Five Steps To Create An Experiment In Azure Machine Learning Studio

Entering Machine Learning Studio

Firstly, open Azure Machine Learning Studio using the following Web address, then click the Sign In button and sign in with your Microsoft account.

Five steps to create an experiment

The following are the five basic steps to build an experiment in Machine Learning Studio in order to create, train, and score your model:

Create a model

• Step 1: Get data
• Step 2: Preprocess data
• Step 3: Define features

Train the model

• Step 4: Choose and apply a learning algorithm

Score and test the model

• Step 5: Predict new automobile prices

Step 1 - Get data

For creating a new experiment by clicking + NEW at the bottom of the Machine Learning Studio window, select EXPERIMENT, and then select Blank Experiment.

 

Then drag the Sample dataset from Saved DataSets, Sample, then Automobile Price data (Raw) to the experiment canvas,

 

Step 2 - Preprocess data

A dataset normally requires some pre-processing (Cleaning) before it can be analyzed. Drag the Clean Missing Data module to the experiment canvas and connect it to the Automobile Price data (Raw) module. In the Properties pane, select Remove entire row in Cleaning mode to clean the data by removing rows that have missing values.

Next Run the experiment by clicking RUN under the experiment canvas.

 

 

Step 3 - Define features

Now Add the project Columns module to the experiment canvas and click Launch column selector.

Now we get the Select columns window, In the column selector, select no columns to begin with, and then select Include and column names in the filter row. Select our list of column names.

  

 

Step 4 - Choose and apply a learning algorithm

Next, add the Spilt Data module to the experiment canvas and connect it to the output of the last Project Columns module and set Fraction of rows in the first output dataset to 0.70.

 

Run the experiment. Then to select the learning algorithm, select the Linear Regression module under the Regression category and drag it to the experiment canvas.

 

Search and drag the Train Model module to the experiment canvas from the Train Category. Connect the left input port to the output of the Linear Regression module. Connect the right input port to the training data output (left port) of the Split Data module.



Select the Train Model module, click Launch column selector in the Properties option, and then select the price column. This is the value that our model is going to predict then click Run the experiment.

 

Step 5 - Predict new automobile prices

Next, Drag the Score Model module to the experiment canvas and connect the left input port to the output of the Train Model module and then connect the right input port to the test data output (right port) of the Split Data module.



To run the experiment and view the output from the Score Model module, click the output port, and then select Visualize.



The output shows the predicted values for price and the known values from the test data.

Finally, to test the quality of the results, select and drag the Evaluate Model module to the experiment canvas, and connect the left input port to the output of the Score Model module.

 

Run the experiment. To view the output from the Evaluate Model module, click the output port, and then select Visualize.