Welcome to ml_toolkit’s documentation!

The ml_toolkit contains many functions to streamline machine learning usecases across different domains. The main modules are:

• llm: Contains a series of functions to use LLMs within Databricks notebooks.

• text_classification: Contains functions to perform supervised training for text classification tasks.

• text_augmentation: Contains functions that use LLMs to perform text augmentation, used to enhance supervised training scenarios.

Attention

The text_classification and text_augmentation modules are bespoke-specific and only available within the corporate and rnd workspaces.

Getting Started

Installing from Databricks

• In Databricks notebooks, the ml_toolkit operates the same way as a repo-based library

• Add this to the top of your notebook as a cell

Install library dependencies

%run /setup_ml_toolkit

• Then, in subsequent cells:

Use the toolkit

import ml_toolkit

Contents:

• llm module
  • Quota Controls
  • Available Models
  • Functions
  • run_llm_batch
  • estimate_token_usage
  • query
  • LLMResponse

Indices and tables

• Index

• Module Index

• Search Page

Contact Us!

This package is maintained by Platform Engineering. If you have any issues, contact us via our Jira Form.