A virtual box script containing Tensorflow-CPU, hadoop docker, python27, numpy and other machine leanring tools.
- Install virtual box.
- Install vargant
- Go to directory with Vagrantfile.
- Type
$ vagrant up - Go and grabe a coffee, wait for a while.
- Type
$ vagrant haltand$ vagrant upagain.
$ cd /docker_files/docker-ambari- switch to bash
$ bash - Activate docker functions
$ source ambari-functions - Type
$ amb-settingsto make sure amb functions are working. - Start 3 clusters
$ amb-start-cluster 3 - Go to ambari shell
$ amb-shell - type
host listto make sure clusters are up. - add blueprint
blueprint add --url https://raw.githubusercontent.com/sequenceiq/ambari-rest-client/2.2.11/src/main/resources/blueprints/multi-node-hdfs-yarn cluster autoAssigncluster create
$ source activate tensorflow$ python> import tensorflow as tf
$ vagrant upSetup virtual machine.$ vagrant haltShut down virtual machine.$ vagrant provisionExecute bootstrap.sh for debugging.
MIT License