1 ### Running an official image
3 You can run one of the automatic [builds](https://hub.docker.com/r/bvlc/caffe). E.g. for the CPU version:
5 `docker run -ti bvlc/caffe:cpu caffe --version`
7 or for GPU support (You need a CUDA 8.0 capable driver and
8 [nvidia-docker](https://github.com/NVIDIA/nvidia-docker)):
10 `nvidia-docker run -ti bvlc/caffe:gpu caffe --version`
12 You might see an error about libdc1394, ignore it.
14 ### Docker run options
16 By default caffe runs as root, thus any output files, e.g. snapshots, will be owned
17 by root. It also runs by default in a container-private folder.
19 You can change this using flags, like user (-u), current directory, and volumes (-w and -v).
20 E.g. this behaves like the usual caffe executable:
22 `docker run --rm -u $(id -u):$(id -g) -v $(pwd):$(pwd) -w $(pwd) bvlc/caffe:cpu caffe train --solver=example_solver.prototxt`
24 Containers can also be used interactively, specifying e.g. `bash` or `ipython`
28 docker run -ti bvlc/caffe:cpu ipython
33 The caffe build requirements are included in the container, so this can be used to
34 build and run custom versions of caffe. Also, `caffe/python` is in PATH, so python
35 utilities can be used directly, e.g. `draw_net.py`, `classify.py`, or `detect.py`.
37 ### Building images yourself
41 `docker build -t caffe:cpu cpu`
43 `docker build -t caffe:gpu gpu`
45 You can also build Caffe and run the tests in the image:
47 `docker run -ti caffe:cpu bash -c "cd /opt/caffe/build; make runtest"`