Skip to main content
Version: 0.96

Getting started with Snorkel Flow

This series of getting started guides is designed to help you quickly get up to speed with the core features and capabilities of Snorkel Flow. Whether you’re new to Snorkel Flow or looking to deepen your understanding, these resources guide you through key workflows to make the most of your experience.

Each of these guides helps you quickly get up and running with the platform. Explore each topic and start building your AI solutions today.

TutorialDescription
Document classificationUse Snorkel Flow to label your training data programmatically, train labeling models, and analyze the results. This tutorial guides you through an iterative development process and introduces key workflows for using Labeling Functions (LFs) to speed up labeling for document classification.
Information extractionSnorkel Flow supports the extraction of structured information from semi-structured documents such as PDFs, HTML, and docx files. You can then prepare your data for analysis and model training by leveraging text, layout, and image modalities to write labeling functions (LFs).
GenAI evaluation tutorialSnorkel's generative AI evaluation module enables enterprise AI teams to develop specialized, fine-grained, actionable GenAI evaluations in an iterative, programmatic way. Implement your genAI evaluation by defining a benchmark, creating and analyzing evaluation dashboards, annotating the datasets, and refining the benchmark or moving into data development. This guide is ideal for users interested in refining generative AI models.
SDK quickstartIf you're a developer, the SDK quickstart shows you how to programmatically classify documents. This guide includes creating your dataset, creating an application, writing labeling functions, training a model, and deploying your model.
Walkthrough for annotatorsThis walkthrough explains the process of manually annotating documents in Annotation Studio, which are used to generate high-quality labeled datasets, ensuring consistency and accuracy across your projects. This guide also includes an interactive demo showing how to annotate your data.