Rocketsled: add MissionControl API to return the x and y values that were tried

Thanks again for making Rocketsled publicly available :slight_smile:

When using it, it seemed to me that it would be useful for the user, in some cases, to have a look at all the x values that were tried during the optimization (and at the corresponding y values).

It is certainly possible to obtain this info from the database (by extracting the spec info of each workflow), but this might be cumbersome.

Would it make sense to add a method to the MissionControl class, that returns the x and y values as numpy arrays (of shape (n_launches, n_x), (n_launches, n_y) for instance)?

+1 for this suggestion (assuming an easy way to get this doesn’t already exist)

···

Best,
Anubhav

This is now a feature!

Use MissionControl’s “fetch_matrices” method to get the X and y matrices now. Let me know if you run into any trouble with it.

Thanks,

Alex

···

On Thursday, February 21, 2019 at 8:55:38 AM UTC-8, RĂ©mi Lehe wrote:

Thanks again for making Rocketsled publicly available :slight_smile:

When using it, it seemed to me that it would be useful for the user, in some cases, to have a look at all the x values that were tried during the optimization (and at the corresponding y values).

It is certainly possible to obtain this info from the database (by extracting the spec info of each workflow), but this might be cumbersome.

Would it make sense to add a method to the MissionControl class, that returns the x and y values as numpy arrays (of shape (n_launches, n_x), (n_launches, n_y) for instance)?

Thanks for adding this! This is exactly what I was looking for.

···

On Thursday, February 21, 2019 at 5:43:15 PM UTC-8, [email protected] wrote:

This is now a feature!

Use MissionControl’s “fetch_matrices” method to get the X and y matrices now. Let me know if you run into any trouble with it.

Thanks,

Alex

On Thursday, February 21, 2019 at 8:55:38 AM UTC-8, RĂ©mi Lehe wrote:

Thanks again for making Rocketsled publicly available :slight_smile:

When using it, it seemed to me that it would be useful for the user, in some cases, to have a look at all the x values that were tried during the optimization (and at the corresponding y values).

It is certainly possible to obtain this info from the database (by extracting the spec info of each workflow), but this might be cumbersome.

Would it make sense to add a method to the MissionControl class, that returns the x and y values as numpy arrays (of shape (n_launches, n_x), (n_launches, n_y) for instance)?