Using the Open Analytics Framework for LLM/DL - Teradata VantageCloud Lake

For this release, use the Teradata Package for Python, teradataml.

1. Connect to your VantageCloud Lake environment using teradataml.
2. Create a user environment. It's a user-configured space where you can run your external language scripts. User environments have user-specified characteristics such as a name, a description, an external language interpreter kind and version, associated language libraries, and user files. See User Environments and APPLY Table Operator.
3. For large language model inference, download LLMs from Hugging Face using a transformers package and compress them into a zip file.
4. Upload the Python language script, LLM/DL models ZIP file, and any other needed files to your target user environment.
5. Execute the APPLY query with the necessary input data, Python script, and output destination defined.