Upload Data to Data Attribute Recommendation

To communicate with your service instance, you need to retrieve an OAuth access token which will grant you access to the Data Attribute Recommendation APIs. This access token is added to all your service instance requests.

Open Postman and make sure that your Data Attribute Recommendation environment is selected. For detailed steps, see Set Up Postman Environment and Collection to call Data Attribute Recommendation APIs.

On the left, expand the Data Attribute Recommendation collection and open the subfolder Setup. In this folder, select the request, Get Authorization.

Click Send to send the request to your service instance.

The response includes your access_token plus its expiration time expires_in. If the token expires, you need to request a new one. There is no need to copy the access token as the collection automatically adds the token to all requests.

Now, you need to create a new dataset schema. A dataset schema describes the structure of datasets.

In these tutorials, you are using a dataset from Best Buy. The original dataset as well as other dataset from Best Buy can be found here. From the original dataset the product information description, manufacturer and price as well as three levels of product categories were picked to illustrate the possibility of the service to deal with such information. In a generic use case you may pick a number and combination of properties yourself.

Expand the subfolder Upload Data and select the request Create new Dataset Schema. click the Body tab to see the dataset schema that you are going to create.

The schema is divided into features and labels. The features are the inputs for the machine learning model whereas the labels are the fields that shall be predicted. Thus, this schema provides the product information as an input and wants to predict the product categories.

To create this dataset schema, click Send to send the request.

You have successfully created a dataset schema.

To see the details of all dataset schemas that you have created, select the request Get Datasets Schemas collection in the folder Upload Data. click Send to send the request.

In the response, you find your newly created dataset schema plus all other schemas that you have created. For every schema, its details are listed including the name, id and features of the schema.

Next, you need to create a dataset using the dataset schema that you have created. The dataset is a table that holds the data that you will upload later.

To create a dataset, select the request Create new Dataset in the folder Upload Data. Click Send to send the request.

In the response, you find the details of your newly created dataset, including the current status of the dataset. The status is NO_DATA as no data file has been uploaded yet.

You have successfully created a dataset.

The final step is to upload data to your dataset.

For that, use the sample data available on GitHub. Download the CSV file that contains the data.

If your browser displays the data instead of downloading it, right-click anywhere and click Save as… to save the file.

Take a moment to look at the dataset. As mentioned in step 2 the dataset contains product information as well as product categories. You might ask why the product categories are in the dataset when we actually want to predict them?

The categories are only necessary for training as the service does not know yet which product information are common for certain categories. Instead, the service will recognize patterns and establish such connections during the training process. This allows the service to make predictions for categories solely based on the product information later on.

In Postman, select the request Upload data in the folder Upload Data. Now select the tab Body where you can add the data. Make sure that the type Binary is selected and click Select File to upload the file that contains the data. Press Send to send the request.

In the response, you see that the status of your dataset has changed to VALIDATING. The service is now validating the data that you have uploaded. When validation is done, the status will change accordingly.

You have successfully uploaded data to your dataset.

To see the current status of your dataset, select the request Get Dataset by id in the folder Upload Data. Then click Send to send the request.

In the response, you can observe the status of your dataset. If the status is SUCCEEDED, your data has been validated successfully. If the status is VALIDATING, your data is still in the validation process. Wait a minute and then send the request again to check whether the status has changed.

Note that you will not be able to Use Data Attribute Recommendation to Train a Machine Learning Model until the dataset is validated and the status changes from VALIDATING to SUCCEEDED.

Your data is now validated and ready to be used to train a machine learning model.