https://github.com/teradata/modelops-demo-models
Teradata ModelOps Demo Models
https://github.com/teradata/modelops-demo-models
Last synced: about 1 year ago
JSON representation
Teradata ModelOps Demo Models
- Host: GitHub
- URL: https://github.com/teradata/modelops-demo-models
- Owner: Teradata
- License: bsd-3-clause
- Created: 2022-06-23T19:18:18.000Z (almost 4 years ago)
- Default Branch: master
- Last Pushed: 2024-11-15T10:22:14.000Z (over 1 year ago)
- Last Synced: 2025-03-29T06:01:36.825Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 2.08 MB
- Stars: 7
- Watchers: 4
- Forks: 47
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# Models
This repository contains the example Diabetes Prediction model code for [Vantage ModelOps](https://docs.teradata.com/r/Teradata-VantageTM-ModelOps-User-Guide/June-2022). We provide a number of demo projects and associated repositories. The goal of these examples is to provide a simple reference implementation for users and not to provide a detailed data science example for each use case.
Please refer to the [Official Documentation](https://docs.teradata.com/r/Teradata-VantageTM-ModelOps-User-Guide/June-2022) for more information.
## Available Models
The example use case we deal with in this demo project is diabetes prediction. More specifically, we use the [PIMA diabetes](http://nrvis.com/data/mldata/pima-indians-diabetes.csv) dataset and build a classification model to predict the likelihood that a given patient has diabetes based on certain medical features.
#### Git
`Git` models are those models where we manage all the code for training evaluation and scoring. We provide examples in both Python and R and we provide notebooks for the Python example.
- [In-database Model Prediction](model_definitions/pima_python_indb_xgboost)
- [Python Diabetes Prediction](model_definitions/python-diabetes)
- [R Diabetes Prediction](model_definitions/r-diabetes)
#### BYOM
`BYOM` is related to [Teradata BYOM](https://docs.teradata.com/r/Teradata-VantageTM-Bring-Your-Own-Model-User-Guide/May-2022/Bring-Your-Own-Model) where we use an open model format such as `PMML` or `ONNX`. In this example, we provide a notebook to produce the `PMML` model artefact along with examples for how to query it dynamically in Teradata.
- [Diabetes Prediction](byom/pima)
#### STO
`STO` allow us to train, evaluate and score micro models (individual models per data partition in Teradata). We provide notebooks for the Python STO example.
- [Diabetes Prediction](model_definitions/STO)