MLflow Transformers Flavor - Tutorials and Guides

Below, you will find a number of guides that focus on different use cases using transformers that leverage MLflow's APIs for tracking and inference capabilities.

Introductory Quickstart to using Transformers with MLflow

If this is your first exposure to transformers or use transformers extensively but are new to MLflow, this is a great place to start.

Quickstart: Text Generation with Transformers

Learn how to leverage the transformers integration with MLflow in this introductory quickstart.

Transformers Fine-Tuning Tutorials with MLflow

Fine-tuning a model is a common task in machine learning workflows. These tutorials are designed to showcase how to fine-tune a model using the transformers library with harnessing MLflow's APIs for tracking experiment configurations and results.

Fine tuning a transformers Foundation Model

Learn how to fine-tune a transformers model using MLflow to keep track of the training process and to log a use-case-specific tuned pipeline.

Fine tuning LLMs efficiently using PEFT and MLflow

Learn how to fine-tune a large foundational models with significantly reduced memory usage using PEFT (QLoRA) and MLflow.

Use Case Tutorials for Transformers with MLflow

Interested in learning about how to leverage transformers for tasks other than basic text generation? Want to learn more about the breadth of problems that you can solve with transformers and MLflow?

These more advanced tutorials are designed to showcase different applications of the transformers model architecture and how to leverage MLflow to track and deploy these models.

Audio Transcription with Transformers

Learn how to leverage the Whisper Model with MLflow to generate accurate audio transcriptions.

Translation with Transformers

Learn about the options for saving and loading transformers models in MLflow for customization of your workflows with a fun translation example!

Chat with Transformers

Learn the basics of stateful chat Conversational Pipelines with Transformers and MLflow.

Building and Serving an OpenAI-Compatible Chatbot

Learn how to build an OpenAI-compatible chatbot using a local Transformers model and MLflow, and serve it with minimal configuration.

Prompt templating with Transformers Pipelines

Learn how to set prompt templates on Transformers Pipelines to optimize your LLM's outputs, and simplify the end-user experience.

Custom PyFunc for Transformers

Learn how to define a custom PyFunc using transformers for advanced, state-of-the-art new models.