arieling / vagrantmachinelearningautobox

Vagrant－VirtualBox－WorkStation

A virtual box script containing Tensorflow-CPU, hadoop docker, python27, numpy and other machine leanring tools.

How to Istall

1. Install virtual box.
2. Install vargant

How to use:

1. Go to directory with Vagrantfile.
2. Type $ vagrant up
3. Go and grabe a coffee, wait for a while.
4. Type $ vagrant halt and $ vagrant up again.

Set up Hadoop Clusters:

1. $ cd /docker_files/docker-ambari
2. switch to bash $ bash
3. Activate docker functions $ source ambari-functions
4. Type $ amb-settings to make sure amb functions are working.
5. Start 3 clusters $ amb-start-cluster 3
6. Go to ambari shell $ amb-shell
7. type host list to make sure clusters are up.
8. add blueprint blueprint add --url https://raw.githubusercontent.com/sequenceiq/ambari-rest-client/2.2.11/src/main/resources/blueprints/multi-node-hdfs-yarn
9. cluster autoAssign
10. cluster create

Use tensorflow:

1. $ source activate tensorflow
2. $ python
3. > import tensorflow as tf

Useful vagrant command:

1. $ vagrant up Setup virtual machine.
2. $ vagrant halt Shut down virtual machine.
3. $ vagrant provision Execute bootstrap.sh for debugging.

License

MIT License