Announcing W&B Tables: Iterate on Your Data
Today, we're excited to launch W&B Tables, a new tool for data iteration and model evaluation. Here's how it works:. Made by Shawn Lewis using W&B
Today, we're excited to launch W&B Tables, a new tool for data iteration and model evaluation. Tables will help you achieve faster model development cycles and a deeper understanding of how your models work.
With Tables, you can log, version, organize and visualize structured data. Tables is designed to work well for a wide range of use cases, from analyzing intermediate data transformations to debugging model predictions.
Our early adopters use Tables to track how their training datasets evolve, and record every prediction made during model evaluation. The powerful new Table Visualizer helps them find difficult slices of data and unearth patterns that guide model development.
Understand your data
Here's an example table visualization from Stacey's Mendeleev project. Check out her report
to see real world examples of how Tables will help you understand your models in the context of your data.
Since this document is a W&B Report, the panel above is interactive. It pulls in tables of model predictions that were created by 4 different model evaluations. In this example, we've filtered to incorrect predictions and grouped by the ground truth label.
We can see that our model frequently confuses Amphibia with Mollusca, by looking at the guess histograms in the second column. Can you build an intuition about this by inspecting the images in those rows?
There is a lot more to unearth here! Try clicking a column header to play with grouping, or create a new derived column as an expression over the data in the other columns, and see what you can find.
So how does Tables work?
Check out our docs
for a deep dive, and instructions on getting started.
For now, here's a brief overview.
A Table is well, a table, with rows and columns. You can add rich media values like images with bounding box controls, and cross-table references for seamlessly connecting one table to another. To store a Table with W&B, construct the wandb.Table class, and then add your Table to a wandb.Artifact, or log it in a run with wandb.log.
Tables comes with a powerful type system that you don't need to learn. It will infer the types of data you give it automatically.
Since Tables is built on top of W&B Artifacts
, you get efficient versioning and data lineage tracking automatically.
Part of the Artifacts DAG in Stacey's Mendeleev project
The Table Visualizer is built using a new system we've developed called Weave, that allows us (and eventually you!) to create flexible, composable and scalable data UIs.
You can build derived columns in the Table Visualizer using Weave's powerful data-aware auto-suggest capabilities. The Table Visualizer keeps track of types and automatically uses appropriate visualizations for each column. And you can filter, group, sort, and compare tables by interacting with the controls.
We're just getting started. Over the coming months, we'll:
Scale Tables up with the Weave backend. We'll remove the current artificial limit of 200k rows per Table, and make everything really fast.
Release a new way to plot your data using Weave. You'll be able to plot data from Tables, runs, artifacts and more. And you'll be able to perform projections like T-SNE to get a new perspective on your data.
And as always, we love any and all feedback. If you've got questions, ideas, bug reports, or anything else you'd like to discuss, please don't hesitate to ping us at firstname.lastname@example.org