Data Visualizations

To get an idea of how the data is distributed, we displayed the data and its hyperparameters using different visualization plots:

Here is an example of the first ten lines of the Task Table

Task Table

Each task has a “Number_of_Jobs”. Although the number of jobs is used as a parameter in the learning process, the tasks that are being deplayed all consists of Jobs. These Jobs are represented in the table below


Job Table



Distribution of the main learning parameters are listed below:

ArgumentGraph



NumberOfJobs



Period



Priority