> ## Documentation Index
> Fetch the complete documentation index at: https://docs.exploratory.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Evaluate Prediction - Binary

Returns a data frame with evaluation score of binary classification including the below.

* AUC
* f\_score
* accuracy
* misclassification\_rate
* precision
* recall
* specificity
* true\_positive - Number of positive predictions that actually are positive.
* false\_positive - Number of positive predictions that actually are negative.
* true\_negative - Number of negative predictions that actually are negative.
* false\_negative - Number of negative predictions that actually are positive.
* test\_size - The number of tested data.
* threshold - threshold value for prediction.

## How to Access This Feature

### From + (plus) Button

There are two ways to access. One is to access from 'Add' (Plus) button.

![](https://mintcdn.com/exploratory-inc/WFw15qmp062pMw1-/en/ml/images/evaluate_binary_add.png?fit=max\&auto=format\&n=WFw15qmp062pMw1-\&q=85\&s=860872a2d5435f6281ed928263f363ac)

Another way is to access from a column header menu.

![](https://mintcdn.com/exploratory-inc/WFw15qmp062pMw1-/en/ml/images/evaluate_binary_col.png?fit=max\&auto=format\&n=WFw15qmp062pMw1-\&q=85\&s=428c82e088bfb59d553e44cffd2bd835)

## How to Use?

![](https://mintcdn.com/exploratory-inc/WFw15qmp062pMw1-/en/ml/images/evaluate_binary_param.png?fit=max\&auto=format\&n=WFw15qmp062pMw1-\&q=85\&s=8a29895d86cae44d92a106d478914313)

* Predicted Probability Column - The column with predicted values. Usually, it's predicted\_probability in the framework of Exploratory.
* Actual Value Column - The column with actual value.
* Threshold Value to Decide Predicted Label - You can choose how to decide threshold for predicted label.
  * Use Optimized Value - This searches threshold to optimize the chosen metric. It can be
    * F Score
    * Accuracy
    * Precision
    * Recall
    * Specificity
  * Enter Manually
    * Set threshold value manually.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.