ROC Curve

A Receiver Operating Characteristic (ROC) Curve visualization node. The view can be accessed either via the “Open view” action on the executed node or on KNIME Hub.
In the node configuration, you can choose the input data you want to display and enable certain controls, which are then available in the view. This includes the ability to choose different columns as reference and prediction columns or the possibility to set a title.
The configuration also offers a preview of the view, which should help to get the ROC curve in the desired shape quickly.
Since missing values as well as NaN (not a number) or infinite values cannot be displayed in the view, they will be omitted. Values outside the range [0,1] will be clipped to the interval boundaries.

Options

Target column

Defines the column that contains the actual nominal target values.

Positive class value

Defines which of the target classes is the positive class.

Prediction columns

Defines the columns that contain the predicted positive class probabilities.

Limit points per line to

Numerical value that restricts the maximum number of data points used to draw each curve. If there are more rows with valid values in the prediction columns they are aggregated accordingly

Title

The title of the graph shown above the generated image. If left blank, no title will be shown.

Horizontal axis label

The label for the X-Axis. If left blank, no label will be shown.

Vertical axis label

The label for the Y-Axis. If left blank, no label will be shown.

Line thickness

Numeric value that defines the line size.

Legend position

Defines whether the legend is shown in the bottom right corner of the plot, below the plot, or not at all.

Enable image download

When checked, an image of the chart can be downloaded by clicking the image download button in the view.

Enable animation

When checked, the animation of the graph is active.

Enable data zoom

When checked, zooming in the chart is enabled.

Show tooltip

When checked, hovering over a line will show information about the data.

Generate image

If checked, adds an image output port and, upon execution, generates an image which is made available via this port.

Image type (SVG, PNG)

Choose between the superior quality provided by SVG vector graphics and the better performance of the PNG raster images. SVG graphics are sharp, also when zoomed, and text labels within the graphic can be searched for. On the other hand, PNG images can be smaller and render faster when there are many data points, for example, in a scatter plot.

Width

The width of the generated image (in pixels).

Height

The height of the generated image (in pixels).

Views

ROC Curve

An ROC Curve visualization of the input data.

Workflows

• 00 Install extensionsKNIME Hub
• 00 Install extensionsKNIME Hub
• 00 Install extensionsKNIME Hub
• 01_Decision_Tree_solutionKNIME Hub
• 01_Decision_TreesKNIME Hub
• Show all 147 workflows

Links

• No links available

Developers