Use AutoAI to predict credit risk with ibm-watson-machine-learning

This notebook demonstrates how to deploy in Watson Machine Learning service an AutoAI model created in Generated Scikit-learn Notebook
which is composed during autoai experiments (in order to learn more about AutoAI experiments go to experiments/autoai).

Some familiarity with bash is helpful. This notebook uses Python.

Learning goals

The learning goals of this notebook are:

• Working with the Watson Machine Learning instance
• Online deployment of AutoAI model
• Scoring data using deployed model

Contents

This notebook contains the following parts:

1. Setup
2. Model upload
3. Web service creation
4. Scoring
5. Clean up
6. Summary and next steps

1. Set up the environment

Before you use the sample code in this notebook, you must perform the following setup tasks:

• Create a Watson Machine Learning (WML) Service instance (a free plan is offered and information about how to create the instance can be found here).

Connection to WML

Authenticate the Watson Machine Learning service on IBM Cloud. You need to provide platform api_key and instance location.

You can use IBM Cloud CLI to retrieve platform API Key and instance location.

API Key can be generated in the following way:

ibmcloud login
ibmcloud iam api-key-create API_KEY_NAME

In result, get the value of api_key from the output.

Location of your WML instance can be retrieved in the following way:

ibmcloud login --apikey API_KEY -a https://cloud.ibm.com
ibmcloud resource service-instance WML_INSTANCE_NAME

In result, get the value of location from the output.

Tip: Your Cloud API key can be generated by going to the Users section of the Cloud console. From that page, click your name, scroll down to the API Keys section, and click Create an IBM Cloud API key. Give your key a name and click Create, then copy the created key and paste it below. You can also get a service specific url by going to the Endpoint URLs section of the Watson Machine Learning docs. You can check your instance location in your Watson Machine Learning (WML) Service instance details.

You can also get service specific apikey by going to the Service IDs section of the Cloud Console. From that page, click Create, then copy the created key and paste it below.

Action: Enter your api_key and location in the following cell.

api_key = 'PASTE YOUR PLATFORM API KEY HERE'
location = 'PASTE YOUR INSTANCE LOCATION HERE'

wml_credentials = {
    "apikey": api_key,
    "url": 'https://' + location + '.ml.cloud.ibm.com'
}

Install and import the ibm-watson-machine-learning package

Note: ibm-watson-machine-learning documentation can be found here.

!pip install -U ibm-watson-machine-learning

from ibm_watson_machine_learning import APIClient

client = APIClient(wml_credentials)

Working with spaces

First, create a space that will be used for your work. If you do not have space already created, you can use Deployment Spaces Dashboard to create one.

• Click New Deployment Space
• Create an empty space
• Select Cloud Object Storage
• Select Watson Machine Learning instance and press Create
• Copy space_id and paste it below

Tip: You can also use SDK to prepare the space for your work. More information can be found here.

Action: Assign space ID below

space_id = 'PASTE YOUR SPACE ID HERE'

You can use list method to print all existing spaces.

client.spaces.list(limit=10)

To be able to interact with all resources available in Watson Machine Learning, you need to set space which you will be using.

client.set.default_space(space_id)

2. Upload model

In this section you will learn how to upload the model to the Cloud.

Download the data as an pandas DataFrame and AutoAI saved as scikit pipeline model using wget.

Hint: To install required packages, execute command !pip install pandas wget numpy.

We can extract model from the executed AutoAI experiment using ibm-watson-machine-learning with following command: experiment.optimizer(...).get_pipeline(astype='sklearn').

!pip install pandas wget numpy

import os, wget
import pandas as pd
import numpy as np

filename = 'german_credit_data_biased_training.csv'
url = 'https://raw.githubusercontent.com/IBM/watson-machine-learning-samples/master/cloud/data/credit_risk/german_credit_data_biased_training.csv'
if not os.path.isfile(filename):
    wget.download(url)

model_name = "model.pickle"
url = 'https://raw.githubusercontent.com/IBM/watson-machine-learning-samples/master/cloud/models/autoai/credit-risk/model.pickle'
if not os.path.isfile(model_name):
    wget.download(url)

credit_risk_df = pd.read_csv(filename)
X = credit_risk_df.drop(['Risk'], axis=1)
y = credit_risk_df['Risk']

credit_risk_df.head()

Custom software_specification

Create new software specification based on default Python environment extended by autoai-libs package.

base_sw_spec_uid = client.software_specifications.get_uid_by_name("runtime-23.1-py3.10")

url = 'https://raw.githubusercontent.com/IBM/watson-machine-learning-samples/master/cloud/configs/config.yaml'
if not os.path.isfile('config.yaml'):
    wget.download(url)

!cat config.yaml

config.yaml file describes details of package extention. Now you need to store new package extention with APIClient.

meta_prop_pkg_extn = {
    client.package_extensions.ConfigurationMetaNames.NAME: "scikt with autoai-libs",
    client.package_extensions.ConfigurationMetaNames.DESCRIPTION: "Extension for autoai-libs",
    client.package_extensions.ConfigurationMetaNames.TYPE: "conda_yml"
}

pkg_extn_details = client.package_extensions.store(meta_props=meta_prop_pkg_extn, file_path="config.yaml")
pkg_extn_uid = client.package_extensions.get_uid(pkg_extn_details)
pkg_extn_url = client.package_extensions.get_href(pkg_extn_details)

Create new software specification and add created package extention to it.

meta_prop_sw_spec = {
    client.software_specifications.ConfigurationMetaNames.NAME: "Mitigated AutoAI bases on scikit spec",
    client.software_specifications.ConfigurationMetaNames.DESCRIPTION: "Software specification for scikt with autoai-libs",
    client.software_specifications.ConfigurationMetaNames.BASE_SOFTWARE_SPECIFICATION: {"guid": base_sw_spec_uid}
}

sw_spec_details = client.software_specifications.store(meta_props=meta_prop_sw_spec)
sw_spec_uid = client.software_specifications.get_uid(sw_spec_details)

client.software_specifications.add_package_extension(sw_spec_uid, pkg_extn_uid)

Get the details of created software specification

client.software_specifications.get_details(sw_spec_uid)

Load the AutoAI model saved as scikit-learn pipeline.

Depending on estimator type in autoai model pipeline may consist models from following frameworks:

• xgboost
• lightgbm
• scikit-learn

from joblib import load

pipeline = load(model_name)

Store the model

model_props = {
    client.repository.ModelMetaNames.NAME: "AutoAI model",
    client.repository.ModelMetaNames.TYPE: 'scikit-learn_1.1',
    client.repository.ModelMetaNames.SOFTWARE_SPEC_UID: sw_spec_uid
    
}
feature_vector = X.columns

published_model = client.repository.store_model(
    model=pipeline, 
    meta_props=model_props,
    training_data=X.values,
    training_target=y.values,
    feature_names=feature_vector,
    label_column_names=['Risk']
)

published_model_uid = client.repository.get_model_id(published_model)

Get model details

client.repository.get_details(published_model_uid)

Note: You can see that model is successfully stored in Watson Machine Learning Service.

client.repository.list_models()

3. Create online deployment

You can use commands bellow to create online deployment for stored model (web service).

metadata = {
    client.deployments.ConfigurationMetaNames.NAME: "Deployment of AutoAI model.",
    client.deployments.ConfigurationMetaNames.ONLINE: {}
}

created_deployment = client.deployments.create(published_model_uid, meta_props=metadata)

Get deployment id.

deployment_id = client.deployments.get_uid(created_deployment)
print(deployment_id)

4. Scoring

You can send new scoring records to web-service deployment using score method.

values = X.values

scoring_payload = {
    "input_data": [{
        'values': values[:5]
    }]
}

predictions = client.deployments.score(deployment_id, scoring_payload)
predictions

5. Clean up

If you want to clean up all created assets:

• experiments
• trainings
• pipelines
• model definitions
• models
• functions
• deployments

see the steps in this sample notebook.

6. Summary and next steps

You successfully completed this notebook! You learned how to use Watson Machine Learning for AutoA model deployment and scoring. Check out our Online Documentation for more samples, tutorials, documentation, how-tos, and blog posts.

Author

Jan Sołtysik Intern in Watson Machine Learning.

Copyright © 2020, 2021 IBM. This notebook and its source code are released under the terms of the MIT License.