Merge pull request #112617 from likebupt/update-trainer-mode-0424

PMEds28 · web-flow · commit 40dcdfb58f59 · 2020-04-24T15:16:43.000+01:00
update trained mode
diff --git a/articles/machine-learning/algorithm-module-reference/boosted-decision-tree-regression.md b/articles/machine-learning/algorithm-module-reference/boosted-decision-tree-regression.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 
 # Boosted Decision Tree Regression module
@@ -74,10 +74,19 @@ The gradient boosting method can also be used for classification problems by red
     By default, the random seed is set to 0, which means the initial seed value is obtained from the system clock.
   
 
-9. Add a training dataset, and one of the training modules:
+9. Train the model:
 
-    - If you set **Create trainer mode** option to **Single Parameter**, use the [Train Model](train-model.md) module.  
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
   
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
     
 
 10. Submit the pipeline.  
diff --git a/articles/machine-learning/algorithm-module-reference/decision-forest-regression.md b/articles/machine-learning/algorithm-module-reference/decision-forest-regression.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 
 # Decision Forest Regression module
@@ -74,9 +74,19 @@ For more information about the theoretical framework for this algorithm and its
      By increasing this value, you increase the threshold for creating new rules. For example, with the default value of 1, even a single case can cause a new rule to be created. If you increase the value to 5, the training data would have to contain at least five cases that meet the same conditions.
 
 
-9. Connect a labeled dataset, select a single label column containing no more than two outcomes, and connect to [Train Model](./train-model.md).
+9. Train the model:
 
-    - If you set **Create trainer mode** option to **Single Parameter**, train the model by using the [Train Model](./train-model.md) module.
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
    
 
diff --git a/articles/machine-learning/algorithm-module-reference/linear-regression.md b/articles/machine-learning/algorithm-module-reference/linear-regression.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Linear Regression module
 This article describes a module in Azure Machine Learning designer (preview).
@@ -122,9 +122,19 @@ After training is complete:
 10. For **Random number seed**, you can optionally type a value to seed the random number generator used by the model. Using a seed value is useful if you want to maintain the same results across different runs of the same pipeline.
 
 
-12. Add a labeled dataset and one of the training modules.
+12. Train the model:
 
-    If you are not using a parameter sweep, use the [Train Model](train-model.md) module.
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
 13. Submit the pipeline.
 
diff --git a/articles/machine-learning/algorithm-module-reference/multiclass-boosted-decision-tree.md b/articles/machine-learning/algorithm-module-reference/multiclass-boosted-decision-tree.md
@@ -52,8 +52,19 @@ You can train this type of model by using the [Train Model](././train-model.md).
 
          The random seed is set by default to 42. Successive runs using different random seeds can have different results.
 
-> [!Note]
-> If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](./train-model.md) module.
+1. Train the model:
+
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
 ## Next steps
 
diff --git a/articles/machine-learning/algorithm-module-reference/multiclass-decision-forest.md b/articles/machine-learning/algorithm-module-reference/multiclass-decision-forest.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Multiclass Decision Forest module
 
@@ -19,7 +19,7 @@ Use this module to create a machine learning model based on the *decision forest
 
 ## More about decision forests
 
-The decision forest algorithm is an ensemble learning method for classification. The algorithm works by building multiple decision trees and then *voting* on the most popular output class. Voting is a form of aggregation, in which each tree in a classification decision forest outputs a non-normalized frequency histogram of labels. The aggregation process sums these histograms and normalizes the result to get the “probabilities” for each label. The trees that have high prediction confidence have a greater weight in the final decision of the ensemble.
+The decision forest algorithm is an ensemble learning method for classification. The algorithm works by building multiple decision trees and then *voting* on the most popular output class. Voting is a form of aggregation, in which each tree in a classification decision forest outputs a non-normalized frequency histogram of labels. The aggregation process sums these histograms and normalizes the result to get the "probabilities" for each label. The trees that have high prediction confidence have a greater weight in the final decision of the ensemble.
 
 Decision trees in general are non-parametric models, meaning they support data with varied distributions. In each tree, a sequence of simple tests is run for each class, increasing the levels of a tree structure until a leaf node (decision) is reached.
 
@@ -66,9 +66,19 @@ The decision forest classifier in Azure Machine Learning consists of an ensemble
 
 
 
-10. Connect a labeled dataset, and one of the training modules:
+10. Connect a labeled dataset, and train the model:
 
-    + If you set **Create trainer mode** to **Single Parameter**, use the [Train Model](./train-model.md) module.
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
 11. Submit the pipeline.
 
diff --git a/articles/machine-learning/algorithm-module-reference/multiclass-logistic-regression.md b/articles/machine-learning/algorithm-module-reference/multiclass-logistic-regression.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Multiclass Logistic Regression module
 
@@ -49,9 +49,19 @@ In multiclass logistic regression, the classifier can be used to predict multipl
 
 6. **Random number seed**: Type an integer value to use as the seed for the algorithm if you want the results to be repeatable over runs. Otherwise, a system clock value is used as the seed, which can produce slightly different results in runs of the same pipeline.
 
-8. Connect a labeled dataset, and one of the train modules:
-
-    + If you set **Create trainer mode** to **Single Parameter**, use the [Train Model](./train-model.md) module.
+8. Connect a labeled dataset, and train the model:
+
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
 9. Submit the pipeline.
 
diff --git a/articles/machine-learning/algorithm-module-reference/multiclass-neural-network.md b/articles/machine-learning/algorithm-module-reference/multiclass-neural-network.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Multiclass Neural Network module
 
@@ -72,9 +72,19 @@ To compute the output of the network for a particular input, a value is calculat
 
 12. **Random number seed**: Type a value to use as the seed, if you want to ensure repeatability across runs of the same pipeline.
 
-14. Connect a training dataset and one of the [training modules](module-reference.md): 
+14. Train the model:
 
-    - If you set **Create trainer mode** to **Single Parameter**, use [Train Model](train-model.md).  
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
+  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.  
   
 
 ## Results
diff --git a/articles/machine-learning/algorithm-module-reference/neural-network-regression.md b/articles/machine-learning/algorithm-module-reference/neural-network-regression.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Neural Network Regression module
 
@@ -76,9 +76,19 @@ Neural networks can be extensively customized. This section describes how to cre
   
 11. For **Random number seed**, you can optionally type a value to use as the seed. Specifying a seed value is useful when you want to ensure repeatability across runs of the same pipeline.
   
-13. Connect a training dataset and one of the [training modules](module-reference.md): 
+13. Connect a training dataset and train the model:
+
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
   
-    -   If you set **Create trainer mode** to **Single Parameter**, use [Train Model](./train-model.md).  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.  
   
    
 14. Submit the pipeline.  
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-averaged-perceptron.md b/articles/machine-learning/algorithm-module-reference/two-class-averaged-perceptron.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Two-Class Averaged Perceptron module
 
@@ -45,9 +45,19 @@ The simpler perceptron models are suited to learning linearly separable patterns
   
 5.  For **Random number seed**, optionally type an integer value to use as the seed. Using a seed is recommended if you want to ensure reproducibility of the pipeline across runs.  
   
-1.  Connect a training dataset, and one of the training modules:
+1.  Connect a training dataset, and train the model:
+
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
   
-    -   If you set **Create trainer mode** to **Single Parameter**, use the [Train Model](train-model.md) module.
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.
 
 
 
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-boosted-decision-tree.md b/articles/machine-learning/algorithm-module-reference/two-class-boosted-decision-tree.md
@@ -9,7 +9,7 @@ ms.topic: reference
 
 author: likebupt
 ms.author: keli19
-ms.date: 02/22/2020
+ms.date: 04/22/2020
 ---
 # Two-Class Boosted Decision Tree module
 
@@ -56,9 +56,19 @@ You can train this type of model using [Train Model](././train-model.md).
      The random seed is set by default to 0, which means the initial seed value is obtained from the system clock.  Successive runs using a random seed can have different results.
   
 
-9. Train the model.
+9. Train the model:
+
+    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](train-model.md) module.  
+  
+    + If you set **Create trainer mode** to **Parameter Range**, connect a tagged dataset and train the model by using [Tune Model Hyperparameters](tune-model-hyperparameters.md).  
   
-    + If you set **Create trainer mode** to **Single Parameter**, connect a tagged dataset and the [Train Model](./train-model.md) module.  
+    > [!NOTE]
+    > 
+    > If you pass a parameter range to [Train Model](train-model.md), it uses only the default value in the single parameter list.  
+    > 
+    > If you pass a single set of parameter values to the [Tune Model Hyperparameters](tune-model-hyperparameters.md) module, when it expects a range of settings for each parameter, it ignores the values, and uses the default values for the learner.  
+    > 
+    > If you select the **Parameter Range** option and enter a single value for any parameter, that single value you specified is used throughout the sweep, even if other parameters change across a range of values.  
    
 ## Results
 
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-decision-forest.md b/articles/machine-learning/algorithm-module-reference/two-class-decision-forest.md
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-logistic-regression.md b/articles/machine-learning/algorithm-module-reference/two-class-logistic-regression.md
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-neural-network.md b/articles/machine-learning/algorithm-module-reference/two-class-neural-network.md
diff --git a/articles/machine-learning/algorithm-module-reference/two-class-support-vector-machine.md b/articles/machine-learning/algorithm-module-reference/two-class-support-vector-machine.md
