Difference between revisions of "Meteos/ExampleKmeans"
(→6. Create a learning job) |
(→7. Online Prediction) |
||
(2 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
== Classify Users by Skill using Meteos == | == Classify Users by Skill using Meteos == | ||
− | In this example, you create a prediction model which classify users by | + | In this example, you create a prediction model which classify users by skill using KMeans Model. |
[[ File:KMeans.png ]] | [[ File:KMeans.png ]] | ||
Line 328: | Line 328: | ||
+--------------------------------------+----------------------+-----------+---------------------+--------+ | +--------------------------------------+----------------------+-----------+---------------------+--------+ | ||
</pre> | </pre> | ||
+ | |||
+ | === 7. Online Prediction === | ||
+ | |||
+ | You can load a Prediction Model in advance for online prediction by using "meteos-load" command. | ||
+ | |||
+ | In online prediction, user can retrieve a predicted data immediately. | ||
+ | |||
+ | For more details, please see [https://wiki.openstack.org/wiki/Meteos/ExampleLinear#7._Online_Prediction here]. |
Latest revision as of 06:51, 10 January 2017
Contents
Classify Users by Skill using Meteos
In this example, you create a prediction model which classify users by skill using KMeans Model.
1. Create a experiment template
Create template of experiment. Experiment is a workspace of Machine Learning.
You have to confirm a glance image id of meteos image, and a neutron network id before creating a template.
You can use a format located in python-meteosclient/sample/json/template.json
$ glance image-list | grep meteos | a6b7de0b-10ff-462c-9c86-25c8a5622a57 | meteos |
$ neutron net-list | grep public | 4222b557-6d9f-405c-b1ff-0f454d2f35bf | public | 1f979ae3-d6b7-4d03-ba0f-9d9112581783 |
$ vim sample/json/template.json
$ cat sample/json/template.json { "display_name": "example-template", "display_description": "This is a sample template of experiment", "image_id" : "a6b7de0b-10ff-462c-9c86-25c8a5622a57", "master_nodes_num": 1, "master_flavor_id": "4", "worker_nodes_num": 2, "worker_flavor_id": "2", "spark_version": "1.6.0", "floating_ip_pool": "4222b557-6d9f-405c-b1ff-0f454d2f35bf" }
$ meteos template-create --json sample/json/template.json +---------------+-----------------------------------------+ | Property | Value | +---------------+-----------------------------------------+ | cluster_id | None | | created_at | 2016-12-15T22:55:03.000000 | | description | This is a sample template of experiment | | id | d3180a78-74cc-432d-9e9d-68640b18abae | | master_flavor | 4 | | master_nodes | 1 | | name | example-template | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | spark_version | 1.6.0 | | status | available | | user_id | 64ad04e101df40b4b51e5f370a83412c | | worker_flavor | 2 | | worker_nodes | 2 | +---------------+-----------------------------------------+
2. Create a experiment from template
Create a experiment by using template created in the above step. You have to confirm a neutron private network id and create keypair before creating a template.
You can use a format located in python-meteosclient/sample/json/experiment.json
$ nova keypair-add key1 > ~/key1.pem && chmod 600 ~/key1.pem
$ neutron net-list | grep private | e4fcc49c-48e5-48f8-9599-bb5eba1339c9 | private | e15c24a5-dfdd-4428-b27d-9827b35600c0 10.0.0.0/26 |
$ vim sample/json/experiment.json
$ cat sample/json/experiment.json { "display_name": "example-experiment", "display_description": "This is a sample experiment", "key_name": "key1", "neutron_management_network": "e4fcc49c-48e5-48f8-9599-bb5eba1339c9", "template_id": "d3180a78-74cc-432d-9e9d-68640b18abae" }
$ meteos experiment-create --json sample/json/experiment.json +--------------------+--------------------------------------+ | Property | Value | +--------------------+--------------------------------------+ | created_at | 2016-12-15T22:56:51.000000 | | description | This is a sample experiment | | id | fcc8d055-e801-4652-af8c-5aabedbf0286 | | key_name | key1 | | management_network | e4fcc49c-48e5-48f8-9599-bb5eba1339c9 | | name | example-experiment | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | status | creating | | user_id | 64ad04e101df40b4b51e5f370a83412c | +--------------------+--------------------------------------+
Meteos creates a experiment using OpenStack Sahara spark plugin.
You can see a sahara cluster and nova VMs created by Meteos as below.
$ openstack dataprocessing cluster list (or sahara cluster-list) +------------------+--------------------------------------+-------------+----------------+----------+ | Name | Id | Plugin name | Plugin version | Status | +------------------+--------------------------------------+-------------+----------------+----------+ | cluster-fcc8d055 | 5736d157-ac7c-41de-8aca-78f7afa7e99c | spark | 1.6.0 | Spawning | +------------------+--------------------------------------+-------------+----------------+----------+
$ openstack server list (or nova list) +--------------------------------------+----------------------------+--------+------------+-------------+------------------+ | ID | Name | Status | Task State | Power State | Networks | +--------------------------------------+----------------------------+--------+------------+-------------+------------------+ | 48a9f429-7756-4bed-8dd6-6dc6140ef897 | cluster-fcc8d055-master-0 | ACTIVE | - | Running | private=10.0.0.5 | | 88ff2070-dfe8-45da-aa5c-02ac3e9de3b8 | cluster-fcc8d055-workers-0 | ACTIVE | - | Running | private=10.0.0.7 | | a57dfa5d-8b55-47c7-aae7-d5b3c8779787 | cluster-fcc8d055-workers-1 | ACTIVE | - | Running | private=10.0.0.4 | +--------------------------------------+----------------------------+--------+------------+-------------+------------------+
3. Upload a raw data
Upload a raw data (in this example user skill data) to OpenStack Swift.
You can use a sample data located in python-meteosclient/sample/data/kmeans_data.txt
Raw data shows "UserID", "Rank of Skill A", "Rank of Skill A", ... from left.
$ cd sample/data/
/sample/data$ head kmeans_data.txt 1,1.0,1.0,0.0,0.0,0.0 2,3.0,3.0,2.0,3.0,3.0 3,5.0,5.0,5.0,4.0,5.0 4,7.0,7.0,6.0,6.0,7.0 5,7.0,8.0,7.0,8.0,8.0 6,0.0,1.0,1.0,0.0,1.0 7,3.0,2.0,2.0,2.0,3.0 8,4.0,4.0,4.0,5.0,5.0 9,6.0,6.0,6.0,7.0,6.0 10,8.0,8.0,8.0,7.0,7.0
/sample/data$ swift upload meteos kmeans_data.txt kmeans_data.txt
4. Parse a raw data
Parse a raw data to enable Prediction Model to handle it.
KMeans model requires only parameters to classify, so you have to elminate UserID from raw data using map method.
You can use a sample format located in python-meteosclient/sample/json/dataset_parse.json
You can see the head data of parsed dataset by executing "meteos dataset-show <dataset-uuid>" command.
$ vim ../python-meteosclient/sample/json/dataset_parse.json
$ cat ../python-meteosclient/sample/json/dataset_parse.json { "source_dataset_url": "swift://meteos/kmeans_data.txt", "display_name": "sample-data", "display_description": "This is a sample dataset", "method": "parse", "params": [{"method": "map", "args": "lambda l: l.split(',',1)[1]"}], "experiment_id": "fcc8d055-e801-4652-af8c-5aabedbf0286", "swift_tenant": "demo", "swift_username": "demo", "swift_password": "nova" }
$ meteos dataset-create --json sample/json/dataset_parse.json +-------------+--------------------------------------+ | Property | Value | +-------------+--------------------------------------+ | created_at | 2016-12-15T23:36:11.000000 | | description | This is a sample dataset | | head | None | | id | b576d46c-01f7-4020-903b-eb79662e3faa | | name | sample-data | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | status | creating | | stderr | None | | user_id | 64ad04e101df40b4b51e5f370a83412c | +-------------+--------------------------------------+
$ meteos dataset-list +--------------------------------------+-------------+-----------+--------------------------------+ | id | name | status | source_dataset_url | +--------------------------------------+-------------+-----------+--------------------------------+ | b576d46c-01f7-4020-903b-eb79662e3faa | sample-data | available | swift://meteos/kmeans_data.txt | +--------------------------------------+-------------+-----------+--------------------------------+
$ meteos dataset-show b576d46c-01f7-4020-903b-eb79662e3faa +-------------+--------------------------------------------------+ | Property | Value | +-------------+--------------------------------------------------+ | created_at | 2016-12-15T23:36:11.000000 | | description | This is a sample dataset | | head | [u'1.0,1.0,0.0,0.0,0.0', u'3.0,3.0,2.0,3.0,3.0', | | | u'5.0,5.0,5.0,4.0,5.0', u'7.0,7.0,6.0,6.0,7.0', | | | u'7.0,8.0,7.0,8.0,8.0', u'0.0,1.0,1.0,0.0,1.0', | | | u'3.0,2.0,2.0,2.0,3.0', u'4.0,4.0,4.0,5.0,5.0', | | | u'6.0,6.0,6.0,7.0,6.0', u'8.0,8.0,8.0,7.0,7.0'] | | id | b576d46c-01f7-4020-903b-eb79662e3faa | | name | sample-data | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | status | available | | stderr | | | user_id | 64ad04e101df40b4b51e5f370a83412c | +-------------+--------------------------------------------------+
5. Create a prediction model
In this example, User creates a Kmeans Model from parsed dataset.
Parsed dataset has been already distributed in hdfs of experiment environment.
So, you spefity the internal url (internal://<dataset-id>) in source_dataset_url parameter.
And you can spefity a numClasses parameter in model_params which shows how many classes you want to classify.
$ vim sample/json/model_kmeans.json
$ cat sample/json/model_kmeans.json { "display_name": "kmeans-model", "display_description": "Sample KMeans Model", "source_dataset_url": "internal://b576d46c-01f7-4020-903b-eb79662e3faa", "model_type": "KMeans", "model_params": "{'numIterations': 5, 'numClasses':5}", "experiment_id": "fcc8d055-e801-4652-af8c-5aabedbf0286" }
$ meteos model-create --json sample/json/model_kmeans.json +-------------+--------------------------------------------------+ | Property | Value | +-------------+--------------------------------------------------+ | created_at | 2016-12-15T23:40:23.000000 | | description | Sample KMeans Model | | id | ef6ba029-c744-4f9f-a702-00ac0098d437 | | name | kmeans-model | | params | eydudW1JdGVyYXRpb25zJzogNSwgJ251bUNsYXNzZXMnOjV9 | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | status | creating | | stderr | None | | stdout | None | | type | KMeans | | user_id | 64ad04e101df40b4b51e5f370a83412c | +-------------+--------------------------------------------------+
$ meteos model-list +--------------------------------------+--------------+-----------+--------+-------------------------------------------------+ | id | name | status | type | source_dataset_url | +--------------------------------------+--------------+-----------+--------+-------------------------------------------------+ | ef6ba029-c744-4f9f-a702-00ac0098d437 | kmeans-model | available | KMeans | internal://b576d46c-01f7-4020-903b-eb79662e3faa | +--------------------------------------+--------------+-----------+--------+-------------------------------------------------+
6. Create a learning job
Create learning job and retrieve class.
Retrieve a output data as a stdout of job execution.
The Class number (stdout) outputted by KMeans Model just indicates a number to classify, not inidicatea a rank.
$ vim sample/json/learning.json
$ cat sample/json/learning.json { "display_name": "example-learning-job", "display_description": "This is a sample job", "model_id": "ef6ba029-c744-4f9f-a702-00ac0098d437", "method": "predict", "args": "7.0,8.0,7.0,8.0,8.0" }
$ meteos learning-create --json sample/json/learning.json +-------------+--------------------------------------+ | Property | Value | +-------------+--------------------------------------+ | args | Ny4wLDguMCw3LjAsOC4wLDguMA== | | created_at | 2016-12-15T23:46:35.000000 | | description | This is a sample job | | id | b236af4a-862f-415d-9e23-f020232fe5cf | | method | predict | | name | example-learning-job | | project_id | 4034bb3cd9324776a68c140fdd18baa4 | | status | creating | | stderr | None | | stdout | None | | user_id | 64ad04e101df40b4b51e5f370a83412c | +-------------+--------------------------------------+
$ meteos learning-list +--------------------------------------+----------------------+-----------+---------------------+--------+ | id | name | status | args | stdout | +--------------------------------------+----------------------+-----------+---------------------+--------+ | 13699fda-36f2-45eb-81ec-dc642dbaccf9 | example-learning-job | available | 1.0,2.0,1.0,3.0,2.0 | 1 | | 7e2a98bd-1ab3-44bb-bbe8-6e4860e0f7f4 | example-learning-job | available | 4.0,4.0,4.0,4.0,4.0 | 2 | | b236af4a-862f-415d-9e23-f020232fe5cf | example-learning-job | available | 7.0,8.0,7.0,8.0,8.0 | 4 | +--------------------------------------+----------------------+-----------+---------------------+--------+
7. Online Prediction
You can load a Prediction Model in advance for online prediction by using "meteos-load" command.
In online prediction, user can retrieve a predicted data immediately.
For more details, please see here.