Last active
July 3, 2018 11:05
-
-
Save bearpelican/e2aa5022351ec1bdcd612986686b2c60 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
python -m torch.distributed.launch --nproc_per_node=8 --nnodes=8 --node_rank=1 --master_addr=192.168.18.149 --master_port=6006 train_imagenet_nv.py ~/data/imagenet --save-dir ~/data/training/nv/2018-07-03T01:13:29-cluster_8_region_b_spot-1-lr12-e68-bs256-warmup-4 --loss-scale 512 --fp16 -b 256 --sz 224 -j 8 --lr 3.2 --warmup 4 --epochs 68 --small --dist-url env:// --dist-backend nccl --distributed | |
~~epoch hours top1Accuracy | |
Distributed: init_process_group success | |
Loaded model | |
Defined loss and optimizer | |
Created data loaders | |
Begin training | |
~~0 0.015604126666666666 2.366 | |
* Prec@1 0.563 Prec@5 2.366 | |
~~1 0.023700757777777776 5.631 | |
* Prec@1 1.588 Prec@5 5.631 | |
~~2 0.031909763333333334 12.580 | |
* Prec@1 4.204 Prec@5 12.580 | |
~~3 0.04014015361111111 22.377 | |
* Prec@1 8.576 Prec@5 22.377 | |
~~4 0.04825068861111111 19.437 | |
* Prec@1 7.361 Prec@5 19.437 | |
~~5 0.056131485833333335 25.506 | |
* Prec@1 10.292 Prec@5 25.506 | |
~~6 0.06431049277777778 30.019 | |
* Prec@1 13.175 Prec@5 30.019 | |
~~7 0.07240750027777779 36.301 | |
* Prec@1 16.520 Prec@5 36.301 | |
~~8 0.08038220916666666 41.744 | |
* Prec@1 20.223 Prec@5 41.744 | |
~~9 0.08838161666666666 29.592 | |
* Prec@1 12.830 Prec@5 29.592 | |
~~10 0.09625070694444444 39.268 | |
* Prec@1 18.492 Prec@5 39.268 | |
~~11 0.10399708055555557 44.353 | |
* Prec@1 21.707 Prec@5 44.353 | |
~~12 0.11212979277777778 51.363 | |
* Prec@1 27.034 Prec@5 51.363 | |
~~13 0.12001623805555554 52.278 | |
* Prec@1 28.365 Prec@5 52.278 | |
~~14 0.12776536916666667 58.288 | |
* Prec@1 32.461 Prec@5 58.288 | |
~~15 0.13556003333333333 55.339 | |
* Prec@1 30.832 Prec@5 55.339 | |
~~16 0.14310172361111112 43.704 | |
* Prec@1 22.241 Prec@5 43.704 | |
~~17 0.15069879472222222 50.603 | |
* Prec@1 26.868 Prec@5 50.603 | |
~~18 0.15827888055555556 47.502 | |
* Prec@1 24.490 Prec@5 47.502 | |
~~19 0.16583665944444445 59.217 | |
* Prec@1 34.087 Prec@5 59.217 | |
~~20 0.17353474861111112 61.899 | |
* Prec@1 35.802 Prec@5 61.899 | |
~~21 0.1811588575 64.057 | |
* Prec@1 37.712 Prec@5 64.057 | |
~~22 0.1889066238888889 62.748 | |
* Prec@1 36.257 Prec@5 62.748 | |
~~23 0.19647308111111111 58.500 | |
* Prec@1 32.661 Prec@5 58.500 | |
~~24 0.204021765 48.655 | |
* Prec@1 25.494 Prec@5 48.655 | |
~~25 0.21159265916666667 62.632 | |
* Prec@1 36.593 Prec@5 62.632 | |
~~26 0.21913230166666667 60.330 | |
* Prec@1 34.829 Prec@5 60.330 | |
~~27 0.22682289111111112 53.684 | |
* Prec@1 29.226 Prec@5 53.684 | |
~~28 0.23455073527777778 63.757 | |
* Prec@1 37.998 Prec@5 63.757 | |
~~29 0.24231406555555557 57.713 | |
* Prec@1 32.689 Prec@5 57.713 | |
~~30 0.24985262694444446 55.834 | |
* Prec@1 31.026 Prec@5 55.834 | |
~~31 0.27500571583333333 70.526 | |
* Prec@1 43.774 Prec@5 70.526 | |
~~32 0.2919476625 63.991 | |
* Prec@1 38.433 Prec@5 63.991 | |
~~33 0.30883913916666667 53.449 | |
* Prec@1 29.580 Prec@5 53.449 | |
~~34 0.3257044086111111 71.292 | |
* Prec@1 44.991 Prec@5 71.292 | |
~~35 0.3422650888888889 57.860 | |
* Prec@1 33.040 Prec@5 57.860 | |
~~36 0.35880357305555555 88.607 | |
* Prec@1 68.452 Prec@5 88.607 | |
~~37 0.3752278866666667 89.031 | |
* Prec@1 69.144 Prec@5 89.031 | |
~~38 0.39160111527777774 88.971 | |
* Prec@1 69.264 Prec@5 88.971 | |
~~39 0.40796110388888884 89.418 | |
* Prec@1 69.807 Prec@5 89.418 | |
~~40 0.4245721666666667 89.632 | |
* Prec@1 70.185 Prec@5 89.632 | |
~~41 0.44108383749999996 89.538 | |
* Prec@1 69.901 Prec@5 89.538 | |
~~42 0.4572433486111111 89.802 | |
* Prec@1 70.097 Prec@5 89.802 | |
~~43 0.47352948777777776 89.730 | |
* Prec@1 70.171 Prec@5 89.730 | |
~~44 0.48974440027777777 89.876 | |
* Prec@1 70.177 Prec@5 89.876 | |
~~45 0.5059532919444445 89.120 | |
* Prec@1 69.361 Prec@5 89.120 | |
~~46 0.522198375 89.484 | |
* Prec@1 69.939 Prec@5 89.484 | |
~~47 0.5385265116666667 89.778 | |
* Prec@1 70.342 Prec@5 89.778 | |
~~48 0.5549079252777778 89.344 | |
* Prec@1 69.613 Prec@5 89.344 | |
~~49 0.5710219061111111 88.723 | |
* Prec@1 68.932 Prec@5 88.723 | |
~~50 0.5872193766666667 88.389 | |
* Prec@1 68.043 Prec@5 88.389 | |
~~51 0.6034064450000001 89.456 | |
* Prec@1 69.763 Prec@5 89.456 | |
~~52 0.6194806166666667 89.080 | |
* Prec@1 69.082 Prec@5 89.080 | |
~~53 0.6356383030555556 88.467 | |
* Prec@1 68.217 Prec@5 88.467 | |
~~54 0.6516842002777778 88.479 | |
* Prec@1 68.352 Prec@5 88.479 | |
~~55 0.6677821680555556 88.873 | |
* Prec@1 68.658 Prec@5 88.873 | |
~~56 0.6840246997222222 88.939 | |
* Prec@1 68.974 Prec@5 88.939 | |
~~57 0.700372101388889 91.428 | |
* Prec@1 73.607 Prec@5 91.428 | |
~~58 0.7166581897222222 91.596 | |
* Prec@1 73.897 Prec@5 91.596 | |
~~59 0.7329163941666667 91.734 | |
* Prec@1 73.873 Prec@5 91.734 | |
~~60 0.7491146797222222 91.682 | |
* Prec@1 73.891 Prec@5 91.682 | |
~~61 0.7654112180555556 91.768 | |
* Prec@1 74.037 Prec@5 91.768 | |
~~62 0.7815943241666667 91.704 | |
* Prec@1 73.847 Prec@5 91.704 | |
~~63 0.7977814427777778 91.836 | |
* Prec@1 74.167 Prec@5 91.836 | |
~~64 0.8139362758333333 91.744 | |
* Prec@1 74.163 Prec@5 91.744 | |
~~65 0.8302545969444445 91.808 | |
* Prec@1 74.219 Prec@5 91.808 | |
~~66 0.84658592 91.756 | |
* Prec@1 74.195 Prec@5 91.756 | |
~~67 0.8806787052777778 92.859 | |
* Prec@1 75.859 Prec@5 92.859 | |
~~68 0.9042658733333333 92.783 | |
* Prec@1 75.701 Prec@5 92.783 | |
~~69 0.927664146111111 92.991 | |
* Prec@1 76.047 Prec@5 92.991 | |
~~70 0.9860236802777778 93.034 | |
* Prec@1 76.051 Prec@5 93.034 | |
~~71 1.0413164147222222 93.050 | |
* Prec@1 76.136 Prec@5 93.050 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment