Tensorboard’s Tensorboard WhatIf
The What-If Tool (WIT) provides an easy-to-use interface for expanding understanding of black-box classification and regression ML models. With the plugin, you can perform inference on a large set of examples and immediately visualize the results in a variety of ways. Additionally, examples can be edited manually or programmatically and re-run through the model in order to see the results of the changes. It contains tooling for investigating model
The Features Overview dashboard of the What-If Tool provides an overview of the distribution of values of each feature in the dataset loaded into the tool. It also provides the same overview for outputs from the model(s) being analyzed by the tool. The dashboard can be found by clicking on the “Features” tab in the What-If Tool.
The dashboard contains two tables: one for numeric features and one for categorical features, along with a control panel. The features are separated into separate tables as the information shown in the tables is different for numeric features versus categorical features. In the tables there is a row for each feature in the dataset. Each row contains some calculated statistics about the values of that feature across the entire dataset, along with a number of charts to show the distribution of values.