Merge pull request #267384 from likebupt/update-designer-docs-20240227

update designer component articles to describe metrics more accurately

Author: prmerger-automator[bot]

articles/machine-learning/component-reference/evaluate-model.md

@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 07/27/2020
+ms.date: 02/27/2024
 ---
 # Evaluate Model component
 
@@ -97,7 +97,7 @@ The following metrics are reported when evaluating binary classification models.
 
 The metrics returned for regression models are designed to estimate the amount of error.  A model is considered to fit the data well if the difference between observed and predicted values is small. However, looking at the pattern of the residuals (the difference between any one predicted point and its corresponding actual value) can tell you a lot about potential bias in the model.  
 
- The following metrics are reported for evaluating regression models.
+ The following metrics are reported for evaluating linear regression models. Other re gression models such as [Fast Forest Quantile Regression](./fast-forest-quantile-regression.md) may have different metrics.
 
 - **Mean absolute error (MAE)** measures how close the predictions are to the actual outcomes; thus, a lower score is better.  
 

articles/machine-learning/component-reference/fast-forest-quantile-regression.md

@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 07/13/2020
+ms.date: 02/27/2024
 ---
 # Fast Forest Quantile Regression
 
@@ -88,6 +88,13 @@ After training is complete:
 
 + To save a snapshot of the trained model, select the training component, then switch to **Outputs+logs** tab in the right panel. Click on the icon **Register dataset**.  You can find the saved model as a component in the component tree.
 
+## Evaluation metrics
+
+You can use [Evaluate Model component](./evaluate-model.md) to evaluate the trained model. For **Fast Forest Quantile Regression**, the metrics are as following.
+
+- **Quantile Loss**: This is a measure of the error for a specific quantile in your model.
+- **Average Quantile Loss**: This is simply the average of the Quantile Loss values across all the quantiles considered in the model. It gives an overall measure of how well the model is performing across all quantiles.
+
 ## Next steps
 
 See the [set of components available](component-reference.md) to Azure Machine Learning.