Lines Matching full:docker

6 [TensorFlow's official Docker images](https://hub.docker.com/r/tensorflow/tensorflow)
11 # Using TensorFlow via Docker
14 TensorFlow via [Docker](http://www.docker.com/).
16 ## Installing Docker
19 [on the Docker site](https://docs.docker.com/installation/), but we give some
22 * [OSX](https://www.docker.com/products/docker#/mac)
23 * [Ubuntu](https://docs.docker.com/engine/installation/linux/ubuntulinux/)
27 We currently maintain two Docker container images:
35 [Docker Hub](https://hub.docker.com/r/tensorflow/tensorflow/tags/).
42     $ docker run -it -p 8888:8888 tensorflow/tensorflow
45 [nvidia-docker](https://github.com/NVIDIA/nvidia-docker). Run using
47     $ nvidia-docker run -it -p 8888:8888 tensorflow/tensorflow:latest-gpu
50 Note: If you would have a problem running nvidia-docker you may try the old method
51 we have used. But it is not recommended. If you find a bug in nvidia-docker, please report
52 it there and try using nvidia-docker as described above.
54     $ # The old, not recommended way to run docker with gpu support:
57     $ docker run -it -p 8888:8888 $CUDA_SO $DEVICES tensorflow/tensorflow:latest-gpu
62 See all available [tags](https://hub.docker.com/r/tensorflow/tensorflow/tags/)
68 Building TensorFlow Docker containers should be done through the
69 …h](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/docker/parameterized_dock…
73 …h](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/docker/parameterized_dock…
74 from a binary docker image such as for example `tensorflow/tensorflow:latest` will
75 not work. One needs to execute the script from a developer docker image since by
76 contrast with a binary docker image it contains not only the compiled solution but
77 also the tensorflow source code. Please select the appropriate developer docker
78 image of tensorflow at `tensorflow/tensorflow:[.](https://hub.docker.com/r/tensorflow/tensorflow/ta…
80 The smallest command line to generate a docker image will then be:
81 ```docker run -it tensorflow/tensorflow:"right_tag"```
83 If you would like to start a jupyter notebook on your docker container, make sure
84 to map the port 8888 of your docker container by adding -p 8888:8888 to the above
88 Python version (`PYTHON2` vs. `PYTHON3`) and whether the developer Docker image
92 For example, to build a CPU-only non-developer Docker image for Python 2, using
105 tensorflow/tools/docker/parameterized_docker_build.sh
110 Rebuilding GPU images requires [nvidia-docker](https://github.com/NVIDIA/nvidia-docker).