loss at step 650: 0.0354459322988987 |
| Loss at step 700: 0.038397882133722305 |
| Loss at step 750: 0.05540142208337784 |
| Loss at step 800: 0.030147867277264595 |
| Loss at step 850: 0.060497745871543884 |
| Loss at step 900: 0.041193459182977676 |
| Mean training loss after epoch 105: 0.04030951934614416 |
|
|
| EPOCH: 106 |
| Loss at step 0: 0.03829112648963928 |
| Loss at step 50: 0.026048794388771057 |
| Loss at step 100: 0.0357942208647728 |
| Loss at step 150: 0.05229229852557182 |
| Loss at step 200: 0.04837630316615105 |
| Loss at step 250: 0.051231976598501205 |
| Loss at step 300: 0.04065598174929619 |
| Loss at step 350: 0.03193920850753784 |
| Loss at step 400: 0.0320788212120533 |
| Loss at step 450: 0.05464372783899307 |
| Loss at step 500: 0.026041459292173386 |
| Loss at step 550: 0.035844311118125916 |
| Loss at step 600: 0.029609395191073418 |
| Loss at step 650: 0.03459584712982178 |
| Loss at step 700: 0.036759935319423676 |
| Loss at step 750: 0.0362151637673378 |
| Loss at step 800: 0.0336429588496685 |
| Loss at step 850: 0.046769093722105026 |
| Loss at step 900: 0.03425263985991478 |
| Mean training loss after epoch 106: 0.04064369485822759 |
|
|
| EPOCH: 107 |
| Loss at step 0: 0.07431833446025848 |
| Loss at step 50: 0.0602848082780838 |
| Loss at step 100: 0.029872529208660126 |
| Loss at step 150: 0.035330332815647125 |
| Loss at step 200: 0.045894455164670944 |
| Loss at step 250: 0.030223330482840538 |
| Loss at step 300: 0.036641091108322144 |
| Loss at step 350: 0.0410873144865036 |
| Loss at step 400: 0.044177498668432236 |
| Loss at step 450: 0.03326188400387764 |
| Loss at step 500: 0.040481384843587875 |
| Loss at step 550: 0.04043734073638916 |
| Loss at step 600: 0.03910863399505615 |
| Loss at step 650: 0.04387463256716728 |
| Loss at step 700: 0.03435670956969261 |
| Loss at step 750: 0.0295183677226305 |
| Loss at step 800: 0.03602305427193642 |
| Loss at step 850: 0.02967149391770363 |
| Loss at step 900: 0.03301510587334633 |
| Mean training loss after epoch 107: 0.040982513295323736 |
|
|
| EPOCH: 108 |
| Loss at step 0: 0.05611558258533478 |
| Loss at step 50: 0.035307254642248154 |
| Loss at step 100: 0.02783018909394741 |
| Loss at step 150: 0.039593588560819626 |
| Loss at step 200: 0.06363610923290253 |
| Loss at step 250: 0.03429553285241127 |
| Loss at step 300: 0.03413379192352295 |
| Loss at step 350: 0.028662709519267082 |
| Loss at step 400: 0.039821330457925797 |
| Loss at step 450: 0.03959920257329941 |
| Loss at step 500: 0.036259181797504425 |
| Loss at step 550: 0.05385353043675423 |
| Loss at step 600: 0.026031946763396263 |
| Loss at step 650: 0.03385019302368164 |
| Loss at step 700: 0.028812086209654808 |
| Loss at step 750: 0.03397990018129349 |
| Loss at step 800: 0.03166824206709862 |
| Loss at step 850: 0.0317801870405674 |
| Loss at step 900: 0.03626967966556549 |
| Mean training loss after epoch 108: 0.040677893166142357 |
|
|
| EPOCH: 109 |
| Loss at step 0: 0.04203055799007416 |
| Loss at step 50: 0.028228793293237686 |
| Loss at step 100: 0.04802173748612404 |
| Loss at step 150: 0.0384133942425251 |
| Loss at step 200: 0.0397668182849884 |
| Loss at step 250: 0.03989081457257271 |
| Loss at step 300: 0.03930295258760452 |
| Loss at step 350: 0.03303258866071701 |
| Loss at step 400: 0.05263761058449745 |
| Loss at step 450: 0.03365996852517128 |
| Loss at step 500: 0.05358096584677696 |
| Loss at step 550: 0.04638715088367462 |
| Loss at step 600: 0.03853611648082733 |
| Loss at step 650: 0.03495076298713684 |
| Loss at step 700: 0.06538189202547073 |
| Loss at step 750: 0.028637005016207695 |
| Loss at step 800: 0.035671524703502655 |
| Loss at step 850: 0.038928259164094925 |
| Loss at step 900: 0.030003154650330544 |
| Mean training loss after epoch 109: 0.040187274618173584 |
|
|
| EPOCH: 110 |
| Loss at step 0: 0.03870801627635956 |
| Loss at step 50: 0.03549281880259514 |
| Loss at step 100: 0.05035026744008064 |
| Loss at step 150: 0.04582242667675018 |
| Loss at step 200: 0.03825569152832031 |
| Loss at step 250: 0.030541272833943367 |
| Loss at step 300: 0.039547670632600784 |
| Loss at step 350: 0.033150963485240936 |
| Loss at step 400: 0.033019401133060455 |
| Loss at step 450: 0.03229507431387901 |
| Loss at step 500: 0.03577958792448044 |
| Loss at step 550: 0.03508676588535309 |
| Loss at step 600: 0.032196156680583954 |
| Loss at step 650: 0.039787814021110535 |
| Loss at step 700: 0.05168083682656288 |
| Loss at step 750: 0.04113023355603218 |
| Loss at step 800: 0.03392178192734718 |
| Loss at step 850: 0.029219821095466614 |
| Loss at step 900: 0.04385785758495331 |
| Mean training loss after epoch 110: 0.04062215005283925 |
|
|
| EPOCH: 111 |
| Loss at step 0: 0.0526747927069664 |
| Loss at step 50: 0.038879457861185074 |
| Loss at step 100: 0.03004498779773712 |
| Loss at step 150: 0.03682366758584976 |
| Loss at step 200: 0.03794126212596893 |
| Loss at step 250: 0.0487305149435997 |
| Loss at step 300: 0.04967457801103592 |
| Loss at step 350: 0.035163119435310364 |
| Loss at step 400: 0.03427344933152199 |
| Loss at step 450: 0.03088877536356449 |
| Loss at step 500: 0.03169834613800049 |
| Loss at step 550: 0.0374300479888916 |
| Loss at step 600: 0.033706407994031906 |
| Loss at step 650: 0.035374246537685394 |
| Loss at step 700: 0.039848484098911285 |
| Loss at step 750: 0.06113433837890625 |
| Loss at step 800: 0.04747779294848442 |
| Loss at step 850: 0.03599917143583298 |
| Loss at step 900: 0.03615652397274971 |
| Mean training loss after epoch 111: 0.039936521320915554 |
|
|
| EPOCH: 112 |
| Loss at step 0: 0.0373336486518383 |
| Loss at step 50: 0.05114150419831276 |
| Loss at step 100: 0.05242919921875 |
| Loss at step 150: 0.03499115630984306 |
| Loss at step 200: 0.04813041538000107 |
| Loss at step 250: 0.03306647390127182 |
| Loss at step 300: 0.0505673922598362 |
| Loss at step 350: 0.03386365622282028 |
| Loss at step 400: 0.03137662261724472 |
| Loss at step 450: 0.03400233015418053 |
| Loss at step 500: 0.03519631177186966 |
| Loss at step 550: 0.04086479917168617 |
| Loss at step 600: 0.03908710181713104 |
| Loss at step 650: 0.05335430055856705 |
| Loss at step 700: 0.036864008754491806 |
| Loss at step 750: 0.051715828478336334 |
| Loss at step 800: 0.051827117800712585 |
| Loss at step 850: 0.03942806273698807 |
| Loss at step 900: 0.04879195615649223 |
| Mean training loss after epoch 112: 0.04063963858104909 |
|
|
| EPOCH: 113 |
| Loss at step 0: 0.03795037046074867 |
| Loss at step 50: 0.0358436293900013 |
| Loss at step 100: 0.04843265190720558 |
| Loss at step 150: 0.05632951855659485 |
| Loss at step 200: 0.03601030632853508 |
| Loss at step 250: 0.03268331661820412 |
| Loss at step 300: 0.038877252489328384 |
| Loss at step 350: 0.045339521020650864 |
| Loss at step 400: 0.032848652452230453 |
| Loss at step 450: 0.04735710099339485 |
| Loss at step 500: 0.04660831019282341 |
| Loss at step 550: 0.05198223516345024 |
| Loss at step 600: 0.04441724345088005 |
| Loss at step 650: 0.05480624735355377 |
| Loss at step 700: 0.06743644922971725 |
| Loss at step 750: 0.03541141003370285 |
| Loss at step 800: 0.03227515518665314 |
| Loss at step 850: 0.03646775335073471 |
| Loss at step 900: 0.03438189998269081 |
| Mean training loss after epoch 113: 0.04005539193471421 |
|
|
| EPOCH: 114 |
| Loss at step 0: 0.036579687148332596 |
| Loss at step 50: 0.03507820516824722 |
| Loss at step 100: 0.029598532244563103 |
| Loss at step 150: 0.03884422406554222 |
| Loss at step 200: 0.05779637023806572 |
| Loss at step 250: 0.04017007723450661 |
| Loss at step 300: 0.04302921146154404 |
| Loss at step 350: 0.033810097724199295 |
| Loss at step 400: 0.03218913450837135 |
| Loss at step 450: 0.03484483063220978 |
| Loss at step 500: 0.05256728455424309 |
| Loss at step 550: 0.032125137746334076 |
| Loss at step 600: 0.034768857061862946 |
| Loss at step 650: 0.036406390368938446 |
| Loss at step 700: 0.05086382478475571 |
| Loss at step 750: 0.03705102577805519 |
| Loss at step 800: 0.03796803578734398 |
| Loss at step 850: 0.037255316972732544 |
| Loss at step 900: 0.03930399939417839 |
| Mean training loss after epoch 114: 0.03997340493960612 |
|
|
| EPOCH: 115 |
| Loss at step 0: 0.03761971741914749 |
| Loss at step 50: 0.02730981633067131 |
| Loss at step 100: 0.029183940961956978 |
| Loss at step 150: 0.029592841863632202 |
| Loss at step 200: 0.06739065796136856 |
| Loss at step 250: 0.038498494774103165 |
| Loss at step 300: 0.0405089445412159 |
| Loss at step 350: 0.036972276866436005 |
| Loss at step 400: 0.034451935440301895 |
| Loss at step 450: 0.045539602637290955 |
| Loss at step 500: 0.03423967957496643 |
| Loss at step 550: 0.03492945432662964 |
| Loss at step 600: 0.04334418848156929 |
| Loss at step 650: 0.03467882424592972 |
| Loss at step 700: 0.03075747936964035 |
| Loss at step 750: 0.038405630737543106 |
| Loss at step 800: 0.05285104736685753 |
| Loss at step 850: 0.03415987268090248 |
| Loss at step 900: 0.03387216106057167 |
| Mean training loss after epoch 115: 0.04025659561435233 |
|
|
| EPOCH: 116 |
| Loss at step 0: 0.036216430366039276 |
| Loss at step 50: 0.05867926776409149 |
| Loss at step 100: 0.04880305752158165 |
| Loss at step 150: 0.024054808542132378 |
| Loss at step 200: 0.03405758738517761 |
| Loss at step 250: 0.0369703583419323 |
| Loss at step 300: 0.043173935264348984 |
| Loss at step 350: 0.03779338672757149 |
| Loss at step 400: 0.0379754900932312 |
| Loss at step 450: 0.04397716745734215 |
| Loss at step 500: 0.02539883553981781 |
| Loss at step 550: 0.03261855989694595 |
| Loss at step 600: 0.036792535334825516 |
| Loss at step 650: 0.03261512145400047 |
| Loss at step 700: 0.048828668892383575 |
| Loss at step 750: 0.03268284723162651 |
| Loss at step 800: 0.05022961646318436 |
| Loss at step 850: 0.04113951325416565 |
| Loss at step 900: 0.03498257324099541 |
| Mean training loss after epoch 116: 0.04056224513894269 |
|
|
| EPOCH: 117 |
| Loss at step 0: 0.0348486453294754 |
| Loss at step 50: 0.04878906533122063 |
| Loss at step 100: 0.044236086308956146 |
| Loss at step 150: 0.03162962570786476 |
| Loss at step 200: 0.03578827157616615 |
| Loss at step 250: 0.030955882743000984 |
| Loss at step 300: 0.03401421010494232 |
| Loss at step 350: 0.03497142344713211 |
| Loss at step 400: 0.035778697580099106 |
| Loss at step 450: 0.03437361121177673 |
| Loss at step 500: 0.04439618065953255 |
| Loss at step 550: 0.03746693581342697 |
| Loss at step 600: 0.034327432513237 |
| Loss at step 650: 0.033807624131441116 |
| Loss at step 700: 0.03443426638841629 |
| Loss at step 750: 0.031515758484601974 |
| Loss at step 800: 0.03196234628558159 |
| Loss at step 850: 0.037271417677402496 |
| Loss at step 900: 0.030908746644854546 |
| Mean training loss after epoch 117: 0.03997806832194328 |
|
|
| EPOCH: 118 |
| Loss at step 0: 0.06313231587409973 |
| Loss at step 50: 0.06051144748926163 |
| Loss at step 100: 0.04856443777680397 |
| Loss at step 150: 0.03401831164956093 |
| Loss at step 200: 0.03721427172422409 |
| Loss at step 250: 0.03343229740858078 |
| Loss at step 300: 0.03522949293255806 |
| Loss at step 350: 0.04258454963564873 |
| Loss at step 400: 0.06014687940478325 |
| Loss at step 450: 0.03544767573475838 |
| Loss at step 500: 0.039289381355047226 |
| Loss at step 550: 0.0608576163649559 |
| Loss at step 600: 0.03271527960896492 |
| Loss at step 650: 0.04725121706724167 |
| Loss at step 700: 0.04453769326210022 |
| Loss at step 750: 0.02915090322494507 |
| Loss at step 800: 0.05556664988398552 |
| Loss at step 850: 0.02730882354080677 |
| Loss at step 900: 0.030028440058231354 |
| Mean training loss after epoch 118: 0.04058234326279303 |
|
|
| EPOCH: 119 |
| Loss at step 0: 0.037382058799266815 |
| Loss at step 50: 0.05149533599615097 |
| Loss at step 100: 0.03135047107934952 |
| Loss at step 150: 0.030940087512135506 |
| Loss at step 200: 0.03823164850473404 |
| Loss at step 250: 0.03651123121380806 |
| Loss at step 300: 0.03233299031853676 |
| Loss at step 350: 0.036667048931121826 |
| Loss at step 400: 0.03249666467308998 |
| Loss at step 450: 0.03559266775846481 |
| Loss at step 500: 0.0458882637321949 |
| Loss at step 550: 0.03821825981140137 |
| Loss at step 600: 0.033053841441869736 |
| Loss at step 650: 0.03599870204925537 |
| Loss at step 700: 0.029937120154500008 |
| Loss at step 750: 0.03885503485798836 |
| Loss at step 800: 0.03379277139902115 |
| Loss at step 850: 0.04492877423763275 |
| Loss at step 900: 0.04187827929854393 |
| Mean training loss after epoch 119: 0.04040607897195417 |
|
|
| EPOCH: 120 |
| Loss at step 0: 0.028167858719825745 |
| Loss at step 50: 0.04726026579737663 |
| Loss at step 100: 0.0361696258187294 |
| Loss at step 150: 0.030510244891047478 |
| Loss at step 200: 0.0330553762614727 |
| Loss at step 250: 0.03535635769367218 |
| Loss at step 300: 0.05735327675938606 |
| Loss at step 350: 0.03452423959970474 |
| Loss at step 400: 0.04450173303484917 |
| Loss at step 450: 0.03576704487204552 |
| Loss at step 500: 0.046488843858242035 |
| Loss at step 550: 0.04781826585531235 |
| Loss at step 600: 0.04467238113284111 |
| Loss at step 650: 0.04754528030753136 |
| Loss at step 700: 0.043728072196245193 |
| Loss at step 750: 0.04102247580885887 |
| Loss at step 800: 0.054334353655576706 |
| Loss at step 850: 0.04936356842517853 |
| Loss at step 900: 0.044593654572963715 |
| Mean training loss after epoch 120: 0.03998779853198256 |
|
|
| EPOCH: 121 |
| Loss at step 0: 0.05285077169537544 |
| Loss at step 50: 0.028145067393779755 |
| Loss at step 100: 0.029211612418293953 |
| Loss at step 150: 0.04488343000411987 |
| Loss at step 200: 0.029937224462628365 |
| Loss at step 250: 0.05586611479520798 |
| Loss at step 300: 0.047047797590494156 |
| Loss at step 350: 0.052246637642383575 |
| Loss at step 400: 0.05250212177634239 |
| Loss at step 450: 0.046561263501644135 |
| Loss at step 500: 0.029796786606311798 |
| Loss at step 550: 0.03729530796408653 |
| Loss at step 600: 0.035116907209157944 |
| Loss at step 650: 0.038523416966199875 |
| Loss at step 700: 0.055753014981746674 |
| Loss at step 750: 0.030627282336354256 |
| Loss at step 800: 0.036290332674980164 |
| Loss at step 850: 0.03514028340578079 |
| Loss at step 900: 0.0453634150326252 |
| Mean training loss after epoch 121: 0.040457542389948996 |
|
|
| EPOCH: 122 |
| Loss at step 0: 0.031591176986694336 |
| Loss at step 50: 0.03447073698043823 |
| Loss at step 100: 0.041707564145326614 |
| Loss at step 150: 0.06623993813991547 |
| Loss at step 200: 0.03063139319419861 |
| Loss at step 250: 0.033697038888931274 |
| Loss at step 300: 0.05644533410668373 |
| Loss at step 350: 0.06907740980386734 |
| Loss at step 400: 0.036720581352710724 |
| Loss at step 450: 0.0389428436756134 |
| Loss at step 500: 0.02795400097966194 |
| Loss at step 550: 0.03139829635620117 |
| Loss at step 600: 0.03670325130224228 |
| Loss at step 650: 0.04355520382523537 |
| Loss at step 700: 0.03025864250957966 |
| Loss at step 750: 0.0512089803814888 |
| Loss at step 800: 0.0327790305018425 |
| Loss at step 850: 0.03410333767533302 |
| Loss at step 900: 0.034768156707286835 |
| Mean training loss after epoch 122: 0.04006110403790022 |
|
|
| EPOCH: 123 |
| Loss at step 0: 0.037590596824884415 |
| Loss at step 50: 0.05157771334052086 |
| Loss at step 100: 0.05391733720898628 |
| Loss at step 150: 0.05016305297613144 |
| Loss at step 200: 0.04134446755051613 |
| Loss at step 250: 0.037453681230545044 |
| Loss at step 300: 0.05308147147297859 |
| Loss at step 350: 0.028448354452848434 |
| Loss at step 400: 0.03311985731124878 |
| Loss at step 450: 0.047355495393276215 |
| Loss at step 500: 0.03463435545563698 |
| Loss at step 550: 0.034671586006879807 |
| Loss at step 600: 0.03247098624706268 |
| Loss at step 650: 0.048032741993665695 |
| Loss at step 700: 0.03710617870092392 |
| Loss at step 750: 0.031204132363200188 |
| Loss at step 800: 0.038457345217466354 |
| Loss at step 850: 0.043605733662843704 |
| Loss at step 900: 0.03489955887198448 |
| Mean training loss after epoch 123: 0.04032010728044551 |
|
|
| EPOCH: 124 |
| Loss at step 0: 0.033983953297138214 |
| Loss at step 50: 0.04215710237622261 |
| Loss at step 100: 0.033741917461156845 |
| Loss at step 150: 0.037536218762397766 |
| Loss at step 200: 0.035390716046094894 |
| Loss at step 250: 0.033167146146297455 |
| Loss at step 300: 0.03190934285521507 |
| Loss at step 350: 0.03862353414297104 |
| Loss at step 400: 0.03355590999126434 |
| Loss at step 450: 0.03378209099173546 |
| Loss at step 500: 0.03174929693341255 |
| Loss at step 550: 0.031022535637021065 |
| Loss at step 600: 0.04976525530219078 |
| Loss at step 650: 0.04322446137666702 |
| Loss at step 700: 0.03885587304830551 |
| Loss at step 750: 0.037186697125434875 |
| Loss at step 800: 0.04601681977510452 |
| Loss at step 850: 0.03291945904493332 |
| Loss at step 900: 0.042094048112630844 |
| Mean training loss after epoch 124: 0.040002900197593644 |
|
|
| EPOCH: 125 |
| Loss at step 0: 0.0500323623418808 |
| Loss at step 50: 0.05065305903553963 |
| Loss at step 100: 0.033764030784368515 |
| Loss at step 150: 0.050167616456747055 |
| Loss at step 200: 0.030769914388656616 |
| Loss at step 250: 0.04813048616051674 |
| Loss at step 300: 0.03515201061964035 |
| Loss at step 350: 0.03375561162829399 |
| Loss at step 400: 0.050982020795345306 |
| Loss at step 450: 0.035507962107658386 |
| Loss at step 500: 0.03745051100850105 |
| Loss at step 550: 0.03788281977176666 |
| Loss at step 600: 0.03262507915496826 |
| Loss at step 650: 0.032084278762340546 |
| Loss at step 700: 0.040814898908138275 |
| Loss at step 750: 0.03551555424928665 |
| Loss at step 800: 0.03950602188706398 |
| Loss at step 850: 0.03731091693043709 |
| Loss at step 900: 0.034966662526130676 |
| Mean training loss after epoch 125: 0.040024525072894244 |
|
|
| EPOCH: 126 |
| Loss at step 0: 0.034171294420957565 |
| Loss at step 50: 0.033561307936906815 |
| Loss at step 100: 0.0351739265024662 |
| Loss at step 150: 0.0548534132540226 |
| Loss at step 200: 0.03751838952302933 |
| Loss at step 250: 0.045688968151807785 |
| Loss at step 300: 0.04876801744103432 |
| Loss at step 350: 0.052135396748781204 |
| Loss at step 400: 0.034007418900728226 |
| Loss at step 450: 0.03813653811812401 |
| Loss at step 500: 0.06047852709889412 |
| Loss at step 550: 0.06819915026426315 |
| Loss at step 600: 0.03307413309812546 |
| Loss at step 650: 0.035687174648046494 |
| Loss at step 700: 0.03782489895820618 |
| Loss at step 750: 0.03208120912313461 |
| Loss at step 800: 0.062991663813591 |
| Loss at step 850: 0.04625096544623375 |
| Loss at step 900: 0.0389222614467144 |
| Mean training loss after epoch 126: 0.03970333728303851 |
|
|
| EPOCH: 127 |
| Loss at step 0: 0.03681530803442001 |
| Loss at step 50: 0.04381631314754486 |
| Loss at step 100: 0.05013523995876312 |
| Loss at step 150: 0.06218918785452843 |
| Loss at step 200: 0.03095322847366333 |
| Loss at step 250: 0.03979748860001564 |
| Loss at step 300: 0.06336919218301773 |
| Loss at step 350: 0.03274674341082573 |
| Loss at step 400: 0.04084060713648796 |
| Loss at step 450: 0.03177779167890549 |
| Loss at step 500: 0.03313849866390228 |
| Loss at step 550: 0.038118936121463776 |
| Loss at step 600: 0.03555034101009369 |
| Loss at step 650: 0.0577574260532856 |
| Loss at step 700: 0.03207573667168617 |
| Loss at step 750: 0.020524565130472183 |
| Loss at step 800: 0.03095831535756588 |
| Loss at step 850: 0.026542063802480698 |
| Loss at step 900: 0.051946502178907394 |
| Mean training loss after epoch 127: 0.039454418913657856 |
|
|
| EPOCH: 128 |
| Loss at step 0: 0.03673482686281204 |
| Loss at step 50: 0.03247862681746483 |
| Loss at step 100: 0.031764429062604904 |
| Loss at step 150: 0.02867070399224758 |
| Loss at step 200: 0.039199747145175934 |
| Loss at step 250: 0.05398840084671974 |
| Loss at step 300: 0.06494774669408798 |
| Loss at step 350: 0.05154864490032196 |
| Loss at step 400: 0.0476166307926178 |
| Loss at step 450: 0.06569583714008331 |
| Loss at step 500: 0.03495299443602562 |
| Loss at step 550: 0.03356633707880974 |
| Loss at step 600: 0.046172380447387695 |
| Loss at step 650: 0.047938257455825806 |
| Loss at step 700: 0.036930497735738754 |
| Loss at step 750: 0.038262851536273956 |
| Loss at step 800: 0.03517230600118637 |
| Loss at step 850: 0.034791771322488785 |
| Loss at step 900: 0.04119005799293518 |
| Mean training loss after epoch 128: 0.04036835033788101 |
|
|
| EPOCH: 129 |
| Loss at step 0: 0.03706024959683418 |
| Loss at step 50: 0.03425043448805809 |
| Loss at step 100: 0.02833772636950016 |
| Loss at step 150: 0.03949480876326561 |
| Loss at step 200: 0.03533528372645378 |
| Loss at step 250: 0.03487789258360863 |
| Loss at step 300: 0.048765428364276886 |
| Loss at step 350: 0.044877734035253525 |
| Loss at step 400: 0.037002693861722946 |
| Loss at step 450: 0.04018561169505119 |
| Loss at step 500: 0.03295544534921646 |
| Loss at step 550: 0.049722786992788315 |
| Loss at step 600: 0.055026598274707794 |
| Loss at step 650: 0.03380793333053589 |
| Loss at step 700: 0.05094536021351814 |
| Loss at step 750: 0.03653072938323021 |
| Loss at step 800: 0.03610605001449585 |
| Loss at step 850: 0.040753696113824844 |
| Loss at step 900: 0.03856388479471207 |
| Mean training loss after epoch 129: 0.039871999711942066 |
|
|
| EPOCH: 130 |
| Loss at step 0: 0.048286810517311096 |
| Loss at step 50: 0.047162044793367386 |
| Loss at step 100: 0.038353268057107925 |
| Loss at step 150: 0.0325942263007164 |
| Loss at step 200: 0.03514799103140831 |
| Loss at step 250: 0.03141734004020691 |
| Loss at step 300: 0.04474622756242752 |
| Loss at step 350: 0.03544167801737785 |
| Loss at step 400: 0.038670141249895096 |
| Loss at step 450: 0.046402618288993835 |
| Loss at step 500: 0.03563407063484192 |
| Loss at step 550: 0.04327638819813728 |
| Loss at step 600: 0.040320947766304016 |
| Loss at step 650: 0.040106456726789474 |
| Loss at step 700: 0.03996186703443527 |
| Loss at step 750: 0.037617284804582596 |
| Loss at step 800: 0.039552800357341766 |
| Loss at step 850: 0.06186339259147644 |
| Loss at step 900: 0.03492647036910057 |
| Mean training loss after epoch 130: 0.03991808345926596 |
|
|
| EPOCH: 131 |
| Loss at step 0: 0.036616213619709015 |
| Loss at step 50: 0.04928402975201607 |
| Loss at step 100: 0.04600382223725319 |
| Loss at step 150: 0.05377383902668953 |
| Loss at step 200: 0.04120911657810211 |
| Loss at step 250: 0.0662044882774353 |
| Loss at step 300: 0.04024979844689369 |
| Loss at step 350: 0.030393140390515327 |
| Loss at step 400: 0.042372897267341614 |
| Loss at step 450: 0.037753261625766754 |
| Loss at step 500: 0.031086349859833717 |
| Loss at step 550: 0.03298509865999222 |
| Loss at step 600: 0.03297559171915054 |
| Loss at step 650: 0.030049888417124748 |
| Loss at step 700: 0.035726603120565414 |
| Loss at step 750: 0.04066278785467148 |
| Loss at step 800: 0.03545417636632919 |
| Loss at step 850: 0.03513539209961891 |
| Loss at step 900: 0.0556241013109684 |
| Mean training loss after epoch 131: 0.0402852121804124 |
|
|
| EPOCH: 132 |
| Loss at step 0: 0.039677299559116364 |
| Loss at step 50: 0.04367973282933235 |
| Loss at step 100: 0.05246155336499214 |
| Loss at step 150: 0.034604262560606 |
| Loss at step 200: 0.049703508615493774 |
| Loss at step 250: 0.03727739304304123 |
| Loss at step 300: 0.049680761992931366 |
| Loss at step 350: 0.0293569453060627 |
| Loss at step 400: 0.0443883016705513 |
| Loss at step 450: 0.0507032610476017 |
| Loss at step 500: 0.03775497153401375 |
| Loss at step 550: 0.0356106273829937 |
| Loss at step 600: 0.06872064620256424 |
| Loss at step 650: 0.03747250884771347 |
| Loss at step 700: 0.03364582359790802 |
| Loss at step 750: 0.05006100982427597 |
| Loss at step 800: 0.049780216068029404 |
| Loss at step 850: 0.035216640681028366 |
| Loss at step 900: 0.03414424508810043 |
| Mean training loss after epoch 132: 0.03970895380750775 |
|
|
| EPOCH: 133 |
| Loss at step 0: 0.036541033536195755 |
| Loss at step 50: 0.03570520505309105 |
| Loss at step 100: 0.03586961328983307 |
| Loss at step 150: 0.033113136887550354 |
| Loss at step 200: 0.031563837081193924 |
| Loss at step 250: 0.03401624783873558 |
| Loss at step 300: 0.0378023199737072 |
| Loss at step 350: 0.03832368552684784 |
| Loss at step 400: 0.02344367839396 |
| Loss at step 450: 0.04743019491434097 |
| Loss at step 500: 0.038056880235672 |
| Loss at step 550: 0.04119366407394409 |
| Loss at step 600: 0.03679410368204117 |
| Loss at step 650: 0.04440126568078995 |
| Loss at step 700: 0.0363314226269722 |
| Loss at step 750: 0.03746131435036659 |
| Loss at step 800: 0.05385933071374893 |
| Loss at step 850: 0.031379908323287964 |
| Loss at step 900: 0.042474813759326935 |
| Mean training loss after epoch 133: 0.039679838541045245 |
|
|
| EPOCH: 134 |
| Loss at step 0: 0.027622492983937263 |
| Loss at step 50: 0.04098658636212349 |
| Loss at step 100: 0.03279804810881615 |
| Loss at step 150: 0.04508473724126816 |
| Loss at step 200: 0.0342780165374279 |
| Loss at step 250: 0.030243201181292534 |
| Loss at step 300: 0.028790751472115517 |
| Loss at step 350: 0.03409457579255104 |
| Loss at step 400: 0.03327581286430359 |
| Loss at step 450: 0.037851233035326004 |
| Loss at step 500: 0.031168678775429726 |
| Loss at step 550: 0.0319376140832901 |
| Loss at step 600: 0.04381627216935158 |
| Loss at step 650: 0.04569874703884125 |
| Loss at step 700: 0.033138565719127655 |
| Loss at step 750: 0.03507937490940094 |
| Loss at step 800: 0.04184393957257271 |
| Loss at step 850: 0.03176693618297577 |
| Loss at step 900: 0.030614210292696953 |
| Mean training loss after epoch 134: 0.03981186858594799 |
|
|
| EPOCH: 135 |
| Loss at step 0: 0.03553043678402901 |
| Loss at step 50: 0.03979399800300598 |
| Loss at step 100: 0.043166384100914 |
| Loss at step 150: 0.04016077518463135 |
| Loss at step 200: 0.04015089198946953 |
| Loss at step 250: 0.038076434284448624 |
| Loss at step 300: 0.032090965658426285 |
| Loss at step 350: 0.05516954883933067 |
| Loss at step 400: 0.0338265635073185 |
| Loss at step 450: 0.05055184289813042 |
| Loss at step 500: 0.0306999534368515 |
| Loss at step 550: 0.046033430844545364 |
| Loss at step 600: 0.03614010661840439 |
| Loss at step 650: 0.03513707220554352 |
| Loss at step 700: 0.055435363203287125 |
| Loss at step 750: 0.033791683614254 |
| Loss at step 800: 0.040268998593091965 |
| Loss at step 850: 0.046138305217027664 |
| Loss at step 900: 0.036269623786211014 |
| Mean training loss after epoch 135: 0.040519412690356596 |
|
|
| EPOCH: 136 |
| Loss at step 0: 0.031145719811320305 |
| Loss at step 50: 0.034706104546785355 |
| Loss at step 100: 0.04843144491314888 |
| Loss at step 150: 0.06583622097969055 |
| Loss at step 200: 0.05224967747926712 |
| Loss at step 250: 0.04415404796600342 |
| Loss at step 300: 0.03567051142454147 |
| Loss at step 350: 0.027567612007260323 |
| Loss at step 400: 0.05148737505078316 |
| Loss at step 450: 0.04981483146548271 |
| Loss at step 500: 0.04140967130661011 |
| Loss at step 550: 0.028734488412737846 |
| Loss at step 600: 0.03323344141244888 |
| Loss at step 650: 0.05476273223757744 |
| Loss at step 700: 0.04226115345954895 |
| Loss at step 750: 0.03669494390487671 |
| Loss at step 800: 0.03222007676959038 |
| Loss at step 850: 0.025839785113930702 |
| Loss at step 900: 0.03102259710431099 |
| Mean training loss after epoch 136: 0.040111369352096686 |
|
|
| EPOCH: 137 |
| Loss at step 0: 0.035307832062244415 |
| Loss at step 50: 0.035859256982803345 |
| Loss at step 100: 0.04564115032553673 |
| Loss at step 150: 0.08037211000919342 |
| Loss at step 200: 0.044677089899778366 |
| Loss at step 250: 0.04060159623622894 |
| Loss at step 300: 0.05918145924806595 |
| Loss at step 350: 0.033060211688280106 |
| Loss at step 400: 0.038067109882831573 |
| Loss at step 450: 0.03555935248732567 |
| Loss at step 500: 0.04896773397922516 |
| Loss at step 550: 0.03716621175408363 |
| Loss at step 600: 0.03253568708896637 |
| Loss at step 650: 0.0386538952589035 |
| Loss at step 700: 0.03709159418940544 |
| Loss at step 750: 0.047217704355716705 |
| Loss at step 800: 0.03762747347354889 |
| Loss at step 850: 0.033402878791093826 |
| Loss at step 900: 0.03984730318188667 |
| Mean training loss after epoch 137: 0.0398748926123354 |
|
|
| EPOCH: 138 |
| Loss at step 0: 0.0351131334900856 |
| Loss at step 50: 0.05153409764170647 |
| Loss at step 100: 0.05654660984873772 |
| Loss at step 150: 0.03171384707093239 |
| Loss at step 200: 0.044744957238435745 |
| Loss at step 250: 0.04206939414143562 |
| Loss at step 300: 0.05348724126815796 |
| Loss at step 350: 0.05097969248890877 |
| Loss at step 400: 0.0321083664894104 |
| Loss at step 450: 0.04224591329693794 |
| Loss at step 500: 0.03608119115233421 |
| Loss at step 550: 0.054212745279073715 |
| Loss at step 600: 0.03262381628155708 |
| Loss at step 650: 0.048176102340221405 |
| Loss at step 700: 0.03508109226822853 |
| Loss at step 750: 0.03597771376371384 |
| Loss at step 800: 0.04042951017618179 |
| Loss at step 850: 0.03498770296573639 |
| Loss at step 900: 0.03275318816304207 |
| Mean training loss after epoch 138: 0.04044409448316674 |
|
|
| EPOCH: 139 |
| Loss at step 0: 0.05238460749387741 |
| Loss at step 50: 0.029966481029987335 |
| Loss at step 100: 0.031206198036670685 |
| Loss at step 150: 0.03607112914323807 |
| Loss at step 200: 0.0320424921810627 |
| Loss at step 250: 0.03686009347438812 |
| Loss at step 300: 0.04129014536738396 |
| Loss at step 350: 0.05626775696873665 |
| Loss at step 400: 0.03750087693333626 |
| Loss at step 450: 0.03866837918758392 |
| Loss at step 500: 0.04019865393638611 |
| Loss at step 550: 0.030603976920247078 |
| Loss at step 600: 0.035492926836013794 |
| Loss at step 650: 0.031438861042261124 |
| Loss at step 700: 0.044327959418296814 |
| Loss at step 750: 0.029902664944529533 |
| Loss at step 800: 0.03317349776625633 |
| Loss at step 850: 0.040507715195417404 |
| Loss at step 900: 0.03306630626320839 |
| Mean training loss after epoch 139: 0.03977981074524523 |
|
|
| EPOCH: 140 |
| Loss at step 0: 0.06723352521657944 |
| Loss at step 50: 0.036315590143203735 |
| Loss at step 100: 0.048940982669591904 |
| Loss at step 150: 0.030496828258037567 |
| Loss at step 200: 0.03590953350067139 |
| Loss at step 250: 0.04100378230214119 |
| Loss at step 300: 0.035402316600084305 |
| Loss at step 350: 0.05464482679963112 |
| Loss at step 400: 0.05612416937947273 |
| Loss at step 450: 0.04811203107237816 |
| Loss at step 500: 0.03293948248028755 |
| Loss at step 550: 0.029463013634085655 |
| Loss at step 600: 0.0330667719244957 |
| Loss at step 650: 0.07245888561010361 |
| Loss at step 700: 0.03625422343611717 |
| Loss at step 750: 0.029948584735393524 |
| Loss at step 800: 0.03603658825159073 |
| Loss at step 850: 0.038006313145160675 |
| Loss at step 900: 0.04900289326906204 |
| Mean training loss after epoch 140: 0.03966290574433453 |
|
|
| EPOCH: 141 |
| Loss at step 0: 0.05389322340488434 |
| Loss at step 50: 0.03978116065263748 |
| Loss at step 100: 0.028913181275129318 |
| Loss at step 150: 0.03143258020281792 |
| Loss at step 200: 0.05115267261862755 |
| Loss at step 250: 0.04871513321995735 |
| Loss at step 300: 0.04140704870223999 |
| Loss at step 350: 0.029615212231874466 |
| Loss at step 400: 0.04913806915283203 |
| Loss at step 450: 0.05245078727602959 |
| Loss at step 500: 0.03655010834336281 |
| Loss at step 550: 0.03750515356659889 |
| Loss at step 600: 0.03648988902568817 |
| Loss at step 650: 0.04609394073486328 |
| Loss at step 700: 0.04012591764330864 |
| Loss at step 750: 0.031112534925341606 |
| Loss at step 800: 0.04239273443818092 |
| Loss at step 850: 0.0286885853856802 |
| Loss at step 900: 0.04067554697394371 |
| Mean training loss after epoch 141: 0.04047492779950216 |
|
|
| EPOCH: 142 |
| Loss at step 0: 0.0341968834400177 |
| Loss at step 50: 0.027987387031316757 |
| Loss at step 100: 0.04370037466287613 |
| Loss at step 150: 0.03250158205628395 |
| Loss at step 200: 0.03320758789777756 |
| Loss at step 250: 0.04050271958112717 |
| Loss at step 300: 0.05204188823699951 |
| Loss at step 350: 0.03396312892436981 |
| Loss at step 400: 0.03425634652376175 |
| Loss at step 450: 0.050483450293540955 |
| Loss at step 500: 0.06107752397656441 |
| Loss at step 550: 0.03602050989866257 |
| Loss at step 600: 0.030771542340517044 |
| Loss at step 650: 0.03399882838129997 |
| Loss at step 700: 0.04370621219277382 |
| Loss at step 750: 0.06062234938144684 |
| Loss at step 800: 0.030899103730916977 |
| Loss at step 850: 0.035059645771980286 |
| Loss at step 900: 0.05461152642965317 |
| Mean training loss after epoch 142: 0.03912595202769044 |
|
|
| EPOCH: 143 |
| Loss at step 0: 0.03069254197180271 |
| Loss at step 50: 0.028873268514871597 |
| Loss at step 100: 0.028288697823882103 |
| Loss at step 150: 0.044573914259672165 |
| Loss at step 200: 0.035409681499004364 |
| Loss at step 250: 0.045168306678533554 |
| Loss at step 300: 0.03995560109615326 |
| Loss at step 350: 0.0302441269159317 |
| Loss at step 400: 0.03496329486370087 |
| Loss at step 450: 0.03343826159834862 |
| Loss at step 500: 0.034524980932474136 |
| Loss at step 550: 0.03748397156596184 |
| Loss at step 600: 0.07241212576627731 |
| Loss at step 650: 0.040051836520433426 |
| Loss at step 700: 0.06440601497888565 |
| Loss at step 750: 0.032848332077264786 |
| Loss at step 800: 0.043311264365911484 |
| Loss at step 850: 0.031081559136509895 |
| Loss at step 900: 0.038504716008901596 |
| Mean training loss after epoch 143: 0.039704615202571535 |
|
|
| EPOCH: 144 |
| Loss at step 0: 0.034515753388404846 |
| Loss at step 50: 0.038643963634967804 |
| Loss at step 100: 0.04919477179646492 |
| Loss at step 150: 0.06632418185472488 |
| Loss at step 200: 0.03776159882545471 |
| Loss at step 250: 0.03524239733815193 |
| Loss at step 300: 0.03229808807373047 |
| Loss at step 350: 0.03772842139005661 |
| Loss at step 400: 0.07176003605127335 |
| Loss at step 450: 0.0438353456556797 |
| Loss at step 500: 0.03729798272252083 |
| Loss at step 550: 0.034596312791109085 |
| Loss at step 600: 0.05316857621073723 |
| Loss at step 650: 0.04838914796710014 |
| Loss at step 700: 0.03394513204693794 |
| Loss at step 750: 0.05002584308385849 |
| Loss at step 800: 0.04865674301981926 |
| Loss at step 850: 0.030374780297279358 |
| Loss at step 900: 0.03489556908607483 |
| Mean training loss after epoch 144: 0.040014465794221424 |
|
|
| EPOCH: 145 |
| Loss at step 0: 0.05304267257452011 |
| Loss at step 50: 0.03572936728596687 |
| Loss at step 100: 0.03844626620411873 |
| Loss at step 150: 0.051656000316143036 |
| Loss at step 200: 0.0419786162674427 |
| Loss at step 250: 0.03228844702243805 |
| Loss at step 300: 0.03717874363064766 |
| Loss at step 350: 0.03276830166578293 |
| Loss at step 400: 0.05304082855582237 |
| Loss at step 450: 0.0370134636759758 |
| Loss at step 500: 0.03658261150121689 |
| Loss at step 550: 0.05247374251484871 |
| Loss at step 600: 0.05243967846035957 |
| Loss at step 650: 0.035890765488147736 |
| Loss at step 700: 0.029134448617696762 |
| Loss at step 750: 0.04950794205069542 |
| Loss at step 800: 0.031324274837970734 |
| Loss at step 850: 0.04002757743000984 |
| Loss at step 900: 0.05876762419939041 |
| Mean training loss after epoch 145: 0.03966649990282588 |
|
|
| EPOCH: 146 |
| Loss at step 0: 0.03870001062750816 |
| Loss at step 50: 0.0340142585337162 |
| Loss at step 100: 0.032529689371585846 |
| Loss at step 150: 0.04962790012359619 |
| Loss at step 200: 0.06994400918483734 |
| Loss at step 250: 0.04276866093277931 |
| Loss at step 300: 0.054688457399606705 |
| Loss at step 350: 0.051464974880218506 |
| Loss at step 400: 0.0338805727660656 |
| Loss at step 450: 0.04141691327095032 |
| Loss at step 500: 0.04455658793449402 |
| Loss at step 550: 0.04114391282200813 |
| Loss at step 600: 0.03700781613588333 |
| Loss at step 650: 0.03334641084074974 |
| Loss at step 700: 0.037004128098487854 |
| Loss at step 750: 0.045164212584495544 |
| Loss at step 800: 0.04131518676877022 |
| Loss at step 850: 0.04450450465083122 |
| Loss at step 900: 0.05313121899962425 |
| Mean training loss after epoch 146: 0.040262286505624176 |
|
|
| EPOCH: 147 |
| Loss at step 0: 0.03320532664656639 |
| Loss at step 50: 0.0375794917345047 |
| Loss at step 100: 0.03766798973083496 |
| Loss at step 150: 0.051330532878637314 |
| Loss at step 200: 0.03195059671998024 |
| Loss at step 250: 0.04082850366830826 |
| Loss at step 300: 0.0481402613222599 |
| Loss at step 350: 0.030774451792240143 |
| Loss at step 400: 0.03871845453977585 |
| Loss at step 450: 0.05088238790631294 |
| Loss at step 500: 0.03520108014345169 |
| Loss at step 550: 0.04160549119114876 |
| Loss at step 600: 0.038120336830616 |
| Loss at step 650: 0.040947865694761276 |
| Loss at step 700: 0.06268515437841415 |
| Loss at step 750: 0.02965262345969677 |
| Loss at step 800: 0.0288058090955019 |
| Loss at step 850: 0.03575523570179939 |
| Loss at step 900: 0.041679564863443375 |
| Mean training loss after epoch 147: 0.0400226015303689 |
|
|
| EPOCH: 148 |
| Loss at step 0: 0.03675468638539314 |
| Loss at step 50: 0.03356386721134186 |
| Loss at step 100: 0.05130539834499359 |
| Loss at step 150: 0.058995168656110764 |
| Loss at step 200: 0.06680206954479218 |
| Loss at step 250: 0.029005855321884155 |
| Loss at step 300: 0.037632159888744354 |
| Loss at step 350: 0.032792042940855026 |
| Loss at step 400: 0.03614620864391327 |
| Loss at step 450: 0.03251120075583458 |
| Loss at step 500: 0.04506837949156761 |
| Loss at step 550: 0.034124311059713364 |
| Loss at step 600: 0.0380428172647953 |
| Loss at step 650: 0.04019821807742119 |
| Loss at step 700: 0.033369410783052444 |
| Loss at step 750: 0.05180910602211952 |
| Loss at step 800: 0.03275255858898163 |
| Loss at step 850: 0.035289227962493896 |
| Loss at step 900: 0.03412739187479019 |
| Mean training loss after epoch 148: 0.03986330867123439 |
|
|
| EPOCH: 149 |
| Loss at step 0: 0.03967485949397087 |
| Loss at step 50: 0.0405656173825264 |
| Loss at step 100: 0.049536626785993576 |
| Loss at step 150: 0.0467953085899353 |
| Loss at step 200: 0.038718242198228836 |
| Loss at step 250: 0.03391097113490105 |
| Loss at step 300: 0.027988167479634285 |
| Loss at step 350: 0.030590718612074852 |
| Loss at step 400: 0.037134964019060135 |
| Loss at step 450: 0.04939684644341469 |
| Loss at step 500: 0.028641855344176292 |
| Loss at step 550: 0.049792949110269547 |
| Loss at step 600: 0.06914780288934708 |
| Loss at step 650: 0.028473980724811554 |
| Loss at step 700: 0.04541952535510063 |
| Loss at step 750: 0.056351106613874435 |
| Loss at step 800: 0.03810013830661774 |
| Loss at step 850: 0.031036950647830963 |
| Loss at step 900: 0.02989376336336136 |
| Mean training loss after epoch 149: 0.03950454026405046 |
|
|
| EPOCH: 150 |
| Loss at step 0: 0.0497492291033268 |
| Loss at step 50: 0.04664904624223709 |
| Loss at step 100: 0.030672777444124222 |
| Loss at step 150: 0.037179891020059586 |
| Loss at step 200: 0.037141673266887665 |
| Loss at step 250: 0.033430758863687515 |
| Loss at step 300: 0.030099069699645042 |
| Loss at step 350: 0.04658585414290428 |
| Loss at step 400: 0.03212271258234978 |
| Loss at step 450: 0.037536367774009705 |
| Loss at step 500: 0.03360248729586601 |
| Loss at step 550: 0.029031341895461082 |
| Loss at step 600: 0.04348360747098923 |
| Loss at step 650: 0.032128266990184784 |
| Loss at step 700: 0.04996559023857117 |
| Loss at step 750: 0.03784072399139404 |
| Loss at step 800: 0.034901782870292664 |
| Loss at step 850: 0.036399587988853455 |
| Loss at step 900: 0.032460201531648636 |
| Mean training loss after epoch 150: 0.03932684691531508 |
|
|
| EPOCH: 151 |
| Loss at step 0: 0.038559745997190475 |
| Loss at step 50: 0.034698486328125 |
| Loss at step 100: 0.034396491944789886 |
| Loss at step 150: 0.03460189327597618 |
| Loss at step 200: 0.03432363644242287 |
| Loss at step 250: 0.044413454830646515 |
| Loss at step 300: 0.0633033886551857 |
| Loss at step 350: 0.028063831850886345 |
| Loss at step 400: 0.03233586624264717 |
| Loss at step 450: 0.050064317882061005 |
| Loss at step 500: 0.04057862237095833 |
| Loss at step 550: 0.03515806421637535 |
| Loss at step 600: 0.03948560357093811 |
| Loss at step 650: 0.03225551173090935 |
| Loss at step 700: 0.0390482172369957 |
| Loss at step 750: 0.03579146787524223 |
| Loss at step 800: 0.03390445560216904 |
| Loss at step 850: 0.05053498223423958 |
| Loss at step 900: 0.04960924759507179 |
| Mean training loss after epoch 151: 0.0394894066931946 |
|
|
| EPOCH: 152 |
| Loss at step 0: 0.04069296270608902 |
| Loss at step 50: 0.0362466461956501 |
| Loss at step 100: 0.03903277963399887 |
| Loss at step 150: 0.03854478523135185 |
| Loss at step 200: 0.030529094859957695 |
| Loss at step 250: 0.03584692254662514 |
| Loss at step 300: 0.03763008117675781 |
| Loss at step 350: 0.03907984122633934 |
| Loss at step 400: 0.032714176923036575 |
| Loss at step 450: 0.04580497741699219 |
| Loss at step 500: 0.030848801136016846 |
| Loss at step 550: 0.03909658268094063 |
| Loss at step 600: 0.0393528938293457 |
| Loss at step 650: 0.034250885248184204 |
| Loss at step 700: 0.039407070726156235 |
| Loss at step 750: 0.06522706151008606 |
| Loss at step 800: 0.02698972448706627 |
| Loss at step 850: 0.033097878098487854 |
| Loss at step 900: 0.039493318647146225 |
| Mean training loss after epoch 152: 0.039979385349081395 |
|
|
| EPOCH: 153 |
| Loss at step 0: 0.03830374777317047 |
| Loss at step 50: 0.03738465532660484 |
| Loss at step 100: 0.03408047929406166 |
| Loss at step 150: 0.08162634074687958 |
| Loss at step 200: 0.047909438610076904 |
| Loss at step 250: 0.04590803012251854 |
| Loss at step 300: 0.03924799710512161 |
| Loss at step 350: 0.049378473311662674 |
| Loss at step 400: 0.03161664679646492 |
| Loss at step 450: 0.060903649777173996 |
| Loss at step 500: 0.0318431593477726 |
| Loss at step 550: 0.030885569751262665 |
| Loss at step 600: 0.04783306270837784 |
| Loss at step 650: 0.03933698311448097 |
| Loss at step 700: 0.03513640910387039 |
| Loss at step 750: 0.03802374005317688 |
| Loss at step 800: 0.03579352796077728 |
| Loss at step 850: 0.039611611515283585 |
| Loss at step 900: 0.032021258026361465 |
| Mean training loss after epoch 153: 0.03981151960806043 |
|
|
| EPOCH: 154 |
| Loss at step 0: 0.03201507031917572 |
| Loss at step 50: 0.03376217558979988 |
| Loss at step 100: 0.030808014795184135 |
| Loss at step 150: 0.05287822335958481 |
| Loss at step 200: 0.035732246935367584 |
| Loss at step 250: 0.03311564028263092 |
| Loss at step 300: 0.033874545246362686 |
| Loss at step 350: 0.0348997600376606 |
| Loss at step 400: 0.03685653209686279 |
| Loss at step 450: 0.034120168536901474 |
| Loss at step 500: 0.027468038722872734 |
| Loss at step 550: 0.03726698458194733 |
| Loss at step 600: 0.03796930983662605 |
| Loss at step 650: 0.052408505231142044 |
| Loss at step 700: 0.03583231568336487 |
| Loss at step 750: 0.038083259016275406 |
| Loss at step 800: 0.03880752623081207 |
| Loss at step 850: 0.03209945186972618 |
| Loss at step 900: 0.04039257392287254 |
| Mean training loss after epoch 154: 0.03945487745201537 |
|
|
| EPOCH: 155 |
| Loss at step 0: 0.03866823762655258 |
| Loss at step 50: 0.031757935881614685 |
| Loss at step 100: 0.03913172706961632 |
| Loss at step 150: 0.04874129965901375 |
| Loss at step 200: 0.05382993072271347 |
| Loss at step 250: 0.041536904871463776 |
| Loss at step 300: 0.030705511569976807 |
| Loss at step 350: 0.03425120934844017 |
| Loss at step 400: 0.03446534648537636 |
| Loss at step 450: 0.04940277710556984 |
| Loss at step 500: 0.03989547863602638 |
| Loss at step 550: 0.03524135425686836 |
| Loss at step 600: 0.045029882341623306 |
| Loss at step 650: 0.0432940311729908 |
| Loss at step 700: 0.039333123713731766 |
| Loss at step 750: 0.04918549954891205 |
| Loss at step 800: 0.052956581115722656 |
| Loss at step 850: 0.03641032800078392 |
| Loss at step 900: 0.03382905200123787 |
| Mean training loss after epoch 155: 0.039653815565380585 |
|
|
| EPOCH: 156 |
| Loss at step 0: 0.052970923483371735 |
| Loss at step 50: 0.05297328904271126 |
| Loss at step 100: 0.04523034766316414 |
| Loss at step 150: 0.052893735468387604 |
| Loss at step 200: 0.06359665840864182 |
| Loss at step 250: 0.03374951332807541 |
| Loss at step 300: 0.033798668533563614 |
| Loss at step 350: 0.03770531713962555 |
| Loss at step 400: 0.06374736875295639 |
| Loss at step 450: 0.04122309386730194 |
| Loss at step 500: 0.03582649677991867 |
| Loss at step 550: 0.03951054438948631 |
| Loss at step 600: 0.03408455848693848 |
| Loss at step 650: 0.03591909632086754 |
| Loss at step 700: 0.028890229761600494 |
| Loss at step 750: 0.050827328115701675 |
| Loss at step 800: 0.03168226033449173 |
| Loss at step 850: 0.06453513354063034 |
| Loss at step 900: 0.05081882327795029 |
| Mean training loss after epoch 156: 0.03980678563981232 |
|
|
| EPOCH: 157 |
| Loss at step 0: 0.03517033904790878 |
| Loss at step 50: 0.03029661625623703 |
| Loss at step 100: 0.04061397910118103 |
| Loss at step 150: 0.04112451896071434 |
| Loss at step 200: 0.05511137843132019 |
| Loss at step 250: 0.034754179418087006 |
| Loss at step 300: 0.03483182564377785 |
| Loss at step 350: 0.02881348878145218 |
| Loss at step 400: 0.03589953854680061 |
| Loss at step 450: 0.040302760899066925 |
| Loss at step 500: 0.02812015265226364 |
| Loss at step 550: 0.03429786115884781 |
| Loss at step 600: 0.042274147272109985 |
| Loss at step 650: 0.03722405433654785 |
| Loss at step 700: 0.06251759082078934 |
| Loss at step 750: 0.03280168026685715 |
| Loss at step 800: 0.033976029604673386 |
| Loss at step 850: 0.03572479635477066 |
| Loss at step 900: 0.03401172161102295 |
| Mean training loss after epoch 157: 0.03961787642494066 |
|
|
| EPOCH: 158 |
| Loss at step 0: 0.038926295936107635 |
| Loss at step 50: 0.06311167031526566 |
| Loss at step 100: 0.039787448942661285 |
| Loss at step 150: 0.03861629217863083 |
| Loss at step 200: 0.04008471593260765 |
| Loss at step 250: 0.03857661783695221 |
| Loss at step 300: 0.03470795974135399 |
| Loss at step 350: 0.05390108749270439 |
| Loss at step 400: 0.047356195747852325 |
| Loss at step 450: 0.04332069307565689 |
| Loss at step 500: 0.03188915550708771 |
| Loss at step 550: 0.06753725558519363 |
| Loss at step 600: 0.05447079613804817 |
| Loss at step 650: 0.039038512855768204 |
| Loss at step 700: 0.034410372376441956 |
| Loss at step 750: 0.0353931188583374 |
| Loss at step 800: 0.040720392018556595 |
| Loss at step 850: 0.030527109280228615 |
| Loss at step 900: 0.03377766162157059 |
| Mean training loss after epoch 158: 0.03973492824121007 |
|
|
| EPOCH: 159 |
| Loss at step 0: 0.04395056143403053 |
| Loss at step 50: 0.03635787218809128 |
| Loss at step 100: 0.052479084581136703 |
| Loss at step 150: 0.0331464521586895 |
| Loss at step 200: 0.03089253231883049 |
| Loss at step 250: 0.038187939673662186 |
| Loss at step 300: 0.030138317495584488 |
| Loss at step 350: 0.03846345096826553 |
| Loss at step 400: 0.049408961087465286 |
| Loss at step 450: 0.05199102312326431 |
| Loss at step 500: 0.03939513862133026 |
| Loss at step 550: 0.025730779394507408 |
| Loss at step 600: 0.040308210998773575 |
| Loss at step 650: 0.04811836779117584 |
| Loss at step 700: 0.03713226318359375 |
| Loss at step 750: 0.03334973752498627 |
| Loss at step 800: 0.054803516715765 |
| Loss at step 850: 0.038321588188409805 |
| Loss at step 900: 0.04296765848994255 |
| Mean training loss after epoch 159: 0.039721751105048254 |
|
|
| EPOCH: 160 |
| Loss at step 0: 0.042155567556619644 |
| Loss at step 50: 0.03570934012532234 |
| Loss at step 100: 0.03494516387581825 |
| Loss at step 150: 0.03171156346797943 |
| Loss at step 200: 0.03144378587603569 |
| Loss at step 250: 0.04876674711704254 |
| Loss at step 300: 0.04549702629446983 |
| Loss at step 350: 0.036487117409706116 |
| Loss at step 400: 0.04203188419342041 |
| Loss at step 450: 0.038777709007263184 |
| Loss at step 500: 0.05074477568268776 |
| Loss at step 550: 0.037049807608127594 |
| Loss at step 600: 0.03319176658987999 |
| Loss at step 650: 0.04365323483943939 |
| Loss at step 700: 0.0491318553686142 |
| Loss at step 750: 0.04497558996081352 |
| Loss at step 800: 0.050616584718227386 |
| Loss at step 850: 0.03367560729384422 |
| Loss at step 900: 0.031395260244607925 |
| Mean training loss after epoch 160: 0.03994718715071932 |
|
|
| EPOCH: 161 |
| Loss at step 0: 0.03787631168961525 |
| Loss at step 50: 0.03632188215851784 |
| Loss at step 100: 0.031822752207517624 |
| Loss at step 150: 0.03723493218421936 |
| Loss at step 200: 0.03751960024237633 |
| Loss at step 250: 0.03722744435071945 |
| Loss at step 300: 0.03279047831892967 |
| Loss at step 350: 0.038861632347106934 |
| Loss at step 400: 0.03202925994992256 |
| Loss at step 450: 0.03544725105166435 |
| Loss at step 500: 0.030634867027401924 |
| Loss at step 550: 0.034504830837249756 |
| Loss at step 600: 0.03699138015508652 |
| Loss at step 650: 0.03438907489180565 |
| Loss at step 700: 0.032897673547267914 |
| Loss at step 750: 0.039568059146404266 |
| Loss at step 800: 0.02716151438653469 |
| Loss at step 850: 0.04132404923439026 |
| Loss at step 900: 0.02859344705939293 |
| Mean training loss after epoch 161: 0.03965161397838707 |
|
|
| EPOCH: 162 |
| Loss at step 0: 0.044058606028556824 |
| Loss at step 50: 0.03533129766583443 |
| Loss at step 100: 0.03633712977170944 |
| Loss at step 150: 0.03775249794125557 |
| Loss at step 200: 0.0342252217233181 |
| Loss at step 250: 0.05188572406768799 |
| Loss at step 300: 0.0261610709130764 |
| Loss at step 350: 0.04466400295495987 |
| Loss at step 400: 0.030652252957224846 |
| Loss at step 450: 0.06206464022397995 |
| Loss at step 500: 0.041849203407764435 |
| Loss at step 550: 0.04211672022938728 |
| Loss at step 600: 0.04404137656092644 |
| Loss at step 650: 0.037888288497924805 |
| Loss at step 700: 0.03815712034702301 |
| Loss at step 750: 0.0420524999499321 |
| Loss at step 800: 0.0497308187186718 |
| Loss at step 850: 0.03643658757209778 |
| Loss at step 900: 0.045854780822992325 |
| Mean training loss after epoch 162: 0.03955368834065158 |
|
|
| EPOCH: 163 |
| Loss at step 0: 0.037637751549482346 |
| Loss at step 50: 0.03754173591732979 |
| Loss at step 100: 0.030969874933362007 |
| Loss at step 150: 0.04777289181947708 |
| Loss at step 200: 0.034243032336235046 |
| Loss at step 250: 0.030788376927375793 |
| Loss at step 300: 0.032340116798877716 |
| Loss at step 350: 0.06834275275468826 |
| Loss at step 400: 0.03906245157122612 |
| Loss at step 450: 0.04159681126475334 |
| Loss at step 500: 0.04897071421146393 |
| Loss at step 550: 0.032994478940963745 |
| Loss at step 600: 0.05192587152123451 |
| Loss at step 650: 0.05311141908168793 |
| Loss at step 700: 0.05736926198005676 |
| Loss at step 750: 0.03613514453172684 |
| Loss at step 800: 0.05063817650079727 |
| Loss at step 850: 0.04703342542052269 |
| Loss at step 900: 0.03148123249411583 |
| Mean training loss after epoch 163: 0.03973168544153542 |
|
|
| EPOCH: 164 |
| Loss at step 0: 0.03522387892007828 |
| Loss at step 50: 0.0354304239153862 |
| Loss at step 100: 0.034576959908008575 |
| Loss at step 150: 0.04489249363541603 |
| Loss at step 200: 0.04102586582303047 |
| Loss at step 250: 0.04082818701863289 |
| Loss at step 300: 0.048556774854660034 |
| Loss at step 350: 0.028994642198085785 |
| Loss at step 400: 0.03966813161969185 |
| Loss at step 450: 0.05221426114439964 |
| Loss at step 500: 0.04641151428222656 |
| Loss at step 550: 0.051005858927965164 |
| Loss at step 600: 0.04715345799922943 |
| Loss at step 650: 0.06296427547931671 |
| Loss at step 700: 0.036891594529151917 |
| Loss at step 750: 0.030493998900055885 |
| Loss at step 800: 0.037916600704193115 |
| Loss at step 850: 0.04149070009589195 |
| Loss at step 900: 0.03550049662590027 |
| Mean training loss after epoch 164: 0.0396854249514274 |
|
|
| EPOCH: 165 |
| Loss at step 0: 0.042220719158649445 |
| Loss at step 50: 0.03374775871634483 |
| Loss at step 100: 0.037937846034765244 |
| Loss at step 150: 0.07947231829166412 |
| Loss at step 200: 0.061438754200935364 |
| Loss at step 250: 0.03606860339641571 |
| Loss at step 300: 0.041866056621074677 |
| Loss at step 350: 0.04028409719467163 |
| Loss at step 400: 0.03585618734359741 |
| Loss at step 450: 0.05177586153149605 |
| Loss at step 500: 0.03241375833749771 |
| Loss at step 550: 0.040101245045661926 |
| Loss at step 600: 0.03462972491979599 |
| Loss at step 650: 0.03558705002069473 |
| Loss at step 700: 0.05475455895066261 |
| Loss at step 750: 0.04315561428666115 |
| Loss at step 800: 0.034266598522663116 |
| Loss at step 850: 0.03024034947156906 |
| Loss at step 900: 0.050545837730169296 |
| Mean training loss after epoch 165: 0.03933941557217064 |
|
|
| EPOCH: 166 |
| Loss at step 0: 0.03422791138291359 |
| Loss at step 50: 0.0342918299138546 |
| Loss at step 100: 0.040826451033353806 |
| Loss at step 150: 0.04773464798927307 |
| Loss at step 200: 0.051334526389837265 |
| Loss at step 250: 0.030052201822400093 |
| Loss at step 300: 0.034113090485334396 |
| Loss at step 350: 0.030852695927023888 |
| Loss at step 400: 0.03177307918667793 |
| Loss at step 450: 0.02808055840432644 |
| Loss at step 500: 0.04130341112613678 |
| Loss at step 550: 0.05406184867024422 |
| Loss at step 600: 0.03329135850071907 |
| Loss at step 650: 0.039528608322143555 |
| Loss at step 700: 0.027340373024344444 |
| Loss at step 750: 0.05258629098534584 |
| Loss at step 800: 0.03488999977707863 |
| Loss at step 850: 0.03175424784421921 |
| Loss at step 900: 0.029860569164156914 |
| Mean training loss after epoch 166: 0.0395889686686652 |
|
|
| EPOCH: 167 |
| Loss at step 0: 0.038849908858537674 |
| Loss at step 50: 0.04868149012327194 |
| Loss at step 100: 0.03719170764088631 |
| Loss at step 150: 0.050611406564712524 |
| Loss at step 200: 0.0537850558757782 |
| Loss at step 250: 0.03979859873652458 |
| Loss at step 300: 0.049695856869220734 |
| Loss at step 350: 0.030528413131833076 |
| Loss at step 400: 0.026854444295167923 |
| Loss at step 450: 0.0315445214509964 |
| Loss at step 500: 0.03543714061379433 |
| Loss at step 550: 0.03773332014679909 |
| Loss at step 600: 0.03539112210273743 |
| Loss at step 650: 0.05348960682749748 |
| Loss at step 700: 0.0349346399307251 |
| Loss at step 750: 0.05097145959734917 |
| Loss at step 800: 0.03509731963276863 |
| Loss at step 850: 0.06194102019071579 |
| Loss at step 900: 0.03483329713344574 |
| Mean training loss after epoch 167: 0.03931984007914564 |
|
|
| EPOCH: 168 |
| Loss at step 0: 0.0442483015358448 |
| Loss at step 50: 0.03344307094812393 |
| Loss at step 100: 0.04714402183890343 |
| Loss at step 150: 0.036182697862386703 |
| Loss at step 200: 0.03450503572821617 |
| Loss at step 250: 0.04949743673205376 |
| Loss at step 300: 0.04699171707034111 |
| Loss at step 350: 0.03356409817934036 |
| Loss at step 400: 0.04914213716983795 |
| Loss at step 450: 0.03713100776076317 |
| Loss at step 500: 0.034673936665058136 |
| Loss at step 550: 0.039094239473342896 |
| Loss at step 600: 0.039589714258909225 |
| Loss at step 650: 0.055458154529333115 |
| Loss at step 700: 0.025569973513484 |
| Loss at step 750: 0.035783909261226654 |
| Loss at step 800: 0.0652955025434494 |
| Loss at step 850: 0.03070709854364395 |
| Loss at step 900: 0.030941976234316826 |
| Mean training loss after epoch 168: 0.03940760542644557 |
|
|
| EPOCH: 169 |
| Loss at step 0: 0.03163331374526024 |
| Loss at step 50: 0.03836061432957649 |
| Loss at step 100: 0.04068070277571678 |
| Loss at step 150: 0.030763309448957443 |
| Loss at step 200: 0.036739546805620193 |
| Loss at step 250: 0.029106061905622482 |
| Loss at step 300: 0.032068174332380295 |
| Loss at step 350: 0.032730091363191605 |
| Loss at step 400: 0.05216984823346138 |
| Loss at step 450: 0.04650463908910751 |
| Loss at step 500: 0.040319133549928665 |
| Loss at step 550: 0.030558280646800995 |
| Loss at step 600: 0.041635192930698395 |
| Loss at step 650: 0.03210076317191124 |
| Loss at step 700: 0.025039872154593468 |
| Loss at step 750: 0.04510484263300896 |
| Loss at step 800: 0.04591650143265724 |
| Loss at step 850: 0.03117387555539608 |
| Loss at step 900: 0.03997860476374626 |
| Mean training loss after epoch 169: 0.039096843947702124 |
|
|
| EPOCH: 170 |
| Loss at step 0: 0.04807845503091812 |
| Loss at step 50: 0.03216809406876564 |
| Loss at step 100: 0.052466776221990585 |
| Loss at step 150: 0.05527348443865776 |
| Loss at step 200: 0.04938875138759613 |
| Loss at step 250: 0.029139377176761627 |
| Loss at step 300: 0.03292334824800491 |
| Loss at step 350: 0.031563788652420044 |
| Loss at step 400: 0.030839821323752403 |
| Loss at step 450: 0.04812578111886978 |
| Loss at step 500: 0.03261122480034828 |
| Loss at step 550: 0.03332837298512459 |
| Loss at step 600: 0.029706457629799843 |
| Loss at step 650: 0.05334772169589996 |
| Loss at step 700: 0.038234107196331024 |
| Loss at step 750: 0.03391353040933609 |
| Loss at step 800: 0.03238190710544586 |
| Loss at step 850: 0.045458365231752396 |
| Loss at step 900: 0.031078660860657692 |
| Mean training loss after epoch 170: 0.03927428957058995 |
|
|
| EPOCH: 171 |
| Loss at step 0: 0.03604511916637421 |
| Loss at step 50: 0.02860851213335991 |
| Loss at step 100: 0.030669499188661575 |
| Loss at step 150: 0.03728640079498291 |
| Loss at step 200: 0.03493388369679451 |
| Loss at step 250: 0.03077596053481102 |
| Loss at step 300: 0.02700110711157322 |
| Loss at step 350: 0.04018703103065491 |
| Loss at step 400: 0.029676662757992744 |
| Loss at step 450: 0.031749628484249115 |
| Loss at step 500: 0.04076163098216057 |
| Loss at step 550: 0.0323580838739872 |
| Loss at step 600: 0.05565223842859268 |
| Loss at step 650: 0.036014098674058914 |
| Loss at step 700: 0.03889094665646553 |
| Loss at step 750: 0.04819978028535843 |
| Loss at step 800: 0.035391367971897125 |
| Loss at step 850: 0.04018506780266762 |
| Loss at step 900: 0.03758300840854645 |
| Mean training loss after epoch 171: 0.03903876623309561 |
|
|
| EPOCH: 172 |
| Loss at step 0: 0.040134530514478683 |
| Loss at step 50: 0.030954569578170776 |
| Loss at step 100: 0.032502155750989914 |
| Loss at step 150: 0.047871530055999756 |
| Loss at step 200: 0.03520025312900543 |
| Loss at step 250: 0.033117249608039856 |
| Loss at step 300: 0.04297640919685364 |
| Loss at step 350: 0.03170880302786827 |
| Loss at step 400: 0.03357839956879616 |
| Loss at step 450: 0.03541633114218712 |
| Loss at step 500: 0.10076101869344711 |
| Loss at step 550: 0.04049905017018318 |
| Loss at step 600: 0.03794636204838753 |
| Loss at step 650: 0.05078333243727684 |
| Loss at step 700: 0.032922934740781784 |
| Loss at step 750: 0.03392742574214935 |
| Loss at step 800: 0.034097425639629364 |
| Loss at step 850: 0.0374067984521389 |
| Loss at step 900: 0.03601730614900589 |
| Mean training loss after epoch 172: 0.039571806728871645 |
|
|
| EPOCH: 173 |
| Loss at step 0: 0.051600147038698196 |
| Loss at step 50: 0.03650055453181267 |
| Loss at step 100: 0.03364503011107445 |
| Loss at step 150: 0.06648823618888855 |
| Loss at step 200: 0.032441750168800354 |
| Loss at step 250: 0.03592945635318756 |
| Loss at step 300: 0.03565963730216026 |
| Loss at step 350: 0.03935347869992256 |
| Loss at step 400: 0.05051932856440544 |
| Loss at step 450: 0.04962983727455139 |
| Loss at step 500: 0.051245179027318954 |
| Loss at step 550: 0.03936230018734932 |
| Loss at step 600: 0.04920635372400284 |
| Loss at step 650: 0.03439559042453766 |
| Loss at step 700: 0.044927652925252914 |
| Loss at step 750: 0.05031128227710724 |
| Loss at step 800: 0.03805975615978241 |
| Loss at step 850: 0.03496336564421654 |
| Loss at step 900: 0.04957449063658714 |
| Mean training loss after epoch 173: 0.03956725788928235 |
|
|
| EPOCH: 174 |
| Loss at step 0: 0.027480291202664375 |
| Loss at step 50: 0.03697388619184494 |
| Loss at step 100: 0.03777739405632019 |
| Loss at step 150: 0.03281679376959801 |
| Loss at step 200: 0.03794809803366661 |
| Loss at step 250: 0.0320734940469265 |
| Loss at step 300: 0.0341799259185791 |
| Loss at step 350: 0.051777079701423645 |
| Loss at step 400: 0.0558072105050087 |
| Loss at step 450: 0.033157266676425934 |
| Loss at step 500: 0.04589252918958664 |
| Loss at step 550: 0.030640888959169388 |
| Loss at step 600: 0.032301343977451324 |
| Loss at step 650: 0.03360531106591225 |
| Loss at step 700: 0.03500358387827873 |
| Loss at step 750: 0.0347106046974659 |
| Loss at step 800: 0.028394857421517372 |
| Loss at step 850: 0.03694157674908638 |
| Loss at step 900: 0.04794781282544136 |
| Mean training loss after epoch 174: 0.03932879738279307 |
|
|
| EPOCH: 175 |
| Loss at step 0: 0.03555794060230255 |
| Loss at step 50: 0.02648364193737507 |
| Loss at step 100: 0.03459541127085686 |
| Loss at step 150: 0.03992311656475067 |
| Loss at step 200: 0.024893423542380333 |
| Loss at step 250: 0.047270677983760834 |
| Loss at step 300: 0.03519446775317192 |
| Loss at step 350: 0.09180052578449249 |
| Loss at step 400: 0.02917756326496601 |
| Loss at step 450: 0.034309279173612595 |
| Loss at step 500: 0.03961813449859619 |
| Loss at step 550: 0.04994247853755951 |
| Loss at step 600: 0.04075918719172478 |
| Loss at step 650: 0.031336862593889236 |
| Loss at step 700: 0.03452292084693909 |
| Loss at step 750: 0.05241398140788078 |
| Loss at step 800: 0.027623998001217842 |
| Loss at step 850: 0.02912183292210102 |
| Loss at step 900: 0.03710698336362839 |
| Mean training loss after epoch 175: 0.039300479576674735 |
|
|
| EPOCH: 176 |
| Loss at step 0: 0.036183230578899384 |
| Loss at step 50: 0.0666927918791771 |
| Loss at step 100: 0.0493197925388813 |
| Loss at step 150: 0.03803710639476776 |
| Loss at step 200: 0.03835935518145561 |
| Loss at step 250: 0.031230589374899864 |
| Loss at step 300: 0.03340180218219757 |
| Loss at step 350: 0.040427062660455704 |
| Loss at step 400: 0.03160826861858368 |
| Loss at step 450: 0.044037673622369766 |
| Loss at step 500: 0.03656746819615364 |
| Loss at step 550: 0.036755748093128204 |
| Loss at step 600: 0.059527214616537094 |
| Loss at step 650: 0.040880002081394196 |
| Loss at step 700: 0.051435764878988266 |
| Loss at step 750: 0.059903018176555634 |
| Loss at step 800: 0.03034154139459133 |
| Loss at step 850: 0.037557195872068405 |
| Loss at step 900: 0.034874048084020615 |
| Mean training loss after epoch 176: 0.040028610705598586 |
|
|
| EPOCH: 177 |
| Loss at step 0: 0.03473574295639992 |
| Loss at step 50: 0.05093587562441826 |
| Loss at step 100: 0.02982262894511223 |
| Loss at step 150: 0.030545443296432495 |
| Loss at step 200: 0.03678588569164276 |
| Loss at step 250: 0.036256153136491776 |
| Loss at step 300: 0.04075793921947479 |
| Loss at step 350: 0.03677338734269142 |
| Loss at step 400: 0.03143291920423508 |
| Loss at step 450: 0.04609094187617302 |
| Loss at step 500: 0.04759456589818001 |
| Loss at step 550: 0.03229653090238571 |
| Loss at step 600: 0.037571556866168976 |
| Loss at step 650: 0.03082422725856304 |
| Loss at step 700: 0.06481219828128815 |
| Loss at step 750: 0.06181209906935692 |
| Loss at step 800: 0.036778099834918976 |
| Loss at step 850: 0.05306657403707504 |
| Loss at step 900: 0.027700865641236305 |
| Mean training loss after epoch 177: 0.03933695467042008 |
|
|
| EPOCH: 178 |
| Loss at step 0: 0.03726204112172127 |
| Loss at step 50: 0.03066483698785305 |
| Loss at step 100: 0.03189198300242424 |
| Loss at step 150: 0.03496934846043587 |
| Loss at step 200: 0.03675440698862076 |
| Loss at step 250: 0.05385400727391243 |
| Loss at step 300: 0.06269004940986633 |
| Loss at step 350: 0.0388777069747448 |
| Loss at step 400: 0.040307316929101944 |
| Loss at step 450: 0.04561549797654152 |
| Loss at step 500: 0.032421860843896866 |
| Loss at step 550: 0.043146658688783646 |
| Loss at step 600: 0.04828475043177605 |
| Loss at step 650: 0.03869273141026497 |
| Loss at step 700: 0.027358677238225937 |
| Loss at step 750: 0.05097721889615059 |
| Loss at step 800: 0.03179521858692169 |
| Loss at step 850: 0.029768288135528564 |
| Loss at step 900: 0.03665889427065849 |
| Mean training loss after epoch 178: 0.03920851013402758 |
|
|
| EPOCH: 179 |
| Loss at step 0: 0.0432792492210865 |
| Loss at step 50: 0.03617312014102936 |
| Loss at step 100: 0.06370623409748077 |
| Loss at step 150: 0.03260226547718048 |
| Loss at step 200: 0.0277892854064703 |
| Loss at step 250: 0.032151203602552414 |
| Loss at step 300: 0.03364957869052887 |
| Loss at step 350: 0.036430612206459045 |
| Loss at step 400: 0.02578619495034218 |
| Loss at step 450: 0.03799568489193916 |
| Loss at step 500: 0.030682628974318504 |
| Loss at step 550: 0.030547630041837692 |
| Loss at step 600: 0.032164983451366425 |
| Loss at step 650: 0.048926692456007004 |
| Loss at step 700: 0.03040003776550293 |
| Loss at step 750: 0.04074382036924362 |
| Loss at step 800: 0.03304985165596008 |
| Loss at step 850: 0.046545110642910004 |
| Loss at step 900: 0.035846710205078125 |
| Mean training loss after epoch 179: 0.03970178106088819 |
|
|
| EPOCH: 180 |
| Loss at step 0: 0.041110847145318985 |
| Loss at step 50: 0.03983008489012718 |
| Loss at step 100: 0.038944534957408905 |
| Loss at step 150: 0.03114279918372631 |
| Loss at step 200: 0.035370562225580215 |
| Loss at step 250: 0.05519220605492592 |
| Loss at step 300: 0.049281731247901917 |
| Loss at step 350: 0.039435915648937225 |
| Loss at step 400: 0.046951428055763245 |
| Loss at step 450: 0.03373648226261139 |
| Loss at step 500: 0.034319255501031876 |
| Loss at step 550: 0.0509195514023304 |
| Loss at step 600: 0.0303181242197752 |
| Loss at step 650: 0.055917058140039444 |
| Loss at step 700: 0.03461319953203201 |
| Loss at step 750: 0.07615417242050171 |
| Loss at step 800: 0.08306694775819778 |
| Loss at step 850: 0.039755940437316895 |
| Loss at step 900: 0.05004990100860596 |
| Mean training loss after epoch 180: 0.04022076416776569 |
|
|
| EPOCH: 181 |
| Loss at step 0: 0.029874665662646294 |
| Loss at step 50: 0.029927344992756844 |
| Loss at step 100: 0.03604910150170326 |
| Loss at step 150: 0.03975125402212143 |
| Loss at step 200: 0.03120443783700466 |
| Loss at step 250: 0.04868396371603012 |
| Loss at step 300: 0.049526773393154144 |
| Loss at step 350: 0.03383695334196091 |
| Loss at step 400: 0.034840282052755356 |
| Loss at step 450: 0.05584263056516647 |
| Loss at step 500: 0.0310691948980093 |
| Loss at step 550: 0.04415060952305794 |
| Loss at step 600: 0.032994214445352554 |
| Loss at step 650: 0.03248259797692299 |
| Loss at step 700: 0.03337256610393524 |
| Loss at step 750: 0.035838667303323746 |
| Loss at step 800: 0.050576042383909225 |
| Loss at step 850: 0.06149858236312866 |
| Loss at step 900: 0.035496558994054794 |
| Mean training loss after epoch 181: 0.03953436933267218 |
|
|
| EPOCH: 182 |
| Loss at step 0: 0.03431776538491249 |
| Loss at step 50: 0.04044733941555023 |
| Loss at step 100: 0.049678489565849304 |
| Loss at step 150: 0.052443064749240875 |
| Loss at step 200: 0.03474302589893341 |
| Loss at step 250: 0.03703489899635315 |
| Loss at step 300: 0.04352117329835892 |
| Loss at step 350: 0.03962820768356323 |
| Loss at step 400: 0.037225913256406784 |
| Loss at step 450: 0.05509385094046593 |
| Loss at step 500: 0.03221511468291283 |
| Loss at step 550: 0.04929589107632637 |
| Loss at step 600: 0.0626567006111145 |
| Loss at step 650: 0.02498018741607666 |
| Loss at step 700: 0.03269823268055916 |
| Loss at step 750: 0.04034951329231262 |
| Loss at step 800: 0.038959357887506485 |
| Loss at step 850: 0.045052941888570786 |
| Loss at step 900: 0.030315130949020386 |
| Mean training loss after epoch 182: 0.03948396699292573 |
|
|
| EPOCH: 183 |
| Loss at step 0: 0.05397326871752739 |
| Loss at step 50: 0.03509332984685898 |
| Loss at step 100: 0.04604775458574295 |
| Loss at step 150: 0.03923487663269043 |
| Loss at step 200: 0.04648161679506302 |
| Loss at step 250: 0.07260828465223312 |
| Loss at step 300: 0.04022159054875374 |
| Loss at step 350: 0.04950942099094391 |
| Loss at step 400: 0.05045710504055023 |
| Loss at step 450: 0.0487130731344223 |
| Loss at step 500: 0.03563311696052551 |
| Loss at step 550: 0.06734095513820648 |
| Loss at step 600: 0.031074846163392067 |
| Loss at step 650: 0.05064196512103081 |
| Loss at step 700: 0.06535758823156357 |
| Loss at step 750: 0.04022606834769249 |
| Loss at step 800: 0.0402403250336647 |
| Loss at step 850: 0.035112425684928894 |
| Loss at step 900: 0.048231590539216995 |
| Mean training loss after epoch 183: 0.03999095734979298 |
|
|
| EPOCH: 184 |
| Loss at step 0: 0.06572314351797104 |
| Loss at step 50: 0.040734417736530304 |
| Loss at step 100: 0.033031612634658813 |
| Loss at step 150: 0.04663529992103577 |
| Loss at step 200: 0.028599590063095093 |
| Loss at step 250: 0.03933103755116463 |
| Loss at step 300: 0.029691778123378754 |
| Loss at step 350: 0.034657254815101624 |
| Loss at step 400: 0.0519569106400013 |
| Loss at step 450: 0.04233112558722496 |
| Loss at step 500: 0.03395779803395271 |
| Loss at step 550: 0.036949314177036285 |
| Loss at step 600: 0.036554135382175446 |
| Loss at step 650: 0.030880143865942955 |
| Loss at step 700: 0.044782429933547974 |
| Loss at step 750: 0.04911411926150322 |
| Loss at step 800: 0.037866562604904175 |
| Loss at step 850: 0.06262623518705368 |
| Loss at step 900: 0.03039371594786644 |
| Mean training loss after epoch 184: 0.039548526968814925 |
|
|
| EPOCH: 185 |
| Loss at step 0: 0.0805806964635849 |
| Loss at step 50: 0.027917854487895966 |
| Loss at step 100: 0.03495066240429878 |
| Loss at step 150: 0.027538307011127472 |
| Loss at step 200: 0.03303145989775658 |
| Loss at step 250: 0.031799983233213425 |
| Loss at step 300: 0.06004394590854645 |
| Loss at step 350: 0.030346397310495377 |
| Loss at step 400: 0.031851932406425476 |
| Loss at step 450: 0.05824219435453415 |
| Loss at step 500: 0.028569376096129417 |
| Loss at step 550: 0.027987563982605934 |
| Loss at step 600: 0.042291879653930664 |
| Loss at step 650: 0.03323879837989807 |
| Loss at step 700: 0.059478759765625 |
| Loss at step 750: 0.03162425383925438 |
| Loss at step 800: 0.033148087561130524 |
| Loss at step 850: 0.03013218380510807 |
| Loss at step 900: 0.0303540900349617 |
| Mean training loss after epoch 185: 0.03991742543518734 |
|
|
| EPOCH: 186 |
| Loss at step 0: 0.03823243826627731 |
| Loss at step 50: 0.04821484163403511 |
| Loss at step 100: 0.0335773304104805 |
| Loss at step 150: 0.03713402897119522 |
| Loss at step 200: 0.059686921536922455 |
| Loss at step 250: 0.030698001384735107 |
| Loss at step 300: 0.05005711689591408 |
| Loss at step 350: 0.03683238849043846 |
| Loss at step 400: 0.031736306846141815 |
| Loss at step 450: 0.02648918889462948 |
| Loss at step 500: 0.03295756131410599 |
| Loss at step 550: 0.055000800639390945 |
| Loss at step 600: 0.03838678076863289 |
| Loss at step 650: 0.03734767809510231 |
| Loss at step 700: 0.034189265221357346 |
| Loss at step 750: 0.031962718814611435 |
| Loss at step 800: 0.04328812658786774 |
| Loss at step 850: 0.04030853882431984 |
| Loss at step 900: 0.030331796035170555 |
| Mean training loss after epoch 186: 0.04037082401602698 |
|
|
| EPOCH: 187 |
| Loss at step 0: 0.03479158505797386 |
| Loss at step 50: 0.027288708835840225 |
| Loss at step 100: 0.03628791123628616 |
| Loss at step 150: 0.04474860802292824 |
| Loss at step 200: 0.046368908137083054 |
| Loss at step 250: 0.03773191571235657 |
| Loss at step 300: 0.03852430358529091 |
| Loss at step 350: 0.043728701770305634 |
| Loss at step 400: 0.03673646226525307 |
| Loss at step 450: 0.04937160760164261 |
| Loss at step 500: 0.03505728766322136 |
| Loss at step 550: 0.03285621479153633 |
| Loss at step 600: 0.03919614851474762 |
| Loss at step 650: 0.03945038095116615 |
| Loss at step 700: 0.03538677096366882 |
| Loss at step 750: 0.044934406876564026 |
| Loss at step 800: 0.03320298716425896 |
| Loss at step 850: 0.04768715798854828 |
| Loss at step 900: 0.03354670852422714 |
| Mean training loss after epoch 187: 0.039411708287624664 |
|
|
| EPOCH: 188 |
| Loss at step 0: 0.03753814473748207 |
| Loss at step 50: 0.04097013920545578 |
| Loss at step 100: 0.030687859281897545 |
| Loss at step 150: 0.0322105698287487 |
| Loss at step 200: 0.03676756098866463 |
| Loss at step 250: 0.038250818848609924 |
| Loss at step 300: 0.03754312917590141 |
| Loss at step 350: 0.03725780174136162 |
| Loss at step 400: 0.047191228717565536 |
| Loss at step 450: 0.06354070454835892 |
| Loss at step 500: 0.032046183943748474 |
| Loss at step 550: 0.04977399855852127 |
| Loss at step 600: 0.036721959710121155 |
| Loss at step 650: 0.05434718355536461 |
| Loss at step 700: 0.038213517516851425 |
| Loss at step 750: 0.024722011759877205 |
| Loss at step 800: 0.02971751242876053 |
| Loss at step 850: 0.0317879281938076 |
| Loss at step 900: 0.04744962230324745 |
| Mean training loss after epoch 188: 0.04035358981235322 |
|
|
| EPOCH: 189 |
| Loss at step 0: 0.040479883551597595 |
| Loss at step 50: 0.036622148007154465 |
| Loss at step 100: 0.034205641597509384 |
| Loss at step 150: 0.04935614764690399 |
| Loss at step 200: 0.029448114335536957 |
| Loss at step 250: 0.03502771258354187 |
| Loss at step 300: 0.05513657256960869 |
| Loss at step 350: 0.04774866998195648 |
| Loss at step 400: 0.0404169000685215 |
| Loss at step 450: 0.06562421470880508 |
| Loss at step 500: 0.035340260714292526 |
| Loss at step 550: 0.0556841604411602 |
| Loss at step 600: 0.030720310285687447 |
| Loss at step 650: 0.036314480006694794 |
| Loss at step 700: 0.03438333421945572 |
| Loss at step 750: 0.033112671226263046 |
| Loss at step 800: 0.03592928498983383 |
| Loss at step 850: 0.03132610768079758 |
| Loss at step 900: 0.03413686528801918 |
| Mean training loss after epoch 189: 0.03904189117379916 |
|
|
| EPOCH: 190 |
| Loss at step 0: 0.04827755317091942 |
| Loss at step 50: 0.029142968356609344 |
| Loss at step 100: 0.036558929830789566 |
| Loss at step 150: 0.0487666018307209 |
| Loss at step 200: 0.052352845668792725 |
| Loss at step 250: 0.05458885058760643 |
| Loss at step 300: 0.03482016548514366 |
| Loss at step 350: 0.029485084116458893 |
| Loss at step 400: 0.039250362664461136 |
| Loss at step 450: 0.051643602550029755 |
| Loss at step 500: 0.03167468681931496 |
| Loss at step 550: 0.04167293757200241 |
| Loss at step 600: 0.03159556910395622 |
| Loss at step 650: 0.04445411637425423 |
| Loss at step 700: 0.04958534613251686 |
| Loss at step 750: 0.03268647566437721 |
| Loss at step 800: 0.06649065017700195 |
| Loss at step 850: 0.06429065763950348 |
| Loss at step 900: 0.04084879532456398 |
| Mean training loss after epoch 190: 0.03958941305250819 |
|
|
| EPOCH: 191 |
| Loss at step 0: 0.035234544426202774 |
| Loss at step 50: 0.037936512380838394 |
| Loss at step 100: 0.043479036539793015 |
| Loss at step 150: 0.03322908282279968 |
| Loss at step 200: 0.049596164375543594 |
| Loss at step 250: 0.04965569078922272 |
| Loss at step 300: 0.0495617538690567 |
| Loss at step 350: 0.03661232441663742 |
| Loss at step 400: 0.03508910536766052 |
| Loss at step 450: 0.036798376590013504 |
| Loss at step 500: 0.054994553327560425 |
| Loss at step 550: 0.03120352514088154 |
| Loss at step 600: 0.048640795052051544 |
| Loss at step 650: 0.0348392091691494 |
| Loss at step 700: 0.055915024131536484 |
| Loss at step 750: 0.05253498628735542 |
| Loss at step 800: 0.031080756336450577 |
| Loss at step 850: 0.04723038524389267 |
| Loss at step 900: 0.03465425223112106 |
| Mean training loss after epoch 191: 0.039508505227929876 |
|
|
| EPOCH: 192 |
| Loss at step 0: 0.028439735993742943 |
| Loss at step 50: 0.04162871092557907 |
| Loss at step 100: 0.031034814193844795 |
| Loss at step 150: 0.05161105841398239 |
| Loss at step 200: 0.03534523397684097 |
| Loss at step 250: 0.03421344980597496 |
| Loss at step 300: 0.03441846743226051 |
| Loss at step 350: 0.05163728818297386 |
| Loss at step 400: 0.038883548229932785 |
| Loss at step 450: 0.031683698296546936 |
| Loss at step 500: 0.044283077120780945 |
| Loss at step 550: 0.0333872064948082 |
| Loss at step 600: 0.03577182814478874 |
| Loss at step 650: 0.035094004124403 |
| Loss at step 700: 0.052114710211753845 |
| Loss at step 750: 0.028549348935484886 |
| Loss at step 800: 0.031279418617486954 |
| Loss at step 850: 0.04974686726927757 |
| Loss at step 900: 0.06214940547943115 |
| Mean training loss after epoch 192: 0.03946408471747884 |
|
|
| EPOCH: 193 |
| Loss at step 0: 0.02789539098739624 |
| Loss at step 50: 0.04128626734018326 |
| Loss at step 100: 0.031125077977776527 |
| Loss at step 150: 0.054317064583301544 |
| Loss at step 200: 0.035284653306007385 |
| Loss at step 250: 0.030020158737897873 |
| Loss at step 300: 0.0426630899310112 |
| Loss at step 350: 0.02814929373562336 |
| Loss at step 400: 0.03576153889298439 |
| Loss at step 450: 0.030557071790099144 |
| Loss at step 500: 0.07169098407030106 |
| Loss at step 550: 0.046338628977537155 |
| Loss at step 600: 0.07676282525062561 |
| Loss at step 650: 0.027175115421414375 |
| Loss at step 700: 0.03226439654827118 |
| Loss at step 750: 0.05648811534047127 |
| Loss at step 800: 0.05417701229453087 |
| Loss at step 850: 0.034749649465084076 |
| Loss at step 900: 0.05131750926375389 |
| Mean training loss after epoch 193: 0.0390426356639308 |
|
|
| EPOCH: 194 |
| Loss at step 0: 0.031341664493083954 |
| Loss at step 50: 0.034224752336740494 |
| Loss at step 100: 0.030920159071683884 |
| Loss at step 150: 0.03192133828997612 |
| Loss at step 200: 0.05085880681872368 |
| Loss at step 250: 0.030830208212137222 |
| Loss at step 300: 0.032128382474184036 |
| Loss at step 350: 0.03746500611305237 |
| Loss at step 400: 0.03606203570961952 |
| Loss at step 450: 0.034037210047245026 |
| Loss at step 500: 0.032156843692064285 |
| Loss at step 550: 0.05800786241889 |
| Loss at step 600: 0.03398191183805466 |
| Loss at step 650: 0.034924451261758804 |
| Loss at step 700: 0.03250403329730034 |
| Loss at step 750: 0.02987152710556984 |
| Loss at step 800: 0.03198182210326195 |
| Loss at step 850: 0.03203301131725311 |
| Loss at step 900: 0.027829548344016075 |
| Mean training loss after epoch 194: 0.03956125608123124 |
|
|
| EPOCH: 195 |
| Loss at step 0: 0.03269430249929428 |
| Loss at step 50: 0.040413130074739456 |
| Loss at step 100: 0.037335216999053955 |
| Loss at step 150: 0.037506330758333206 |
| Loss at step 200: 0.04935978725552559 |
| Loss at step 250: 0.04005909338593483 |
| Loss at step 300: 0.051902711391448975 |
| Loss at step 350: 0.04994789883494377 |
| Loss at step 400: 0.06731879711151123 |
| Loss at step 450: 0.041173528879880905 |
| Loss at step 500: 0.03356955945491791 |
| Loss at step 550: 0.03218766674399376 |
| Loss at step 600: 0.03400886803865433 |
| Loss at step 650: 0.046790432184934616 |
| Loss at step 700: 0.0261378213763237 |
| Loss at step 750: 0.04084371030330658 |
| Loss at step 800: 0.03878629207611084 |
| Loss at step 850: 0.031192457303404808 |
| Loss at step 900: 0.04036341607570648 |
| Mean training loss after epoch 195: 0.03942633575714156 |
|
|
| EPOCH: 196 |
| Loss at step 0: 0.05267094075679779 |
| Loss at step 50: 0.051652032881975174 |
| Loss at step 100: 0.030373184010386467 |
| Loss at step 150: 0.04002588614821434 |
| Loss at step 200: 0.033278606832027435 |
| Loss at step 250: 0.03507424145936966 |
| Loss at step 300: 0.037266410887241364 |
| Loss at step 350: 0.039247218519449234 |
| Loss at step 400: 0.029236620292067528 |
| Loss at step 450: 0.03595879673957825 |
| Loss at step 500: 0.042438700795173645 |
| Loss at step 550: 0.04551468789577484 |
| Loss at step 600: 0.04240725561976433 |
| Loss at step 650: 0.03342801705002785 |
| Loss at step 700: 0.0306172464042902 |
| Loss at step 750: 0.03251747041940689 |
| Loss at step 800: 0.037676624953746796 |
| Loss at step 850: 0.03884509578347206 |
| Loss at step 900: 0.05178022384643555 |
| Mean training loss after epoch 196: 0.03961125413166371 |
|
|
| EPOCH: 197 |
| Loss at step 0: 0.03382141515612602 |
| Loss at step 50: 0.036528680473566055 |
| Loss at step 100: 0.03256684169173241 |
| Loss at step 150: 0.04024352878332138 |
| Loss at step 200: 0.03290350362658501 |
| Loss at step 250: 0.05601493641734123 |
| Loss at step 300: 0.03842312470078468 |
| Loss at step 350: 0.03384149447083473 |
| Loss at step 400: 0.03558668866753578 |
| Loss at step 450: 0.03347795829176903 |
| Loss at step 500: 0.03716802969574928 |
| Loss at step 550: 0.03866909071803093 |
| Loss at step 600: 0.04315279796719551 |
| Loss at step 650: 0.030964933335781097 |
| Loss at step 700: 0.031178439036011696 |
| Loss at step 750: 0.030622651800513268 |
| Loss at step 800: 0.03555544838309288 |
| Loss at step 850: 0.04361952841281891 |
| Loss at step 900: 0.029841801151633263 |
| Mean training loss after epoch 197: 0.03926740740058519 |
|
|
| EPOCH: 198 |
| Loss at step 0: 0.03077380359172821 |
| Loss at step 50: 0.0331268385052681 |
| Loss at step 100: 0.028607361018657684 |
| Loss at step 150: 0.027219180017709732 |
| Loss at step 200: 0.03859898820519447 |
| Loss at step 250: 0.043153245002031326 |
| Loss at step 300: 0.037560559809207916 |
| Loss at step 350: 0.03113088570535183 |
| Loss at step 400: 0.05178512632846832 |
| Loss at step 450: 0.03197965398430824 |
| Loss at step 500: 0.036399465054273605 |
| Loss at step 550: 0.039263416081666946 |
| Loss at step 600: 0.03154432773590088 |
| Loss at step 650: 0.033167921006679535 |
| Loss at step 700: 0.03971502184867859 |
| Loss at step 750: 0.029375536367297173 |
| Loss at step 800: 0.08838345855474472 |
| Loss at step 850: 0.054228276014328 |
| Loss at step 900: 0.034299567341804504 |
| Mean training loss after epoch 198: 0.039270908790213596 |
|
|
| EPOCH: 199 |
| Loss at step 0: 0.032876938581466675 |
| Loss at step 50: 0.04957316815853119 |
| Loss at step 100: 0.04838715121150017 |
| Loss at step 150: 0.06222357973456383 |
| Loss at step 200: 0.03767170011997223 |
| Loss at step 250: 0.03373800590634346 |
| Loss at step 300: 0.03463589772582054 |
| Loss at step 350: 0.036685265600681305 |
| Loss at step 400: 0.037816766649484634 |
| Loss at step 450: 0.04994241148233414 |
| Loss at step 500: 0.05309994891285896 |
| Loss at step 550: 0.050136446952819824 |
| Loss at step 600: 0.034272655844688416 |
| Loss at step 650: 0.04841490089893341 |
| Loss at step 700: 0.032551348209381104 |
| Loss at step 750: 0.04803447797894478 |
| Loss at step 800: 0.027424441650509834 |
| Loss at step 850: 0.03509342297911644 |
| Loss at step 900: 0.04698752239346504 |
| Mean training loss after epoch 199: 0.039368446850811624 |
|
|
| EPOCH: 200 |
| Loss at step 0: 0.033516522496938705 |
| Loss at step 50: 0.033314839005470276 |
| Loss at step 100: 0.028926312923431396 |
| Loss at step 150: 0.045188531279563904 |
| Loss at step 200: 0.03707105666399002 |
| Loss at step 250: 0.04511125758290291 |
| Loss at step 300: 0.041642624884843826 |
| Loss at step 350: 0.03727297484874725 |
| Loss at step 400: 0.04820782691240311 |
| Loss at step 450: 0.05873133987188339 |
| Loss at step 500: 0.03943607211112976 |
| Loss at step 550: 0.03903554007411003 |
| Loss at step 600: 0.03445577993988991 |
| Loss at step 650: 0.03482406958937645 |
| Loss at step 700: 0.03471191227436066 |
| Loss at step 750: 0.0357685312628746 |
| Loss at step 800: 0.046298760920763016 |
| Loss at step 850: 0.03532588854432106 |
| Loss at step 900: 0.04996367171406746 |
| Mean training loss after epoch 200: 0.03945373677646618 |
|
|
| EPOCH: 201 |
| Loss at step 0: 0.035214416682720184 |
| Loss at step 50: 0.035346150398254395 |
| Loss at step 100: 0.034103572368621826 |
| Loss at step 150: 0.038458410650491714 |
| Loss at step 200: 0.038615066558122635 |
| Loss at step 250: 0.03179283067584038 |
| Loss at step 300: 0.035539235919713974 |
| Loss at step 350: 0.03436723351478577 |
| Loss at step 400: 0.05592014268040657 |
| Loss at step 450: 0.02996472455561161 |
| Loss at step 500: 0.03577533736824989 |
| Loss at step 550: 0.025224490091204643 |
| Loss at step 600: 0.03710426017642021 |
| Loss at step 650: 0.04781566187739372 |
| Loss at step 700: 0.03749268129467964 |
| Loss at step 750: 0.03389711305499077 |
| Loss at step 800: 0.03144014626741409 |
| Loss at step 850: 0.038093969225883484 |
| Loss at step 900: 0.04576469212770462 |
| Mean training loss after epoch 201: 0.03969399575620636 |
|
|
| EPOCH: 202 |
| Loss at step 0: 0.0476631335914135 |
| Loss at step 50: 0.031282223761081696 |
| Loss at step 100: 0.044516369700431824 |
| Loss at step 150: 0.06186264380812645 |
| Loss at step 200: 0.0393553301692009 |
| Loss at step 250: 0.03279627859592438 |
| Loss at step 300: 0.0629049614071846 |
| Loss at step 350: 0.030579999089241028 |
| Loss at step 400: 0.031155675649642944 |
| Loss at step 450: 0.06583017855882645 |
| Loss at step 500: 0.05303068459033966 |
| Loss at step 550: 0.036140043288469315 |
| Loss at step 600: 0.03231840208172798 |
| Loss at step 650: 0.05249636247754097 |
| Loss at step 700: 0.051017142832279205 |
| Loss at step 750: 0.04280295595526695 |
| Loss at step 800: 0.03303103521466255 |
| Loss at step 850: 0.03351632133126259 |
| Loss at step 900: 0.038658663630485535 |
| Mean training loss after epoch 202: 0.03924750817268451 |
|
|
| EPOCH: 203 |
| Loss at step 0: 0.04230416938662529 |
| Loss at step 50: 0.03197874128818512 |
| Loss at step 100: 0.034516096115112305 |
| Loss at step 150: 0.0965784564614296 |
| Loss at step 200: 0.036621060222387314 |
| Loss at step 250: 0.033483728766441345 |
| Loss at step 300: 0.05883985385298729 |
| Loss at step 350: 0.03732655197381973 |
| Loss at step 400: 0.03222108259797096 |
| Loss at step 450: 0.0372038297355175 |
| Loss at step 500: 0.031740378588438034 |
| Loss at step 550: 0.032624147832393646 |
| Loss at step 600: 0.035175345838069916 |
| Loss at step 650: 0.0395626574754715 |
| Loss at step 700: 0.04586372524499893 |
| Loss at step 750: 0.032561711966991425 |
| Loss at step 800: 0.06468833237886429 |
| Loss at step 850: 0.03629869595170021 |
| Loss at step 900: 0.030181996524333954 |
| Mean training loss after epoch 203: 0.03954556779161509 |
|
|
| EPOCH: 204 |
| Loss at step 0: 0.029719999060034752 |
| Loss at step 50: 0.03435388207435608 |
| Loss at step 100: 0.035306770354509354 |
| Loss at step 150: 0.04551118612289429 |
| Loss at step 200: 0.03435719758272171 |
| Loss at step 250: 0.034615252166986465 |
| Loss at step 300: 0.030884413048624992 |
| Loss at step 350: 0.050709690898656845 |
| Loss at step 400: 0.038177862763404846 |
| Loss at step 450: 0.031270384788513184 |
| Loss at step 500: 0.035266853868961334 |
| Loss at step 550: 0.07006863504648209 |
| Loss at step 600: 0.03182707354426384 |
| Loss at step 650: 0.05542439594864845 |
| Loss at step 700: 0.04853573441505432 |
| Loss at step 750: 0.027383018285036087 |
| Loss at step 800: 0.03046734444797039 |
| Loss at step 850: 0.04207098111510277 |
| Loss at step 900: 0.03459654748439789 |
| Mean training loss after epoch 204: 0.03929966339456247 |
|
|
| EPOCH: 205 |
| Loss at step 0: 0.0312785767018795 |
| Loss at step 50: 0.036261215806007385 |
| Loss at step 100: 0.04925072193145752 |
| Loss at step 150: 0.04340329393744469 |
| Loss at step 200: 0.039639346301555634 |
| Loss at step 250: 0.03776752948760986 |
| Loss at step 300: 0.042972952127456665 |
| Loss at step 350: 0.030083203688263893 |
| Loss at step 400: 0.04385426640510559 |
| Loss at step 450: 0.0570792555809021 |
| Loss at step 500: 0.05148709565401077 |
| Loss at step 550: 0.0342877060174942 |
| Loss at step 600: 0.037329480051994324 |
| Loss at step 650: 0.03023405931890011 |
| Loss at step 700: 0.027055177837610245 |
| Loss at step 750: 0.026842977851629257 |
| Loss at step 800: 0.050824299454689026 |
| Loss at step 850: 0.032287370413541794 |
| Loss at step 900: 0.02937036007642746 |
| Mean training loss after epoch 205: 0.03916449878197997 |
|
|
| EPOCH: 206 |
| Loss at step 0: 0.03259656950831413 |
| Loss at step 50: 0.04985092580318451 |
| Loss at step 100: 0.02949802204966545 |
| Loss at step 150: 0.036611951887607574 |
| Loss at step 200: 0.039992108941078186 |
| Loss at step 250: 0.0345536433160305 |
| Loss at step 300: 0.032152608036994934 |
| Loss at step 350: 0.04637967050075531 |
| Loss at step 400: 0.02919534221291542 |
| Loss at step 450: 0.031057648360729218 |
| Loss at step 500: 0.03315690532326698 |
| Loss at step 550: 0.05003923177719116 |
| Loss at step 600: 0.045779258012771606 |
| Loss at step 650: 0.02811567671597004 |
| Loss at step 700: 0.038607172667980194 |
| Loss at step 750: 0.04060652107000351 |
| Loss at step 800: 0.0453539676964283 |
| Loss at step 850: 0.035857975482940674 |
| Loss at step 900: 0.031072260811924934 |
| Mean training loss after epoch 206: 0.038921635300079895 |
|
|
| EPOCH: 207 |
| Loss at step 0: 0.03965787962079048 |
| Loss at step 50: 0.0446898527443409 |
| Loss at step 100: 0.028009677305817604 |
| Loss at step 150: 0.038405463099479675 |
| Loss at step 200: 0.06563003361225128 |
| Loss at step 250: 0.03391384705901146 |
| Loss at step 300: 0.04727596417069435 |
| Loss at step 350: 0.03483771160244942 |
| Loss at step 400: 0.036268409341573715 |
| Loss at step 450: 0.030389249324798584 |
| Loss at step 500: 0.0310251172631979 |
| Loss at step 550: 0.0379030704498291 |
| Loss at step 600: 0.040424078702926636 |
| Loss at step 650: 0.0521022230386734 |
| Loss at step 700: 0.033831410109996796 |
| Loss at step 750: 0.033530574291944504 |
| Loss at step 800: 0.036986444145441055 |
| Loss at step 850: 0.032420702278614044 |
| Loss at step 900: 0.03470515087246895 |
| Mean training loss after epoch 207: 0.03863498132858576 |
|
|
| EPOCH: 208 |
| Loss at step 0: 0.033557191491127014 |
| Loss at step 50: 0.03412749990820885 |
| Loss at step 100: 0.048487868160009384 |
| Loss at step 150: 0.05572287738323212 |
| Loss at step 200: 0.04506176337599754 |
| Loss at step 250: 0.06600682437419891 |
| Loss at step 300: 0.05004393309354782 |
| Loss at step 350: 0.05113781616091728 |
| Loss at step 400: 0.03928277641534805 |
| Loss at step 450: 0.056767385452985764 |
| Loss at step 500: 0.04085297882556915 |
| Loss at step 550: 0.0346493199467659 |
| Loss at step 600: 0.031214237213134766 |
| Loss at step 650: 0.04923402890563011 |
| Loss at step 700: 0.06293520331382751 |
| Loss at step 750: 0.03206819295883179 |
| Loss at step 800: 0.037167709320783615 |
| Loss at step 850: 0.03659123182296753 |
| Loss at step 900: 0.06300326436758041 |
| Mean training loss after epoch 208: 0.03929860922279579 |
|
|
| EPOCH: 209 |
| Loss at step 0: 0.03361137583851814 |
| Loss at step 50: 0.026428446173667908 |
| Loss at step 100: 0.034164272248744965 |
| Loss at step 150: 0.0364573635160923 |
| Loss at step 200: 0.041490085422992706 |
| Loss at step 250: 0.05014254152774811 |
| Loss at step 300: 0.04537137597799301 |
| Loss at step 350: 0.03958693891763687 |
| Loss at step 400: 0.03602161258459091 |
| Loss at step 450: 0.028751106932759285 |
| Loss at step 500: 0.04826076701283455 |
| Loss at step 550: 0.02965088002383709 |
| Loss at step 600: 0.03778247535228729 |
| Loss at step 650: 0.05167883262038231 |
| Loss at step 700: 0.029958970844745636 |
| Loss at step 750: 0.028692517429590225 |
| Loss at step 800: 0.029456930235028267 |
| Loss at step 850: 0.04437252879142761 |
| Loss at step 900: 0.03329407423734665 |
| Mean training loss after epoch 209: 0.03891603271367707 |
|
|
| EPOCH: 210 |
| Loss at step 0: 0.034966133534908295 |
| Loss at step 50: 0.03536030650138855 |
| Loss at step 100: 0.07317616790533066 |
| Loss at step 150: 0.0346195213496685 |
| Loss at step 200: 0.027271442115306854 |
| Loss at step 250: 0.024927347898483276 |
| Loss at step 300: 0.031957872211933136 |
| Loss at step 350: 0.033988021314144135 |
| Loss at step 400: 0.0324539951980114 |
| Loss at step 450: 0.027275538071990013 |
| Loss at step 500: 0.034581758081912994 |
| Loss at step 550: 0.05167417973279953 |
| Loss at step 600: 0.03554034233093262 |
| Loss at step 650: 0.03346965089440346 |
| Loss at step 700: 0.03680897504091263 |
| Loss at step 750: 0.03390275314450264 |
| Loss at step 800: 0.03475474938750267 |
| Loss at step 850: 0.03505280613899231 |
| Loss at step 900: 0.030334360897541046 |
| Mean training loss after epoch 210: 0.03880712026352885 |
|
|
| EPOCH: 211 |
| Loss at step 0: 0.036252912133932114 |
| Loss at step 50: 0.039842527359724045 |
| Loss at step 100: 0.05772003158926964 |
| Loss at step 150: 0.03693200275301933 |
| Loss at step 200: 0.04155736789107323 |
| Loss at step 250: 0.07334703207015991 |
| Loss at step 300: 0.03494369238615036 |
| Loss at step 350: 0.03264397382736206 |
| Loss at step 400: 0.029918232932686806 |
| Loss at step 450: 0.02940177358686924 |
| Loss at step 500: 0.03305615857243538 |
| Loss at step 550: 0.038510505110025406 |
| Loss at step 600: 0.03817708045244217 |
| Loss at step 650: 0.06494858860969543 |
| Loss at step 700: 0.030305013060569763 |
| Loss at step 750: 0.031748753041028976 |
| Loss at step 800: 0.03335447981953621 |
| Loss at step 850: 0.029417529702186584 |
| Loss at step 900: 0.03409936651587486 |
| Mean training loss after epoch 211: 0.03944876101582861 |
|
|
| EPOCH: 212 |
| Loss at step 0: 0.03463529050350189 |
| Loss at step 50: 0.04601624608039856 |
| Loss at step 100: 0.03157833218574524 |
| Loss at step 150: 0.036684878170490265 |
| Loss at step 200: 0.02828829362988472 |
| Loss at step 250: 0.032570887356996536 |
| Loss at step 300: 0.03535538911819458 |
| Loss at step 350: 0.03101705014705658 |
| Loss at step 400: 0.04566393792629242 |
| Loss at step 450: 0.03168395534157753 |
| Loss at step 500: 0.0468168742954731 |
| Loss at step 550: 0.03592550754547119 |
| Loss at step 600: 0.030794337391853333 |
| Loss at step 650: 0.039325911551713943 |
| Loss at step 700: 0.03656806796789169 |
| Loss at step 750: 0.03873779997229576 |
| Loss at step 800: 0.03337247669696808 |
| Loss at step 850: 0.05102049559354782 |
| Loss at step 900: 0.03313032537698746 |
| Mean training loss after epoch 212: 0.038925153630247504 |
|
|
| EPOCH: 213 |
| Loss at step 0: 0.03312905132770538 |
| Loss at step 50: 0.034856270998716354 |
| Loss at step 100: 0.03638290613889694 |
| Loss at step 150: 0.030590932816267014 |
| Loss at step 200: 0.03178534656763077 |
| Loss at step 250: 0.03628033027052879 |
| Loss at step 300: 0.03542640805244446 |
| Loss at step 350: 0.03868675231933594 |
| Loss at step 400: 0.03446945920586586 |
| Loss at step 450: 0.031632207334041595 |
| Loss at step 500: 0.03109968639910221 |
| Loss at step 550: 0.03217053413391113 |
| Loss at step 600: 0.031699612736701965 |
| Loss at step 650: 0.031050097197294235 |
| Loss at step 700: 0.03135169669985771 |
| Loss at step 750: 0.05426996573805809 |
| Loss at step 800: 0.031804606318473816 |
| Loss at step 850: 0.037588391453027725 |
| Loss at step 900: 0.058043185621500015 |
| Mean training loss after epoch 213: 0.038873043518934426 |
|
|
| EPOCH: 214 |
| Loss at step 0: 0.028386587277054787 |
| Loss at step 50: 0.050295133143663406 |
| Loss at step 100: 0.046243395656347275 |
| Loss at step 150: 0.03635299205780029 |
| Loss at step 200: 0.032567527145147324 |
| Loss at step 250: 0.026220694184303284 |
| Loss at step 300: 0.029877593740820885 |
| Loss at step 350: 0.036172349005937576 |
| Loss at step 400: 0.028854016214609146 |
| Loss at step 450: 0.033716216683387756 |
| Loss at step 500: 0.022720973938703537 |
| Loss at step 550: 0.03392207995057106 |
| Loss at step 600: 0.03281420096755028 |
| Loss at step 650: 0.06516284495592117 |
| Loss at step 700: 0.038289912045001984 |
| Loss at step 750: 0.03463584557175636 |
| Loss at step 800: 0.03736652433872223 |
| Loss at step 850: 0.049522798508405685 |
| Loss at step 900: 0.03491751104593277 |
| Mean training loss after epoch 214: 0.0391494517685222 |
|
|
| EPOCH: 215 |
| Loss at step 0: 0.028759721666574478 |
| Loss at step 50: 0.0389941930770874 |
| Loss at step 100: 0.030850136652588844 |
| Loss at step 150: 0.0318099707365036 |
| Loss at step 200: 0.034005057066679 |
| Loss at step 250: 0.052801866084337234 |
| Loss at step 300: 0.035044100135564804 |
| Loss at step 350: 0.052397601306438446 |
| Loss at step 400: 0.04918781667947769 |
| Loss at step 450: 0.0334693007171154 |
| Loss at step 500: 0.0463688038289547 |
| Loss at step 550: 0.033775392919778824 |
| Loss at step 600: 0.030683718621730804 |
| Loss at step 650: 0.03664135932922363 |
| Loss at step 700: 0.03418098762631416 |
| Loss at step 750: 0.03744911029934883 |
| Loss at step 800: 0.029192639514803886 |
| Loss at step 850: 0.05244458466768265 |
| Loss at step 900: 0.03115343302488327 |
| Mean training loss after epoch 215: 0.039243683344059026 |
|
|
| EPOCH: 216 |
| Loss at step 0: 0.0490545816719532 |
| Loss at step 50: 0.04258950799703598 |
| Loss at step 100: 0.03089594468474388 |
| Loss at step 150: 0.051975492388010025 |
| Loss at step 200: 0.051118168979883194 |
| Loss at step 250: 0.03654303401708603 |
| Loss at step 300: 0.05530314892530441 |
| Loss at step 350: 0.048319995403289795 |
| Loss at step 400: 0.03275396302342415 |
| Loss at step 450: 0.03487536683678627 |
| Loss at step 500: 0.03248892351984978 |
| Loss at step 550: 0.03192398324608803 |
| Loss at step 600: 0.0315568670630455 |
| Loss at step 650: 0.033777620643377304 |
| Loss at step 700: 0.04205819591879845 |
| Loss at step 750: 0.05035315454006195 |
| Loss at step 800: 0.032998502254486084 |
| Loss at step 850: 0.03232625871896744 |
| Loss at step 900: 0.04717819020152092 |
| Mean training loss after epoch 216: 0.039436278183624815 |
|
|
| EPOCH: 217 |
| Loss at step 0: 0.03221813589334488 |
| Loss at step 50: 0.03685177117586136 |
| Loss at step 100: 0.03466470539569855 |
| Loss at step 150: 0.031917549669742584 |
| Loss at step 200: 0.03797820210456848 |
| Loss at step 250: 0.04220956936478615 |
| Loss at step 300: 0.02914607711136341 |
| Loss at step 350: 0.05011358857154846 |
| Loss at step 400: 0.03618037700653076 |
| Loss at step 450: 0.03652055934071541 |
| Loss at step 500: 0.029459647834300995 |
| Loss at step 550: 0.031061850488185883 |
| Loss at step 600: 0.03643409535288811 |
| Loss at step 650: 0.04160930588841438 |
| Loss at step 700: 0.03050972893834114 |
| Loss at step 750: 0.03366422280669212 |
| Loss at step 800: 0.04819341376423836 |
| Loss at step 850: 0.04103286191821098 |
| Loss at step 900: 0.03597785905003548 |
| Mean training loss after epoch 217: 0.03884100024379901 |
|
|
| EPOCH: 218 |
| Loss at step 0: 0.034691356122493744 |
| Loss at step 50: 0.0512816347181797 |
| Loss at step 100: 0.0498422309756279 |
| Loss at step 150: 0.034464772790670395 |
| Loss at step 200: 0.03690893575549126 |
| Loss at step 250: 0.027062783017754555 |
| Loss at step 300: 0.037605658173561096 |
| Loss at step 350: 0.034629516303539276 |
| Loss at step 400: 0.033738873898983 |
| Loss at step 450: 0.05193618685007095 |
| Loss at step 500: 0.0349096916615963 |
| Loss at step 550: 0.03417463228106499 |
| Loss at step 600: 0.03664156794548035 |
| Loss at step 650: 0.050338178873062134 |
| Loss at step 700: 0.04119217023253441 |
| Loss at step 750: 0.04932699352502823 |
| Loss at step 800: 0.032882045954465866 |
| Loss at step 850: 0.05195748060941696 |
| Loss at step 900: 0.056095194071531296 |
| Mean training loss after epoch 218: 0.039159961821221466 |
|
|
| EPOCH: 219 |
| Loss at step 0: 0.03401932865381241 |
| Loss at step 50: 0.033970147371292114 |
| Loss at step 100: 0.02930084988474846 |
| Loss at step 150: 0.038355208933353424 |
| Loss at step 200: 0.02973788045346737 |
| Loss at step 250: 0.031476765871047974 |
| Loss at step 300: 0.03217240050435066 |
| Loss at step 350: 0.03983880579471588 |
| Loss at step 400: 0.035470906645059586 |
| Loss at step 450: 0.06286390125751495 |
| Loss at step 500: 0.03506041690707207 |
| Loss at step 550: 0.033520910888910294 |
| Loss at step 600: 0.03043210878968239 |
| Loss at step 650: 0.029626408591866493 |
| Loss at step 700: 0.03785299137234688 |
| Loss at step 750: 0.029762817546725273 |
| Loss at step 800: 0.03253250569105148 |
| Loss at step 850: 0.03220551460981369 |
| Loss at step 900: 0.03191516175866127 |
| Mean training loss after epoch 219: 0.03961604257533227 |
|
|
| EPOCH: 220 |
| Loss at step 0: 0.035007499158382416 |
| Loss at step 50: 0.03676850348711014 |
| Loss at step 100: 0.04776357114315033 |
| Loss at step 150: 0.05322731286287308 |
| Loss at step 200: 0.0326763354241848 |
| Loss at step 250: 0.02881242334842682 |
| Loss at step 300: 0.06452496349811554 |
| Loss at step 350: 0.04654304310679436 |
| Loss at step 400: 0.033597029745578766 |
| Loss at step 450: 0.06089720129966736 |
| Loss at step 500: 0.05325775966048241 |
| Loss at step 550: 0.034849777817726135 |
| Loss at step 600: 0.03910364955663681 |
| Loss at step 650: 0.04495980218052864 |
| Loss at step 700: 0.03320116922259331 |
| Loss at step 750: 0.031099630519747734 |
| Loss at step 800: 0.02967253513634205 |
| Loss at step 850: 0.05618506297469139 |
| Loss at step 900: 0.023359069600701332 |
| Mean training loss after epoch 220: 0.03917579456711057 |
|
|
| EPOCH: 221 |
| Loss at step 0: 0.033655982464551926 |
| Loss at step 50: 0.03283160179853439 |
| Loss at step 100: 0.06974532455205917 |
| Loss at step 150: 0.03525441884994507 |
| Loss at step 200: 0.033504560589790344 |
| Loss at step 250: 0.05242004245519638 |
| Loss at step 300: 0.049112677574157715 |
| Loss at step 350: 0.03913629800081253 |
| Loss at step 400: 0.030687332153320312 |
| Loss at step 450: 0.038405779749155045 |
| Loss at step 500: 0.055703528225421906 |
| Loss at step 550: 0.030176103115081787 |
| Loss at step 600: 0.0338451974093914 |
| Loss at step 650: 0.03483753651380539 |
| Loss at step 700: 0.046717628836631775 |
| Loss at step 750: 0.030152037739753723 |
| Loss at step 800: 0.051735568791627884 |
| Loss at step 850: 0.037322036921978 |
| Loss at step 900: 0.03671986982226372 |
| Mean training loss after epoch 221: 0.039529672089114246 |
|
|
| EPOCH: 222 |
| Loss at step 0: 0.040746379643678665 |
| Loss at step 50: 0.03421521559357643 |
| Loss at step 100: 0.03173847496509552 |
| Loss at step 150: 0.027398396283388138 |
| Loss at step 200: 0.06909073889255524 |
| Loss at step 250: 0.02943158522248268 |
| Loss at step 300: 0.04270821437239647 |
| Loss at step 350: 0.036493346095085144 |
| Loss at step 400: 0.03201095387339592 |
| Loss at step 450: 0.03346439078450203 |
| Loss at step 500: 0.03585893660783768 |
| Loss at step 550: 0.03641887009143829 |
| Loss at step 600: 0.029378972947597504 |
| Loss at step 650: 0.03823823109269142 |
| Loss at step 700: 0.03458854556083679 |
| Loss at step 750: 0.03295740485191345 |
| Loss at step 800: 0.03287142887711525 |
| Loss at step 850: 0.027882471680641174 |
| Loss at step 900: 0.037205036729574203 |
| Mean training loss after epoch 222: 0.038900254218817265 |
|
|
| EPOCH: 223 |
| Loss at step 0: 0.0465879812836647 |
| Loss at step 50: 0.04563435539603233 |
| Loss at step 100: 0.04606280103325844 |
| Loss at step 150: 0.03840605169534683 |
| Loss at step 200: 0.05436091497540474 |
| Loss at step 250: 0.03147437050938606 |
| Loss at step 300: 0.053010765463113785 |
| Loss at step 350: 0.048824213445186615 |
| Loss at step 400: 0.03178839758038521 |
| Loss at step 450: 0.05775563046336174 |
| Loss at step 500: 0.0351368747651577 |
| Loss at step 550: 0.03580415993928909 |
| Loss at step 600: 0.03660053759813309 |
| Loss at step 650: 0.04339035227894783 |
| Loss at step 700: 0.029888346791267395 |
| Loss at step 750: 0.06932703405618668 |
| Loss at step 800: 0.06762956827878952 |
| Loss at step 850: 0.047050874680280685 |
| Loss at step 900: 0.0409468412399292 |
| Mean training loss after epoch 223: 0.03930669576962238 |
|
|
| EPOCH: 224 |
| Loss at step 0: 0.03919639065861702 |
| Loss at step 50: 0.03524477407336235 |
| Loss at step 100: 0.0306867565959692 |
| Loss at step 150: 0.036834586411714554 |
| Loss at step 200: 0.039752665907144547 |
| Loss at step 250: 0.03726755827665329 |
| Loss at step 300: 0.035303812474012375 |
| Loss at step 350: 0.052437279373407364 |
| Loss at step 400: 0.04689083993434906 |
| Loss at step 450: 0.05424738675355911 |
| Loss at step 500: 0.03356577455997467 |
| Loss at step 550: 0.033016446977853775 |
| Loss at step 600: 0.04336623102426529 |
| Loss at step 650: 0.03267418593168259 |
| Loss at step 700: 0.0531155988574028 |
| Loss at step 750: 0.03437129035592079 |
| Loss at step 800: 0.05175966024398804 |
| Loss at step 850: 0.028837081044912338 |
| Loss at step 900: 0.05922437086701393 |
| Mean training loss after epoch 224: 0.0386647883195009 |
|
|
| EPOCH: 225 |
| Loss at step 0: 0.0487726666033268 |
| Loss at step 50: 0.03738081455230713 |
| Loss at step 100: 0.03074018657207489 |
| Loss at step 150: 0.03496900945901871 |
| Loss at step 200: 0.04817022383213043 |
| Loss at step 250: 0.04730185121297836 |
| Loss at step 300: 0.034258127212524414 |
| Loss at step 350: 0.034974534064531326 |
| Loss at step 400: 0.04086778685450554 |
| Loss at step 450: 0.03467993810772896 |
| Loss at step 500: 0.033085793256759644 |
| Loss at step 550: 0.09303003549575806 |
| Loss at step 600: 0.035393036901950836 |
| Loss at step 650: 0.035260267555713654 |
| Loss at step 700: 0.02982688695192337 |
| Loss at step 750: 0.032229792326688766 |
| Loss at step 800: 0.0330289825797081 |
| Loss at step 850: 0.03003479540348053 |
| Loss at step 900: 0.029361525550484657 |
| Mean training loss after epoch 225: 0.03890162737551592 |
|
|
| EPOCH: 226 |
| Loss at step 0: 0.032596834003925323 |
| Loss at step 50: 0.03327403590083122 |
| Loss at step 100: 0.05183485522866249 |
| Loss at step 150: 0.035328250378370285 |
| Loss at step 200: 0.03449290245771408 |
| Loss at step 250: 0.03304436802864075 |
| Loss at step 300: 0.0349038764834404 |
| Loss at step 350: 0.034197621047496796 |
| Loss at step 400: 0.03215165063738823 |
| Loss at step 450: 0.05494612455368042 |
| Loss at step 500: 0.03127453476190567 |
| Loss at step 550: 0.039228979498147964 |
| Loss at step 600: 0.03493810445070267 |
| Loss at step 650: 0.031990375369787216 |
| Loss at step 700: 0.05123075097799301 |
| Loss at step 750: 0.02953650988638401 |
| Loss at step 800: 0.06226984038949013 |
| Loss at step 850: 0.0387241505086422 |
| Loss at step 900: 0.04094972461462021 |
| Mean training loss after epoch 226: 0.03866499135576522 |
|
|
| EPOCH: 227 |
| Loss at step 0: 0.036986760795116425 |
| Loss at step 50: 0.03572157025337219 |
| Loss at step 100: 0.03125032037496567 |
| Loss at step 150: 0.03796594962477684 |
| Loss at step 200: 0.03207015246152878 |
| Loss at step 250: 0.04670996591448784 |
| Loss at step 300: 0.04626971855759621 |
| Loss at step 350: 0.032793719321489334 |
| Loss at step 400: 0.05023602768778801 |
| Loss at step 450: 0.04747440665960312 |
| Loss at step 500: 0.029485540464520454 |
| Loss at step 550: 0.032925404608249664 |
| Loss at step 600: 0.04897836595773697 |
| Loss at step 650: 0.03635069355368614 |
| Loss at step 700: 0.031118150800466537 |
| Loss at step 750: 0.04991304501891136 |
| Loss at step 800: 0.03546535223722458 |
| Loss at step 850: 0.04668305441737175 |
| Loss at step 900: 0.03488590568304062 |
| Mean training loss after epoch 227: 0.03901965910398058 |
|
|
| EPOCH: 228 |
| Loss at step 0: 0.04974370822310448 |
| Loss at step 50: 0.03373951464891434 |
| Loss at step 100: 0.033895570784807205 |
| Loss at step 150: 0.03443300724029541 |
| Loss at step 200: 0.03293636441230774 |
| Loss at step 250: 0.04655670002102852 |
| Loss at step 300: 0.02601798251271248 |
| Loss at step 350: 0.03469035029411316 |
| Loss at step 400: 0.02628219872713089 |
| Loss at step 450: 0.034008219838142395 |
| Loss at step 500: 0.037605393677949905 |
| Loss at step 550: 0.04931284487247467 |
| Loss at step 600: 0.047764576971530914 |
| Loss at step 650: 0.03390823304653168 |
| Loss at step 700: 0.036051489412784576 |
| Loss at step 750: 0.03190223500132561 |
| Loss at step 800: 0.027517585083842278 |
| Loss at step 850: 0.026676075533032417 |
| Loss at step 900: 0.035612836480140686 |
| Mean training loss after epoch 228: 0.039303831541652616 |
|
|
| EPOCH: 229 |
| Loss at step 0: 0.0288917887955904 |
| Loss at step 50: 0.06252633035182953 |
| Loss at step 100: 0.03233402222394943 |
| Loss at step 150: 0.03316564857959747 |
| Loss at step 200: 0.034948062151670456 |
| Loss at step 250: 0.02844369038939476 |
| Loss at step 300: 0.031051771715283394 |
| Loss at step 350: 0.04054409638047218 |
| Loss at step 400: 0.03256020322442055 |
| Loss at step 450: 0.03426603227853775 |
| Loss at step 500: 0.033789653331041336 |
| Loss at step 550: 0.027340136468410492 |
| Loss at step 600: 0.03498152643442154 |
| Loss at step 650: 0.033199094235897064 |
| Loss at step 700: 0.04883769899606705 |
| Loss at step 750: 0.041650619357824326 |
| Loss at step 800: 0.04393523558974266 |
| Loss at step 850: 0.03471892699599266 |
| Loss at step 900: 0.04146659001708031 |
| Mean training loss after epoch 229: 0.03924599493831905 |
|
|
| EPOCH: 230 |
| Loss at step 0: 0.03501119837164879 |
| Loss at step 50: 0.031957730650901794 |
| Loss at step 100: 0.06566370278596878 |
| Loss at step 150: 0.03400876373052597 |
| Loss at step 200: 0.03895947337150574 |
| Loss at step 250: 0.03220095485448837 |
| Loss at step 300: 0.03845254331827164 |
| Loss at step 350: 0.03425909951329231 |
| Loss at step 400: 0.03328506276011467 |
| Loss at step 450: 0.03391508758068085 |
| Loss at step 500: 0.0358009859919548 |
| Loss at step 550: 0.03223380073904991 |
| Loss at step 600: 0.028488216921687126 |
| Loss at step 650: 0.03499921038746834 |
| Loss at step 700: 0.028844734653830528 |
| Loss at step 750: 0.04680425301194191 |
| Loss at step 800: 0.029129402711987495 |
| Loss at step 850: 0.05072740092873573 |
| Loss at step 900: 0.03400781750679016 |
| Mean training loss after epoch 230: 0.03910631895152681 |
|
|
| EPOCH: 231 |
| Loss at step 0: 0.029594002291560173 |
| Loss at step 50: 0.033388398587703705 |
| Loss at step 100: 0.030222948640584946 |
| Loss at step 150: 0.03679189831018448 |
| Loss at step 200: 0.034237805753946304 |
| Loss at step 250: 0.03863455355167389 |
| Loss at step 300: 0.03193581849336624 |
| Loss at step 350: 0.04040062427520752 |
| Loss at step 400: 0.042988717555999756 |
| Loss at step 450: 0.048941951245069504 |
| Loss at step 500: 0.036740753799676895 |
| Loss at step 550: 0.04671098291873932 |
| Loss at step 600: 0.02967640571296215 |
| Loss at step 650: 0.03679099678993225 |
| Loss at step 700: 0.03040301240980625 |
| Loss at step 750: 0.048260949552059174 |
| Loss at step 800: 0.04979100823402405 |
| Loss at step 850: 0.03750582039356232 |
| Loss at step 900: 0.0348072350025177 |
| Mean training loss after epoch 231: 0.03864938818784093 |
|
|
| EPOCH: 232 |
| Loss at step 0: 0.04080262780189514 |
| Loss at step 50: 0.0347711406648159 |
| Loss at step 100: 0.04852057993412018 |
| Loss at step 150: 0.035218965262174606 |
| Loss at step 200: 0.03393496945500374 |
| Loss at step 250: 0.047152820974588394 |
| Loss at step 300: 0.03276395425200462 |
| Loss at step 350: 0.03969300538301468 |
| Loss at step 400: 0.031070873141288757 |
| Loss at step 450: 0.0518072172999382 |
| Loss at step 500: 0.03238266333937645 |
| Loss at step 550: 0.03421289101243019 |
| Loss at step 600: 0.04663264751434326 |
| Loss at step 650: 0.033952854573726654 |
| Loss at step 700: 0.050659772008657455 |
| Loss at step 750: 0.02905084565281868 |
| Loss at step 800: 0.0381062813103199 |
| Loss at step 850: 0.03878077492117882 |
| Loss at step 900: 0.033104512840509415 |
| Mean training loss after epoch 232: 0.03919413653033565 |
|
|
| EPOCH: 233 |
| Loss at step 0: 0.030844537541270256 |
| Loss at step 50: 0.05130849406123161 |
| Loss at step 100: 0.06363049894571304 |
| Loss at step 150: 0.032386112958192825 |
| Loss at step 200: 0.03350917622447014 |
| Loss at step 250: 0.04133734479546547 |
| Loss at step 300: 0.03743404150009155 |
| Loss at step 350: 0.030576646327972412 |
| Loss at step 400: 0.03490522503852844 |
| Loss at step 450: 0.046362556517124176 |
| Loss at step 500: 0.03323175013065338 |
| Loss at step 550: 0.042132873088121414 |
| Loss at step 600: 0.05004531517624855 |
| Loss at step 650: 0.036296043545007706 |
| Loss at step 700: 0.03690037876367569 |
| Loss at step 750: 0.041600458323955536 |
| Loss at step 800: 0.047059088945388794 |
| Loss at step 850: 0.03368044272065163 |
| Loss at step 900: 0.04677114263176918 |
| Mean training loss after epoch 233: 0.039063130267091524 |
|
|
| EPOCH: 234 |
| Loss at step 0: 0.03360550105571747 |
| Loss at step 50: 0.03290083259344101 |
| Loss at step 100: 0.034932930022478104 |
| Loss at step 150: 0.03593425080180168 |
| Loss at step 200: 0.0440426729619503 |
| Loss at step 250: 0.05007448047399521 |
| Loss at step 300: 0.036359675228595734 |
| Loss at step 350: 0.031192978844046593 |
| Loss at step 400: 0.02669164352118969 |
| Loss at step 450: 0.0325494147837162 |
| Loss at step 500: 0.02928798645734787 |
| Loss at step 550: 0.037536706775426865 |
| Loss at step 600: 0.04039769247174263 |
| Loss at step 650: 0.03147507831454277 |
| Loss at step 700: 0.05367570370435715 |
| Loss at step 750: 0.044787436723709106 |
| Loss at step 800: 0.028223801404237747 |
| Loss at step 850: 0.05118735879659653 |
| Loss at step 900: 0.03116217814385891 |
| Mean training loss after epoch 234: 0.03894189336120701 |
|
|
| EPOCH: 235 |
| Loss at step 0: 0.03430939093232155 |
| Loss at step 50: 0.03122139908373356 |
| Loss at step 100: 0.03365548327565193 |
| Loss at step 150: 0.029249092563986778 |
| Loss at step 200: 0.055865611881017685 |
| Loss at step 250: 0.04746852442622185 |
| Loss at step 300: 0.0332975871860981 |
| Loss at step 350: 0.04182635620236397 |
| Loss at step 400: 0.0324370302259922 |
| Loss at step 450: 0.03309211879968643 |
| Loss at step 500: 0.03604499250650406 |
| Loss at step 550: 0.02579680271446705 |
| Loss at step 600: 0.0383726991713047 |
| Loss at step 650: 0.03601831570267677 |
| Loss at step 700: 0.03624669089913368 |
| Loss at step 750: 0.03240237012505531 |
| Loss at step 800: 0.037177249789237976 |
| Loss at step 850: 0.033456508070230484 |
| Loss at step 900: 0.031031902879476547 |
| Mean training loss after epoch 235: 0.03901019287921155 |
|
|
| EPOCH: 236 |
| Loss at step 0: 0.04972589761018753 |
| Loss at step 50: 0.044508837163448334 |
| Loss at step 100: 0.035206809639930725 |
| Loss at step 150: 0.03639085963368416 |
| Loss at step 200: 0.04725930094718933 |
| Loss at step 250: 0.05124089494347572 |
| Loss at step 300: 0.05921793729066849 |
| Loss at step 350: 0.0671318843960762 |
| Loss at step 400: 0.03563682362437248 |
| Loss at step 450: 0.03774426132440567 |
| Loss at step 500: 0.05267181992530823 |
| Loss at step 550: 0.03393133357167244 |
| Loss at step 600: 0.03175343573093414 |
| Loss at step 650: 0.05360681936144829 |
| Loss at step 700: 0.05347166955471039 |
| Loss at step 750: 0.035956475883722305 |
| Loss at step 800: 0.03781764209270477 |
| Loss at step 850: 0.031263165175914764 |
| Loss at step 900: 0.03862721472978592 |
| Mean training loss after epoch 236: 0.03922938913512014 |
|
|
| EPOCH: 237 |
| Loss at step 0: 0.06566672772169113 |
| Loss at step 50: 0.031792644411325455 |
| Loss at step 100: 0.03735481947660446 |
| Loss at step 150: 0.034779392182826996 |
| Loss at step 200: 0.06874541938304901 |
| Loss at step 250: 0.05580836907029152 |
| Loss at step 300: 0.04530530422925949 |
| Loss at step 350: 0.03510119020938873 |
| Loss at step 400: 0.03770944103598595 |
| Loss at step 450: 0.06360543519258499 |
| Loss at step 500: 0.05799640715122223 |
| Loss at step 550: 0.03166048228740692 |
| Loss at step 600: 0.04853644594550133 |
| Loss at step 650: 0.03736702352762222 |
| Loss at step 700: 0.03405768796801567 |
| Loss at step 750: 0.03659890219569206 |
| Loss at step 800: 0.025363899767398834 |
| Loss at step 850: 0.03285384550690651 |
| Loss at step 900: 0.027758479118347168 |
| Mean training loss after epoch 237: 0.03887855290953539 |
|
|
| EPOCH: 238 |
| Loss at step 0: 0.03370613232254982 |
| Loss at step 50: 0.04008691385388374 |
| Loss at step 100: 0.03578205406665802 |
| Loss at step 150: 0.05115322396159172 |
| Loss at step 200: 0.05261386185884476 |
| Loss at step 250: 0.03729558736085892 |
| Loss at step 300: 0.03199843317270279 |
| Loss at step 350: 0.05246369168162346 |
| Loss at step 400: 0.0468270368874073 |
| Loss at step 450: 0.03319402039051056 |
| Loss at step 500: 0.03305567055940628 |
| Loss at step 550: 0.05719682201743126 |
| Loss at step 600: 0.051412034779787064 |
| Loss at step 650: 0.03428887948393822 |
| Loss at step 700: 0.039233673363924026 |
| Loss at step 750: 0.028749994933605194 |
| Loss at step 800: 0.035130735486745834 |
| Loss at step 850: 0.03354177996516228 |
| Loss at step 900: 0.035492267459630966 |
| Mean training loss after epoch 238: 0.03845155325287314 |
|
|
| EPOCH: 239 |
| Loss at step 0: 0.03153827786445618 |
| Loss at step 50: 0.028729312121868134 |
| Loss at step 100: 0.06446194648742676 |
| Loss at step 150: 0.05377689749002457 |
| Loss at step 200: 0.08363740891218185 |
| Loss at step 250: 0.028110457584261894 |
| Loss at step 300: 0.03177076205611229 |
| Loss at step 350: 0.02922712452709675 |
| Loss at step 400: 0.03213167190551758 |
| Loss at step 450: 0.03212027996778488 |
| Loss at step 500: 0.03912307322025299 |
| Loss at step 550: 0.032081134617328644 |
| Loss at step 600: 0.04772243648767471 |
| Loss at step 650: 0.029768118634819984 |
| Loss at step 700: 0.05186186730861664 |
| Loss at step 750: 0.0529065765440464 |
| Loss at step 800: 0.0341126024723053 |
| Loss at step 850: 0.051606934517621994 |
| Loss at step 900: 0.06367186456918716 |
| Mean training loss after epoch 239: 0.03928863341961779 |
|
|
| EPOCH: 240 |
| Loss at step 0: 0.03887162730097771 |
| Loss at step 50: 0.050880786031484604 |
| Loss at step 100: 0.039547499269247055 |
| Loss at step 150: 0.028909575194120407 |
| Loss at step 200: 0.03536584973335266 |
| Loss at step 250: 0.03659897297620773 |
| Loss at step 300: 0.033108536154031754 |
| Loss at step 350: 0.02675941213965416 |
| Loss at step 400: 0.03612198680639267 |
| Loss at step 450: 0.05426546931266785 |
| Loss at step 500: 0.0330321379005909 |
| Loss at step 550: 0.05149666219949722 |
| Loss at step 600: 0.03463984280824661 |
| Loss at step 650: 0.028366481885313988 |
| Loss at step 700: 0.044816501438617706 |
| Loss at step 750: 0.04147651046514511 |
| Loss at step 800: 0.033853594213724136 |
| Loss at step 850: 0.03742264583706856 |
| Loss at step 900: 0.03427162766456604 |
| Mean training loss after epoch 240: 0.03903400417028079 |
|
|
| EPOCH: 241 |
| Loss at step 0: 0.03618745505809784 |
| Loss at step 50: 0.03541552275419235 |
| Loss at step 100: 0.03832630813121796 |
| Loss at step 150: 0.031097810715436935 |
| Loss at step 200: 0.05100856348872185 |
| Loss at step 250: 0.05207272991538048 |
| Loss at step 300: 0.03732670843601227 |
| Loss at step 350: 0.03722361475229263 |
| Loss at step 400: 0.035946499556303024 |
| Loss at step 450: 0.04949504882097244 |
| Loss at step 500: 0.04036898910999298 |
| Loss at step 550: 0.03357747942209244 |
| Loss at step 600: 0.03259134665131569 |
| Loss at step 650: 0.037669405341148376 |
| Loss at step 700: 0.03358788415789604 |
| Loss at step 750: 0.05227615311741829 |
| Loss at step 800: 0.03881647065281868 |
| Loss at step 850: 0.033830676227808 |
| Loss at step 900: 0.04052143916487694 |
| Mean training loss after epoch 241: 0.03923677153257864 |
|
|
| EPOCH: 242 |
| Loss at step 0: 0.031095046550035477 |
| Loss at step 50: 0.036116208881139755 |
| Loss at step 100: 0.029616231098771095 |
| Loss at step 150: 0.04071303829550743 |
| Loss at step 200: 0.05126992613077164 |
| Loss at step 250: 0.02957015484571457 |
| Loss at step 300: 0.04773343354463577 |
| Loss at step 350: 0.045865584164857864 |
| Loss at step 400: 0.03263732045888901 |
| Loss at step 450: 0.032689858227968216 |
| Loss at step 500: 0.04566469416022301 |
| Loss at step 550: 0.0382232740521431 |
| Loss at step 600: 0.0350029319524765 |
| Loss at step 650: 0.032325662672519684 |
| Loss at step 700: 0.029781801626086235 |
| Loss at step 750: 0.049002304673194885 |
| Loss at step 800: 0.05032792314887047 |
| Loss at step 850: 0.04957038536667824 |
| Loss at step 900: 0.03759152814745903 |
| Mean training loss after epoch 242: 0.03931932861029085 |
|
|
| EPOCH: 243 |
| Loss at step 0: 0.038250263780355453 |
| Loss at step 50: 0.03833189606666565 |
| Loss at step 100: 0.03076758421957493 |
| Loss at step 150: 0.037498589605093 |
| Loss at step 200: 0.03554268926382065 |
| Loss at step 250: 0.029882796108722687 |
| Loss at step 300: 0.035731032490730286 |
| Loss at step 350: 0.030183272436261177 |
| Loss at step 400: 0.049779996275901794 |
| Loss at step 450: 0.040445808321237564 |
| Loss at step 500: 0.047706544399261475 |
| Loss at step 550: 0.047556981444358826 |
| Loss at step 600: 0.03681233897805214 |
| Loss at step 650: 0.02908683754503727 |
| Loss at step 700: 0.02920670434832573 |
| Loss at step 750: 0.030341751873493195 |
| Loss at step 800: 0.030939605087041855 |
| Loss at step 850: 0.03304276987910271 |
| Loss at step 900: 0.03191913291811943 |
| Mean training loss after epoch 243: 0.03905273881206698 |
|
|
| EPOCH: 244 |
| Loss at step 0: 0.06598794460296631 |
| Loss at step 50: 0.02893797680735588 |
| Loss at step 100: 0.03549230843782425 |
| Loss at step 150: 0.03681396320462227 |
| Loss at step 200: 0.03673600032925606 |
| Loss at step 250: 0.04525090008974075 |
| Loss at step 300: 0.03606670722365379 |
| Loss at step 350: 0.024963827803730965 |
| Loss at step 400: 0.05136997625231743 |
| Loss at step 450: 0.06354724615812302 |
| Loss at step 500: 0.03802937641739845 |
| Loss at step 550: 0.04526302590966225 |
| Loss at step 600: 0.03441905602812767 |
| Loss at step 650: 0.04524055868387222 |
| Loss at step 700: 0.02656574547290802 |
| Loss at step 750: 0.07117734104394913 |
| Loss at step 800: 0.036995913833379745 |
| Loss at step 850: 0.03178925812244415 |
| Loss at step 900: 0.037087518721818924 |
| Mean training loss after epoch 244: 0.039370466573739736 |
|
|
| EPOCH: 245 |
| Loss at step 0: 0.048373542726039886 |
| Loss at step 50: 0.033451057970523834 |
| Loss at step 100: 0.028194300830364227 |
| Loss at step 150: 0.047540027648210526 |
| Loss at step 200: 0.035591401159763336 |
| Loss at step 250: 0.047648560255765915 |
| Loss at step 300: 0.02961515262722969 |
| Loss at step 350: 0.036826133728027344 |
| Loss at step 400: 0.036649350076913834 |
| Loss at step 450: 0.046033795922994614 |
| Loss at step 500: 0.038278304040431976 |
| Loss at step 550: 0.05737805739045143 |
| Loss at step 600: 0.034337978810071945 |
| Loss at step 650: 0.047907229512929916 |
| Loss at step 700: 0.029077300801873207 |
| Loss at step 750: 0.026141535490751266 |
| Loss at step 800: 0.040920063853263855 |
| Loss at step 850: 0.03886491060256958 |
| Loss at step 900: 0.03947722166776657 |
| Mean training loss after epoch 245: 0.0392171511811806 |
|
|
| EPOCH: 246 |
| Loss at step 0: 0.03227032348513603 |
| Loss at step 50: 0.03560794144868851 |
| Loss at step 100: 0.03743302822113037 |
| Loss at step 150: 0.030393380671739578 |
| Loss at step 200: 0.05443951487541199 |
| Loss at step 250: 0.050018344074487686 |
| Loss at step 300: 0.0319996252655983 |
| Loss at step 350: 0.03510141000151634 |
| Loss at step 400: 0.05677584186196327 |
| Loss at step 450: 0.05292211100459099 |
| Loss at step 500: 0.04954027011990547 |
| Loss at step 550: 0.036719441413879395 |
| Loss at step 600: 0.037654343992471695 |
| Loss at step 650: 0.045069120824337006 |
| Loss at step 700: 0.04070558398962021 |
| Loss at step 750: 0.03152972832322121 |
| Loss at step 800: 0.032653797417879105 |
| Loss at step 850: 0.032190416008234024 |
| Loss at step 900: 0.03875408694148064 |
| Mean training loss after epoch 246: 0.03970623201827632 |
|
|
| EPOCH: 247 |
| Loss at step 0: 0.03510020300745964 |
| Loss at step 50: 0.028460653498768806 |
| Loss at step 100: 0.027597110718488693 |
| Loss at step 150: 0.0368226058781147 |
| Loss at step 200: 0.03382598236203194 |
| Loss at step 250: 0.0317443311214447 |
| Loss at step 300: 0.04087996855378151 |
| Loss at step 350: 0.029916564002633095 |
| Loss at step 400: 0.03555725887417793 |
| Loss at step 450: 0.047546543180942535 |
| Loss at step 500: 0.03285599499940872 |
| Loss at step 550: 0.03534551337361336 |
| Loss at step 600: 0.06219349056482315 |
| Loss at step 650: 0.0399542972445488 |
| Loss at step 700: 0.028029421344399452 |
| Loss at step 750: 0.0445200577378273 |
| Loss at step 800: 0.03424978256225586 |
| Loss at step 850: 0.025731278583407402 |
| Loss at step 900: 0.038073569536209106 |
| Mean training loss after epoch 247: 0.039303670599580065 |
|
|
| EPOCH: 248 |
| Loss at step 0: 0.030523089691996574 |
| Loss at step 50: 0.03858974575996399 |
| Loss at step 100: 0.03324702009558678 |
| Loss at step 150: 0.033646613359451294 |
| Loss at step 200: 0.028181303292512894 |
| Loss at step 250: 0.03686067834496498 |
| Loss at step 300: 0.02987145632505417 |
| Loss at step 350: 0.02894950844347477 |
| Loss at step 400: 0.06689367443323135 |
| Loss at step 450: 0.029212500900030136 |
| Loss at step 500: 0.03293319419026375 |
| Loss at step 550: 0.032734330743551254 |
| Loss at step 600: 0.028700964525341988 |
| Loss at step 650: 0.0392322838306427 |
| Loss at step 700: 0.048850029706954956 |
| Loss at step 750: 0.05006673187017441 |
| Loss at step 800: 0.03270864486694336 |
| Loss at step 850: 0.03546704351902008 |
| Loss at step 900: 0.03319983929395676 |
| Mean training loss after epoch 248: 0.03888359318163667 |
|
|
| EPOCH: 249 |
| Loss at step 0: 0.036848295480012894 |
| Loss at step 50: 0.03344952315092087 |
| Loss at step 100: 0.030375657603144646 |
| Loss at step 150: 0.04395264759659767 |
| Loss at step 200: 0.032356906682252884 |
| Loss at step 250: 0.03715647757053375 |
| Loss at step 300: 0.047168467193841934 |
| Loss at step 350: 0.028079552575945854 |
| Loss at step 400: 0.06304143369197845 |
| Loss at step 450: 0.03134738653898239 |
| Loss at step 500: 0.03943207114934921 |
| Loss at step 550: 0.02837996557354927 |
| Loss at step 600: 0.043990012258291245 |
| Loss at step 650: 0.05276433750987053 |
| Loss at step 700: 0.03396638482809067 |
| Loss at step 750: 0.02971942536532879 |
| Loss at step 800: 0.034827522933483124 |
| Loss at step 850: 0.05165525898337364 |
| Loss at step 900: 0.04828884080052376 |
| Mean training loss after epoch 249: 0.03892020832127663 |
|
|
| EPOCH: 250 |
| Loss at step 0: 0.03388892486691475 |
| Loss at step 50: 0.03131557255983353 |
| Loss at step 100: 0.05106263980269432 |
| Loss at step 150: 0.034251436591148376 |
| Loss at step 200: 0.05417611449956894 |
| Loss at step 250: 0.035362519323825836 |
| Loss at step 300: 0.03308453783392906 |
| Loss at step 350: 0.03788948431611061 |
| Loss at step 400: 0.03544377163052559 |
| Loss at step 450: 0.036667026579380035 |
| Loss at step 500: 0.05221394822001457 |
| Loss at step 550: 0.04554745554924011 |
| Loss at step 600: 0.030167769640684128 |
| Loss at step 650: 0.0340573750436306 |
| Loss at step 700: 0.04917588457465172 |
| Loss at step 750: 0.03632596880197525 |
| Loss at step 800: 0.03073965013027191 |
| Loss at step 850: 0.027795109897851944 |
| Loss at step 900: 0.03306938707828522 |
| Mean training loss after epoch 250: 0.03891344393081248 |
|
|
| EPOCH: 251 |
| Loss at step 0: 0.039425771683454514 |
| Loss at step 50: 0.030994482338428497 |
| Loss at step 100: 0.05115703493356705 |
| Loss at step 150: 0.03247106820344925 |
| Loss at step 200: 0.036715537309646606 |
| Loss at step 250: 0.04827232286334038 |
| Loss at step 300: 0.04005703702569008 |
| Loss at step 350: 0.03334115818142891 |
| Loss at step 400: 0.03590434789657593 |
| Loss at step 450: 0.035763200372457504 |
| Loss at step 500: 0.04415174201130867 |
| Loss at step 550: 0.040398284792900085 |
| Loss at step 600: 0.0336516872048378 |
| Loss at step 650: 0.06026959791779518 |
| Loss at step 700: 0.05111640691757202 |
| Loss at step 750: 0.03100166656076908 |
| Loss at step 800: 0.030193939805030823 |
| Loss at step 850: 0.03528062254190445 |
| Loss at step 900: 0.0353427454829216 |
| Mean training loss after epoch 251: 0.03915926600014096 |
|
|
| EPOCH: 252 |
| Loss at step 0: 0.047453004866838455 |
| Loss at step 50: 0.03374917432665825 |
| Loss at step 100: 0.028373297303915024 |
| Loss at step 150: 0.06039762496948242 |
| Loss at step 200: 0.030490513890981674 |
| Loss at step 250: 0.026922443881630898 |
| Loss at step 300: 0.023987047374248505 |
| Loss at step 350: 0.03156132623553276 |
| Loss at step 400: 0.0331733338534832 |
| Loss at step 450: 0.028657851740717888 |
| Loss at step 500: 0.03813005983829498 |
| Loss at step 550: 0.028067875653505325 |
| Loss at step 600: 0.038628771901130676 |
| Loss at step 650: 0.033749524503946304 |
| Loss at step 700: 0.035677965730428696 |
| Loss at step 750: 0.028826430439949036 |
| Loss at step 800: 0.035348471254110336 |
| Loss at step 850: 0.03522353619337082 |
| Loss at step 900: 0.03503499925136566 |
| Mean training loss after epoch 252: 0.03881135373028802 |
|
|
| EPOCH: 253 |
| Loss at step 0: 0.05614131689071655 |
| Loss at step 50: 0.02495838701725006 |
| Loss at step 100: 0.047003425657749176 |
| Loss at step 150: 0.029494477435946465 |
| Loss at step 200: 0.030162688344717026 |
| Loss at step 250: 0.06391389667987823 |
| Loss at step 300: 0.02871713973581791 |
| Loss at step 350: 0.03513010963797569 |
| Loss at step 400: 0.06640415638685226 |
| Loss at step 450: 0.03930817171931267 |
| Loss at step 500: 0.029120804741978645 |
| Loss at step 550: 0.02939414419233799 |
| Loss at step 600: 0.0732555091381073 |
| Loss at step 650: 0.030200641602277756 |
| Loss at step 700: 0.03482096642255783 |
| Loss at step 750: 0.030384305864572525 |
| Loss at step 800: 0.034591879695653915 |
| Loss at step 850: 0.03804127871990204 |
| Loss at step 900: 0.04978170618414879 |
| Mean training loss after epoch 253: 0.038376694193669854 |
|
|
| EPOCH: 254 |
| Loss at step 0: 0.03105386532843113 |
| Loss at step 50: 0.04386408254504204 |
| Loss at step 100: 0.0490780733525753 |
| Loss at step 150: 0.035760875791311264 |
| Loss at step 200: 0.03169526532292366 |
| Loss at step 250: 0.033175766468048096 |
| Loss at step 300: 0.04693169519305229 |
| Loss at step 350: 0.033079493790864944 |
| Loss at step 400: 0.0317985899746418 |
| Loss at step 450: 0.03733246773481369 |
| Loss at step 500: 0.07064293324947357 |
| Loss at step 550: 0.03856551647186279 |
| Loss at step 600: 0.04498601704835892 |
| Loss at step 650: 0.050545915961265564 |
| Loss at step 700: 0.032745424658060074 |
| Loss at step 750: 0.03153373301029205 |
| Loss at step 800: 0.031938981264829636 |
| Loss at step 850: 0.06265679746866226 |
| Loss at step 900: 0.06484922766685486 |
| Mean training loss after epoch 254: 0.038683842222240056 |
|
|
| EPOCH: 255 |
| Loss at step 0: 0.06329744309186935 |
| Loss at step 50: 0.04864448308944702 |
| Loss at step 100: 0.04800715669989586 |
| Loss at step 150: 0.05042517930269241 |
| Loss at step 200: 0.03493422269821167 |
| Loss at step 250: 0.033742208033800125 |
| Loss at step 300: 0.04552479088306427 |
| Loss at step 350: 0.03966813534498215 |
| Loss at step 400: 0.035213813185691833 |
| Loss at step 450: 0.037367723882198334 |
| Loss at step 500: 0.03437703102827072 |
| Loss at step 550: 0.036713361740112305 |
| Loss at step 600: 0.04268074035644531 |
| Loss at step 650: 0.03121114708483219 |
| Loss at step 700: 0.04252978786826134 |
| Loss at step 750: 0.028949955478310585 |
| Loss at step 800: 0.03465810418128967 |
| Loss at step 850: 0.030163034796714783 |
| Loss at step 900: 0.03492124006152153 |
| Mean training loss after epoch 255: 0.03891279746387114 |
|
|
| EPOCH: 256 |
| Loss at step 0: 0.034355781972408295 |
| Loss at step 50: 0.051526568830013275 |
| Loss at step 100: 0.05003800615668297 |
| Loss at step 150: 0.03800094127655029 |
| Loss at step 200: 0.06754325330257416 |
| Loss at step 250: 0.03478672355413437 |
| Loss at step 300: 0.051567237824201584 |
| Loss at step 350: 0.0353977233171463 |
| Loss at step 400: 0.050872646272182465 |
| Loss at step 450: 0.04051677882671356 |
| Loss at step 500: 0.03876348212361336 |
| Loss at step 550: 0.04580125957727432 |
| Loss at step 600: 0.03618134185671806 |
| Loss at step 650: 0.025788437575101852 |
| Loss at step 700: 0.03462769463658333 |
| Loss at step 750: 0.0296726506203413 |
| Loss at step 800: 0.035372376441955566 |
| Loss at step 850: 0.06778371334075928 |
| Loss at step 900: 0.04379701986908913 |
| Mean training loss after epoch 256: 0.039542128389546356 |
|
|
