Train an Unsupervised Model¶

First, initialize a client with your server credentials and store them in the variable dai.

In [ ]:

import driverlessai

dai = driverlessai.Client(address='http://mr-dl28:12353/', username='user', password='user')

import driverlessai dai = driverlessai.Client(address='http://mr-dl28:12353/', username='user', password='user')

Loading Data¶

Import the iris.csv file from S3 to the Driverless AI server and name the dataset 'iris-getting-started'.

In [ ]:

ds = dai.datasets.create(
    data='s3://h2o-public-test-data/smalldata/iris/iris.csv',
    data_source='s3',
    name='iris-getting-started'
)

ds = dai.datasets.create( data='s3://h2o-public-test-data/smalldata/iris/iris.csv', data_source='s3', name='iris-getting-started' )

Running Unsupervised Experiment¶

Next, create an experiment preview. To run an unsupervised experiment, you need to set task='unsupervised' and specify which unsupervised algorithms to use. For example, models=['KMeans', ...]

In [ ]:

dai.experiments.preview(
    train_dataset=ds,
    target_column=None,
    task="unsupervised",
    models=['KMeans'],
    name='unsupervised_model_101'
)

dai.experiments.preview( train_dataset=ds, target_column=None, task="unsupervised", models=['KMeans'], name='unsupervised_model_101' )

Finally, run the experiment:

In [ ]:

ex = dai.experiments.create(
    train_dataset=ds,
    target_column=None,
    task="unsupervised",
    models=['KMeans'],
    name='unsupervised_model_101'
)

ex = dai.experiments.create( train_dataset=ds, target_column=None, task="unsupervised", models=['KMeans'], name='unsupervised_model_101' )