Docker 的 TensorFlow

Question

我已经为运行 TensorFlow 创建了一个 Docker 机器：

$ docker-machine create --driver virtualbox tensorflow
$ eval $(docker-machine env tensorflow)
$ docker-machine ls
NAME               ACTIVE   URL          STATE     URL                         SWARM   DOCKER   ERRORS  
tensorflow   *        virtualbox   Running   tcp://193.168.99.101:2376           v1.9.1

然后我尝试使用 TesnorFlow 开发镜像启动一个容器：

$ docker run -it b.gcr.io/tensorflow/tensorflow:latest-devel

但我得到以下打印输出和错误：

Unable to find image 'b.gcr.io/tensorflow/tensorflow:latest-devel' locally
latest-devel: Pulling from tensorflow/tensorflow
d5abe3500fe7: Verifying Checksum 
b4906bf7cb96: Download complete 
bfd1864b7d78: Verifying Checksum 
b5834ff0edff: Download complete 
37c70795c272: Download complete 
fd1295db72c3: Verifying Checksum 
0851dd369c75: Download complete 
f090bc83193e: Download complete 
3505c3e79922: Download complete 
5f7f94f4a89a: Download complete 
82d880401459: Download complete 
3f8ab9378c1a: Verifying Checksum 
cd2e7af6dd36: Verifying Checksum 
0fbd1aaa7f93: Download complete 
a073e9eadc31: Verifying Checksum 
eac8972ce5cd: Download complete 
d20000bda44f: Download complete 
7abc7a47a494: Download complete 
6a90f0a0e005: Download complete 
Pulling repository b.gcr.io/tensorflow/tensorflow
74c849ee103e: Error pulling image (latest-devel) from b.gcr.io/tensorflow/tensorflow, Untar re-exec error: exit status 1: output: write /usr/lib/gcc/x86_64-linux-gnu/4.8/libitm.a: no space left on device bitm.a: no space left on device 
a82f81f25750: Download complete 
b207c06aba70: Download complete 
d55e68e6cc9c: Download complete 
0aa714ad37e0: Download complete 
28f77615a692: Error downloading dependent layers 
Error pulling image (latest-devel) from b.gcr.io/tensorflow/tensorflow, Untar re-exec error: exit status 1: output: write /usr/lib/gcc/x86_64-linux-gnu/4.8/libitm.a: no space left on device

我不知道该怎么办...有人能解释一下这意味着什么以及我应该如何进行吗？

额外信息：我在 Mac OSX 10.9，

$ docker info
Containers: 0
Images: 12
Server Version: 1.9.1
Storage Driver: aufs
 Root Dir: /mnt/sda1/var/lib/docker/aufs
 Backing Filesystem: tmpfs
 Dirs: 12
 Dirperm1 Supported: true
Execution Driver: native-0.2
Logging Driver: json-file
Kernel Version: 4.1.13-boot2docker
Operating System: Boot2Docker 1.9.1 (TCL 6.4.1); master : cef800b - Fri Nov 20 19:33:59 UTC 2015
CPUs: 1
Total Memory: 996.2 MiB
Name: tensorflow
Debug mode (server): true
 File Descriptors: 11
 Goroutines: 19
 System Time: 2016-01-24T02:54:29.677797424Z
 EventsListeners: 0
 Init SHA1: 
 Init Path: /usr/local/bin/docker
 Docker Root Dir: /mnt/sda1/var/lib/docker
Labels:
 provider=virtualbox

Answer 1

而不是默认的 docker 创建，试试这个：

docker-machine create --driver virtualbox --virtualbox-disk-size "20000" tensorflow

然后：

eval $(docker-machine env tensorflow)
docker run -it b.gcr.io/tensorflow/tensorflow:latest-devel

注意：

见Docker increase disk space
确保您的物理磁盘上有 20 GB

Docker 的 TensorFlow

TensorFlow with Docker

docker

tensorflow