Evaluate¶
Evaluate Node icon
The Clario Evaluate node is used to generate model documentation, including a model description, a model profile and related graphs. Evaluate will work with either the Logistic Regression or Linear regression nodes. There are two ways to use Evaluate. These are outlined below as Method A and Method B.
Method A
Two input data streams are connected to the evaluate node: one dataset created as a result of a Logistic or Linear node (this dataset contains the model information only), and one dataset containing the actual modeling data (dependent and predictor attributes). Connect the Linear or Logistic node to the top connector. Connect the data stream to the bottom connector.
Method B
Only one input data stream is connected to the evaluate node: a dataset that already contains a model score, which you wish to use to generate the model description, profile and graphs. Connect the data stream to the bottom connector.
Configuration¶
The Evaluate node has two configuration tabs: Configure, and Select Attributes.
Configure Tab¶
Configure Tab, Method A
Method A - two inputs
The Configure Tab contains an Available Attributes list box, a Weight Attribute field, and a Settings area.
First, drag and drop an attribute from the Available Attributes list box into the Weight Attribute Field. The Weight attribute MUST be an integer. If you do not have a weight attribute, skip this step. Next, under settings, select the number of Segments you want for the model documentation, the profile and the graphs. The valid range is 1-20 segments, and the default value is 10. Finally, also under Settings, type in the attribute name you want for the Model Score Name, see tips on Valid Characters for Attribute Names; if invalid keys are pressed in this field, nothing will appear. The Model Score will be calculated based on your top connector model equation (linear or logistic).
- The model stream node must be connected to the top connector.
- ‘Model Score Name’ cannot be <NULL>.
Configure Tab, Method B
Method B - one input
The Configure Tab contains an Available Attributes list box, a Model Attribute field, a Weight Attribute field, and a Settings area.
First, drag and drop the attribute you will to use to evaluate the model performance from the Available Attributes list box into the Model Attribute Field. Next, drag and drop an attribute from the Available Attributes list box into the Weight Attribute Field. The Weight attribute MUST be an integer. If you do not have a weight attribute, skip this step. Finally, under Settings, select the number of segments you want to display in the model documentation, profile and graphs. The valid range is 1-20 segments, and the default value is 10.
Select Attributes Tab¶
The Select Attributes Tab contains Available Attributes and Selected Attributes list boxes. Drag and drop the predictor attributes that are in the model (as well as any other attributes you want to analyze) from the Available Attributes box to the Selected Attributes box. See tips on Finding and Selecting Attributes. You must put at least one attribute in the Selected Attributes list box.
Select Attributes Tab
Results¶
When you use Method A (two inputs), the Evaluate node results set contains three results screens: Model Equation, Model Profile, and Graphs. When you use Method B (one input), this set contains two results screens: Model Profile and Graph.
Model Equation¶
Model Equation Tab
This shows the actual model equation using Pseudo Code in the top box. It also shows each attribute in the model, along with the coefficient and score contribution (computed using the Absolute Value of the standardized estimate). Note that the score contribution is a percentage value, and all values add to 100. You can easily export the model description and coefficients to a spreadsheet by clicking on the Export to Spreadsheet button on the Toolbar.
Model Profile¶
Model Profile Tab, Method A
This tab profiles the results of your model, split into the number of segments specified (between 1 and 20), using model score. The model scores are ordered low to high, and an attempt is made to rank all rows in your data into equally-sized segments. The profile then reports the number of rows for each segment, along with the means of the Dependent Attribute Predicted Score (using the attribute name you specified on the Configure Tab) and All the Predictor Attributes. The top table, consisting of one row, shows values for all records or your dataset. You can use this model profile to evaluate how well your model performs, by comparing the dependent attribute and predicted scores by segment. You can also use the predictor attribute means by segment to describe the model. You can easily export the model profile to a spreadsheet by clicking on the Export to Spreadsheet button on the Toolbar.
Model Profile Tab, Method B
Graph¶
Graph Tab
This results tab is actually a graphing tool, which you can use to graph any of the data in the Model Profile. The x-axis represents model segment, and you can graph one or two attributes on the y-axis. Just select the attribute you want to graph in blue in the drop down box next to the blue square. Then, if you want to graph a second attribute, select the attribute you want to graph in green in the drop down box next to the green square. The blue y-axis values are on the left side of the graph, and the green y-axis values are on the right side of the graph. One common graph will be to compare the Dependent Attribute and Predicted Score by segment.
Output Stream¶
There is no data file output from Evaluate, as it is a terminal node. However, two of the Evaluate results tables, Model Equation and Model Profile, can be exported into Excel by clicking on the Export to Spreadsheet button on the Toolbar.
