[chibi@archlinux ~]$ sudo nvidia-docker run --rm -ti nvcr.io/nvidia/tensorflow:18.04-py3 [sudo] chibi のパスワード: ================ == TensorFlow == ================ NVIDIA Release 18.04 (build 377127) Container image Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved. Copyright 2017 The TensorFlow Authors. All rights reserved. Various files include modifications (c) NVIDIA CORPORATION. All rights reserved. NVIDIA modifications are covered by the license terms that apply to the underlying project or file. NOTE: The SHMEM allocation limit is set to the default of 64MB. This may be insufficient for TensorFlow. NVIDIA recommends the use of the following flags: nvidia-docker run --shm-size=1g --ulimit memlock=-1 --ulimit stack=67108864 ... root@d77f3fdcf198:/workspace# ls README.md docker-examples nvidia-examples root@d77f3fdcf198:/workspace# cd nvidia-examples/cnn root@d77f3fdcf198:/workspace/nvidia-examples/cnn# time python nvcnn.py --model=googlenet \ > --batch_size=256 \ > --num_gpus=1 \ > --fp16 WARNING:tensorflow:From /usr/local/lib/python3.5/dist-packages/tensorflow/contrib/learn/python/learn/datasets/base.py:198: retry (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version. Instructions for updating: Use the retry module or similar alternatives. TensorFlow: 1.7.0 This script: nvcnn.py v1.6 Cmd line args: --model=googlenet --batch_size=256 --num_gpus=1 --fp16 Num images: Synthetic Model: googlenet Batch size: 256 global 256.0 per device Devices: ['/gpu:0'] Data format: NCHW Data type: fp16 Have NCCL: True Using NCCL: True Using XLA: False Building training graph Creating session 2018-06-09 06:51:06.626030: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1344] Found device 0 with properties: name: TITAN V major: 7 minor: 0 memoryClockRate(GHz): 1.455 pciBusID: 0000:02:00.0 totalMemory: 11.78GiB freeMemory: 11.25GiB 2018-06-09 06:51:06.831653: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1344] Found device 1 with properties: name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.582 pciBusID: 0000:82:00.0 totalMemory: 10.92GiB freeMemory: 10.76GiB 2018-06-09 06:51:07.021042: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1344] Found device 2 with properties: name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.582 pciBusID: 0000:83:00.0 totalMemory: 10.92GiB freeMemory: 10.76GiB 2018-06-09 06:51:07.022487: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1423] Adding visible gpu devices: 0, 1, 2 2018-06-09 06:51:09.103310: I tensorflow/core/common_runtime/gpu/gpu_device.cc:911] Device interconnect StreamExecutor with strength 1 edge matrix: 2018-06-09 06:51:09.103389: I tensorflow/core/common_runtime/gpu/gpu_device.cc:917] 0 1 2 2018-06-09 06:51:09.103398: I tensorflow/core/common_runtime/gpu/gpu_device.cc:930] 0: N N N 2018-06-09 06:51:09.103404: I tensorflow/core/common_runtime/gpu/gpu_device.cc:930] 1: N N Y 2018-06-09 06:51:09.103412: I tensorflow/core/common_runtime/gpu/gpu_device.cc:930] 2: N Y N 2018-06-09 06:51:09.104904: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1041] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10871 MB memory) -> physical GPU (device: 0, name: TITAN V, pci bus id: 0000:02:00.0, compute capability: 7.0) 2018-06-09 06:51:09.290784: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1041] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 10415 MB memory) -> physical GPU (device: 1, name: GeForce GTX 1080 Ti, pci bus id: 0000:82:00.0, compute capability: 6.1) 2018-06-09 06:51:09.479407: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1041] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:2 with 10415 MB memory) -> physical GPU (device: 2, name: GeForce GTX 1080 Ti, pci bus id: 0000:83:00.0, compute capability: 6.1) Initializing variables Pre-filling input pipeline Training Step Epoch Img/sec Loss LR 1 1 20.0 7.369 0.04000 2 1 75.6 7.369 0.03842 3 1 161.9 7.369 0.03686 4 1 164.8 7.369 0.03534 5 1 426.1 7.369 0.03386 6 1 425.4 7.369 0.03240 7 1 695.3 7.369 0.03098 8 1 696.2 7.369 0.02958 9 1 1205.7 7.369 0.02822 10 1 1205.8 7.369 0.02690 11 1 1346.4 7.369 0.02560 12 1 1351.4 7.369 0.02434 13 1 1350.1 7.369 0.02310 14 1 1347.3 7.369 0.02190 15 1 1339.8 7.369 0.02074 16 1 1361.8 7.369 0.01960 17 1 1353.5 7.369 0.01850 18 1 1350.8 7.369 0.01742 19 1 1346.8 7.369 0.01638 20 1 1348.9 7.369 0.01538 21 1 1346.3 7.369 0.01440 22 1 1359.0 7.369 0.01346 23 1 1349.3 7.369 0.01254 24 1 1352.8 7.369 0.01166 25 1 1349.0 7.369 0.01082 26 1 1348.0 7.369 0.01000 27 1 1347.7 7.369 0.00922 28 1 1353.4 7.369 0.00846 29 1 1353.7 7.369 0.00774 30 1 1352.3 7.369 0.00706 31 1 1347.8 7.369 0.00640 32 1 1349.8 7.369 0.00578 33 1 1347.6 7.369 0.00518 34 1 1347.1 7.369 0.00462 35 1 1351.0 7.369 0.00410 36 1 1353.3 7.369 0.00360 37 1 1348.5 7.369 0.00314 38 1 1351.3 7.369 0.00270 39 1 1345.2 7.369 0.00230 40 1 1345.4 7.369 0.00194 41 1 1348.4 7.369 0.00160 42 1 1345.9 7.368 0.00130 43 1 1341.9 7.368 0.00102 44 1 1348.5 7.368 0.00078 45 1 1346.4 7.368 0.00058 46 1 1351.0 7.368 0.00040 47 1 1356.8 7.369 0.00026 48 1 1343.7 7.369 0.00014 49 1 1352.5 7.368 0.00006 50 1 1355.8 7.368 0.00002 ---------------------------------------------------------------- Images/sec: 1349.7 +/- 0.7 (jitter = 3.7) ---------------------------------------------------------------- real 0m40.439s user 0m27.049s sys 0m17.376s root@d77f3fdcf198:/workspace/nvidia-examples/cnn# cat /etc/os-release NAME="Ubuntu" VERSION="16.04.3 LTS (Xenial Xerus)" ID=ubuntu ID_LIKE=debian PRETTY_NAME="Ubuntu 16.04.3 LTS" VERSION_ID="16.04" HOME_URL="http://www.ubuntu.com/" SUPPORT_URL="http://help.ubuntu.com/" BUG_REPORT_URL="http://bugs.launchpad.net/ubuntu/" VERSION_CODENAME=xenial UBUNTU_CODENAME=xenial root@d77f3fdcf198:/workspace/nvidia-examples/cnn# nvcc -V nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2017 NVIDIA Corporation Built on Fri_Sep__1_21:08:03_CDT_2017 Cuda compilation tools, release 9.0, V9.0.176 root@d77f3fdcf198:/workspace/nvidia-examples/cnn# exit exit [chibi@archlinux ~]$ cat /etc/os-release NAME="Arch Linux" PRETTY_NAME="Arch Linux" ID=arch ID_LIKE=archlinux ANSI_COLOR="0;36" HOME_URL="https://www.archlinux.org/" SUPPORT_URL="https://bbs.archlinux.org/" BUG_REPORT_URL="https://bugs.archlinux.org/" [chibi@archlinux ~]$ nvcc -V nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2018 NVIDIA Corporation Built on Wed_Apr_11_23:16:29_CDT_2018 Cuda compilation tools, release 9.2, V9.2.88 [chibi@archlinux ~]$