Skip to main content

Tables Quickstart

The following Quickstart demonstrates how to log data tables, visualize data, and query data.

Select the button below to try a PyTorch Quickstart example project on MNIST data. Open In Colab

1. Log a table​

Follow the procedure outlined below to log a Table with W&B:

  1. Initialize a W&B Run with wandb.init().
  2. Create a wandb.Table() object instance. Pass the name of the columns in your table along with the data for the columns and data parameters, respectively.
  3. Log the table with run.log() as a key-value pair. Provide a name for your table for the key, and pass the object instance of wandb.Table as the value.
run = wandb.init(project="table-test")
my_table = wandb.Table(columns=["a", "b"], data=[["a1", "b1"], ["a2", "b2"]])
run.log({"Table Name": my_table})

You can optionally pass in a Pandas DataFrame to wandb.Table() Class. For more information on supported data types, see the wandb.Table in the W&B API Reference Guide.

2. Visualize tables in the workspace​

View the resulting table in your workspace. Navigate to the W&B App and select the name of your Run in your Project workspace. A new panel is added for each unique table key.

In this example, my_table, is logged under the key "Table Name".

3. Compare across model versions​

Log sample tables from multiple W&B Runs and compare results in the project workspace. In this example workspace, we show how to combine rows from multiple different versions in the same table.

Use the table filter, sort, and grouping features to explore and evaluate model results.

Was this page helpful?👍👎