example-plugins
Directory actions
More options
Directory actions
More options
example-plugins
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
parent directory.. | ||||
To compile the project and create the sample jars do
mvn package
you should see a example-plugins.jar in the target directory.
The hello world plugin will add a method to the frame class that prints hello world.
The word-count plugin will add a method to the frame class that counts words in the frame.
The vertex-degree count plugin will add a method to the graph class that counts the vertex out-degree.
To register the plugins with the Analytics Toolkit, update the following configuration files:
src/main/resources/atk-module.conf
src/main/resources/atk-plugin.conf
The 3 mentioned plugins are commented out in the default atk-plugin.conf file.
To enable them, uncomment those lines and rebuild.
Installing plugins:
To add the plugins to your current installation of trusted analytics copy the example-plugins.jar to
/usr/lib/trustedanalytics/lib/. Plugins should go in a jar with "plugins" in the name of the jar.
Now restart the trusted analytics service and try the code example below.
sudo service trustedanalytics restart
Sample code:
from trustedanalytics import *
#the default home directory is hdfs://user/atkuser all the sample data sets are saved to hdfs://user/atkuser/datasets
dataset = r"datasets/movie_data_random.csv"
#csv schema definition
schema = [("user_id", int32),
("movie_id", int32),
("rating", int32),
("splits", str)]
csv_file = CsvFile(dataset, schema, skip_header_lines=1)
print "Building data frame"
frame = Frame(csv_file)
print "View auto-generated Python documentation for plugin"
help(frame.helloworld)
print "Run hello-world plugin"
frame.helloworld()
>>u'Frame says Hello World'
frame.helloworld("test")
>> u'Frame says test'
More examples are available in the python-examples folder.