how to use the existing Azure pipeline

Question

Hello, I try to use the existing Azure machine learning pipeline "Image Object Detection Pipeline" to do my own object detection. I have many questions. May I ask where I can find the manual how to use these existing pipeline? I have studied https://learn.microsoft.com/en-us/azure/machine-learning/how-to-manage-inputs-outputs-pipeline?view=azureml-api-2&tabs=cli, Does Azure provide a more specific lectures how to use these existing pipelines?

For example, in its inputs, there is no way to specify the data input . I expand this pipeline. in its training or validaton_data, I can NOT find where I can specify the data asset or datastore for data input. I also noticed the "training_data" or "validation_data" looks purple, what does it mean?

I click "Configure &Submit". It is NOT allowed to input the data