1z0-1096-23 Oracle Machine Learning using Autonomous Database 2023 Associate Questions and Answers

• The task that is NOT required by an administrator when adding a new user to Oracle Machine Learning Notebooks is providing the user with an Autonomous Data Warehouse client wallet for remote authentication3.
• The client wallet is only needed for remote access to the database using tools such as SQL Developer or Python. For accessing Oracle Machine Learning Notebooks, the user only needs an OML username and password, which are created by the administrator in the Oracle Machine Learning User Management interface3.

Explanation

https://docs.oracle.com/en/cloud/paas/autonomous-database/adbsa/autonomous-sql-commands.html

Explanation

https://www.doag.org/formes/pubfiles/13151859/OE-DAC-Oracle-Machine-Learning-Overview-Whats-New-Customers-Apps_MLconcepts_Berger.pdf

Explanation

https://docs.oracle.com/en/database/oracle/machine-learning/oml-notebooks/omlug/use-scratchpad.html

When managing machine learning models using the OML AutoML UI, three actions that can be performed on Oracle Machine Learning (OML) models from within OML AutoML UI are:

• Create a notebook with auto-generated OML4Py code from an OML in-database model to allow for further model tweaking and batch scoring. This feature enables users to export the selected model as a notebook that contains the OML4Py code to reproduce the model building process and perform additional tasks such as model evaluation, scoring, or deployment1.
• Deploy an OML in-database model to a REST endpoint. This feature enables users to deploy the selected model as a RESTful web service that can be accessed by external applications or tools for real-time scoring or predictions2.
• Review the model statistics associated with the experiment of an OML in-database model. This feature enables users to view the details of the model such as the algorithm name, hyperparameters, performance metrics, feature importance, and confusion matrix3.

Explanation

https://oracle-base.com/articles/misc/sqlcl-format-query-results-with-the-set-sqlformat-command#loader

Explanation

https://learn.oracle.com/ols/course/using-oracle-machine-learning-with-autonomous-database/35644/98086/149312 This is from Oracle Course : Using Oracle Machine Learning with Autonomous Database Regression is a supervised learning technique used when the output attribute is a continuous numerical value, for example, the price of a property based on the attributes of the property and its locality, the humidity of an area, and so on. Here is a list of all the algorithms offered by Oracle Machine Learning. Unsupervised learning does not involve direct control from the developer. In terms of defining an output attribute, the output attribute is a dependent attribute. And as there is no output attribute in unsupervised learning, there is no distinction between dependent and independent attributes.

Explanation

https://oralytics.com/2021/03/15/oml4py-automl-an-example/

Explanation

https://docs.oracle.com/cd/F32587_01/dmcon/oracle-machine-learning-sql-concepts.pdf

• The three Oracle Machine Learning (OML) components that are available with Oracle Autonomous Database are OML Notebooks, OML4SQL, and OML Services2.
• OML Notebooks are web-based interfaces for data analysis, data discovery, data visualization, and collaboration based on Apache Zeppelin notebook technology. They support SQL, PL/SQL, Python, R, and markdown interpreters on Oracle Autonomous Database2.
• OML4SQL is a set of in-database machine learning algorithms and SQL functions that enable data exploration, data preparation, model building, evaluation, and deployment using SQL and PL/SQL on Oracle Autonomous Database2.
• OML Services are RESTful web services that enable users to build, deploy, and manage machine learning models using REST APIs on Oracle Autonomous Database. They support native in-database models and ONNX-format models for classification, regression, and clustering2

Explanation

https://docs.oracle.com/en/database/oracle/machine-learning/oml-automl-ui/amlui/create-experiment.html

Explanation

https://docs.oracle.com/en/database/oracle/machine-learning/oml-notebooks/omlug/administer-oracle-machine-learning.html

• View notebooks. An Administrator can view notebooks in their own workspace or in workspaces where they have collaboration rights. However, an Administrator cannot run or modify notebooks1.
• Reassign workspaces to users. An Administrator can reassign workspaces from one user to another user in the User Data page. This can be useful when a user leaves the organization or changes roles2.
• Create, edit and delete OML users. An Administrator can create new OML user accounts and passwords, edit existing OML user information, and delete OML users in the User Management interface.

Explanation

https://docs.oracle.com/en/database/oracle/machine-learning/oml-notebooks/omlug/administer-oracle-machine-learning.html#GUID-E74F0E2E-EEE5-4421-A0BB-96A58811C04A Create User: An administrator creates a new user account and user credentials for Oracle Machine Learning in the User Management interface. Note: You must have the administrator role to access the Oracle Machine Learning User Management interface. To create a user account: Select an Autonomous Da-ta Warehouse instance and on the details page click Service Console. On the Service Console click Administration. Click Manage OML Users to open the Oracle Machine Learning User Administration page. Click Create on the Oracle Machine Learning User Administration page. In the Username field, enter a username for the account. Using the username, the user will log in to an Oracle Ma-chine Learning instance. Enter a name in the First Name field. Enter a name in the Last Name field. In the Email Address field, enter the email ID of the user. Select the option Generate password and email account details to user. User will be required to reset the password on first sign in. to auto generate a temporary password and send an email with the account credentials to the user. If you select this option, you need not enter values in the Password and Confirm Password fields; the fields are grayed out. In the Password field, enter a password for the user, if you choose to create a password for the user.This option is disabled if you select the Generate password... option to auto generate a temporary password for the user. In the Confirm Password field, enter a password to confirm the value that you entered in the Password field.By doing so, you create the password for the user. The user can change the password when first logging in. Click Create. This creates a new database user and grants the required privileges to use Oracle Machine Learning.

• To segment your customer data for marketing research purposes and identify homogeneous groups to build supervised models, you should use clustering12.
• Clustering is a type of unsupervised machine learning that groups data points based on their similarities in terms of features or attributes. Clustering can help discover the underlying structure of the data and reveal the natural segments or categories within it12.
• Clustering can be useful for marketing research because it can help identify different types of customers based on their demographics, preferences, behaviors, or needs. Clustering can also help create customer profiles or personas that can be used to target specific segments with tailored marketing campaigns or offers12.
• Clustering can also be used as a preliminary step for building supervised models, such as classification or regression. By using the cluster labels as an additional feature or a target variable, supervised models can learn from the cluster information and improve their accuracy or performance12.

• To share a notebook with a colleague to run on a different Autonomous Database instance, you should export the notebook in .json format and have your colleague import this notebook4.
• Exporting a notebook in .json format preserves the content and format of the notebook, including the code cells, text cells, images, charts, and graphs. You can export a notebook from the Notebook menu or from the Notebooks page4.
• Importing a notebook in .json format allows you to upload an existing notebook file to your workspace and run it on your Autonomous Database instance. You can import a notebook from the Notebooks page or from the Create Notebook dialog box4.

Explanation

https://apexapps.oracle.com/pls/apex/dbpm/r/livelabs/view-workshop?wid=891 &clear=180&session=109742845155042

Explanation

The three data sources that can be immediately used for model building in Oracle Machine Learning are: Oracle Database tables: Oracle Machine Learning can directly access and use data from tables within an Oracle Database for model building. This allows for efficient utilization of the database's structured data for machine learning tasks. Oracle Database views: Similar to tables, Oracle Machine Learning can also utilize data from database views. Views provide a virtual representation of the data, allowing users to access and analyze specific subsets or transformations of the underlying tables. CSV files: Oracle Machine Learning supports the use of CSV (Comma-Separated Values) files as a data source for model building. CSV files provide a simple and widely-used format for storing tabular data, making it easy to import and use within Oracle Machine Learning. Therefore, the correct options are: Oracle Database tables Oracle Database views CSV files

• The reason why a user with Developer permission is unable to create a job on an existing notebook that is shared is that the user requires the Create Job role1.
• The Create Job role is a database role that grants the privilege to create and manage jobs on Oracle Machine Learning Notebooks. This role is not granted by default to any user, including the ADMIN user. An administrator needs to explicitly grant this role to users who need to create jobs1