As an Open Analytics Framework user, you can use the framework for use cases such as scoring with existing models, clustering analysis, simulation, data manipulation and analysis, natural language processing, and gradient boosting.
In this section, several step-by-step use cases help you get started with working in the Open Analytics Framework, and illustrates how to use the Open Analytics Framework APIs.
Teradata Python client library, teradataml, is used to manage both Python and R language environments in Open Analytics Framework.
See Working with Open Analytics in the teradataml User Guide for detailed description of these Open Analytics Framework APIs and additional code examples in Jupyter notebooks.
The following use cases use the Teradata Package for Python (teradataml) and its Apply class:
- Using Python environments:
- A scoring example with a Random Forest classifier model: Scoring with a Model File
- A K-Means clustering example: K-Means Clustering Classification
- A simulation example: Simulation
- Two brief examples illustrate the apply method of the teradata.DataFrame class as an alternative, seamless way to interact with the APPLY table operator Examples with teradataml.DataFrame.apply Method.
- Using Python conda environments:
- An example to predict the next value in a Time Series Sequence with Multilayer Perceptron (MLP): Predicting Next Value Using Keras
- An example to perform sentiment analysis using Naive Bayes classifier: Sentiment Analysis
- Using R environments:
- A XGBoost example: Gradient Boosting
- A tokenization example: Natural Language Processing
- An example combining mulidimentional arrays using 'abind' package: Combining Multidimensional Arrays