Notebooks are the main user interface of the SKA SDP parametric model. To allow exploration of the design space, the parametric model provides a number of separate notebooks for different purposes.
If you wish to use the notebooks interactively, clone the parametric model repository and run the notebooks yourself:
$ git clone https://github.com/ska-telescope/sdp-par-model $ cd sdp-par-model $ pipenv install $ pipenv shell $ jupyter notebook notebooks
This should allow you to interact with the notebooks using a web browser.
List of Notebooks¶
The following notebooks come with the parametric model. Follow the links for non-interactive version of the notebooks:
SKA1_Imaging_Performance_Model: Contains interactive cells for computing telescope parameters (data rates, FLOPs, image sizes etc).
SKA1_Sensitivity_Analysis: Investigates the sensitivity of telescope parameters on inputs such as baseline length and ionospheric time scale.
SKA1_SDP_Performance_Dashboard: Overview of FLOP rate predictions for different telescopes, pipelines and high-priority science objectives (HPSOs).
SKA1_Export: Allows exporting detailed parametric model output for usage with external programs. This notebook can also be used to compare existing parametric model exports to allow fine-grained change tracking of calculated parameters.
SKA1_System_Sizing: Uses the HPSOs to estimate the average computational requirements and output data rates of the SDP instances at the two telescopes. This informs the system sizing.
Absolute_Baseline_length_distribution: Compute and visualize baseline distributions of SKA telescopes. This notebook is used for determining the baseline bins that go into the parametric model calculations.