We currently maintain two Docker container images:
-* `gcr.io/tensorflow/tensorflow` - TensorFlow with all dependencies - CPU only!
+* `tensorflow/tensorflow` - TensorFlow with all dependencies - CPU only!
-* `gcr.io/tensorflow/tensorflow:latest-gpu` - TensorFlow with all dependencies
+* `tensorflow/tensorflow:latest-gpu` - TensorFlow with all dependencies
and support for NVidia CUDA
-Note: We also publish the same containers into
+Note: We store all our containers on
[Docker Hub](https://hub.docker.com/r/tensorflow/tensorflow/tags/).
Run non-GPU container using
- $ docker run -it -p 8888:8888 gcr.io/tensorflow/tensorflow
+ $ docker run -it -p 8888:8888 tensorflow/tensorflow
For GPU support install NVidia drivers (ideally latest) and
[nvidia-docker](https://github.com/NVIDIA/nvidia-docker). Run using
- $ nvidia-docker run -it -p 8888:8888 gcr.io/tensorflow/tensorflow:latest-gpu
+ $ nvidia-docker run -it -p 8888:8888 tensorflow/tensorflow:latest-gpu
Note: If you would have a problem running nvidia-docker you may try the old method
$ # The old, not recommended way to run docker with gpu support:
$ export CUDA_SO=$(\ls /usr/lib/x86_64-linux-gnu/libcuda.* | xargs -I{} echo '-v {}:{}')
$ export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
- $ docker run -it -p 8888:8888 $CUDA_SO $DEVICES gcr.io/tensorflow/tensorflow:latest-gpu
+ $ docker run -it -p 8888:8888 $CUDA_SO $DEVICES tensorflow/tensorflow:latest-gpu
## More containers