Many data engineers, business analysts and data scientists are developing data models and analyzing data in the containerized enterprise AI and Data platform IBM Cloud Pak for Data. It provides two very important components Watson Studio and Watson Machine Learning for data science projects. Watson Studio is the comprehensive toolkit for data engineers and data scientists to develop models with the popular tools like Jupyter notebook, SPSS modeler, AutoAI, Data Refinery etc. Watson Machine Learning (WML) is the advanced model deployment toolkit with deployment space management and easy model deployment capabilities. Meanwhile, each component provides broad range of RESTful APIs for uses to access and integrate with Cloud Pak for Data.

from ibm_watson_machine_learning import APIClient
from project_lib.utils import environment
url = environment.get_common_api_url()

import sys,os,os.path
token = os.environ['USER_ACCESS_TOKEN']

wml_credentials = {
     "instance_id": "openshift",
     "token": token,
     "url": url,
     "version": "3.5"
}

client = APIClient(wml_credentials)​

space_name = 'churn-analysis-space'
space_uid = ''

for space in client.spaces.get_details()['resources']:
    if space['entity']['name'] ==space_name:
        space_uid=space['metadata']['id']

if space_uid == '':
    space_meta_data = {
        client.spaces.ConfigurationMetaNames.NAME : space_name
        }
    stored_space_details = client.spaces.store(space_meta_data)
    space_uid = stored_space_details['metadata']['id']

space_uid

client.set.default_space(space_uid)

# Model Metadata
software_spec_uid = client.software_specifications.get_uid_by_name('default_py3.7')

meta_props={
    client.repository.ModelMetaNames.NAME: "churn-sklearn-model",
    client.repository.ModelMetaNames.SOFTWARE_SPEC_UID: software_spec_uid,
    client.repository.ModelMetaNames.TYPE: "scikit-learn_0.23"
}

model_artifact = client.repository.store_model(logreg_pipe,
                                               meta_props=meta_props,
                                               training_data=X_train,
                                               training_target=y_train)
model_uid = client.repository.get_model_uid(model_artifact)
print("Model UID = " + model_uid)

# deploy the model pipeline
metadata_fields = X_train.columns.tolist()

meta_props = {
    client.deployments.ConfigurationMetaNames.NAME: "churn-sklearn-model",
    client.deployments.ConfigurationMetaNames.ONLINE: {},
    client.deployments.ConfigurationMetaNames.CUSTOM: metadata_fields
}

# deploy the model
deployment_details = client.deployments.create( artifact_uid=model_uid, meta_props=meta_props)

deployment_uid=client.deployments.get_uid(deployment_details)

fields = ["AGE", "ACTIVITY", "EDUCATION", "NEGTWEETS" , "INCOME", "SEX", "STATE"]
values = [(30, 1, 1, 12, 30000, 'M', 'TX'), (40, 2, 3, 2, 80000, 'F', 'TX')]
payload_scoring={"input_data": [{"fields": fields,"values": values}]}

payload = {
    client.deployments.ScoringMetaNames.INPUT_DATA: [payload_scoring]
}

# Pass the payload to wml client to predict churn for the sample records
scoring_response = client.deployments.score(deployment_uid, payload_scoring)
scoring_response