Santiago Soler

During yesterday's meeting we decided that would be better to make `dobs` a required argument that only takes an array (without nans or infs). To make that happen, we would...

Hi @swpucwf. Thanks for asking this question. No, currently SimPEG doesn't have 2.5D FDEM simulation, only 1D and 3D. Maybe @dccowan might have more insight on how to run FDEM...

Thanks for opening this @prisae. I agree with the idea of detaching the inversion framework in SimPEG from the rest of the machinery in the package, so it could easily...

> I still often use the `Data` container myself. It is very useful (but not well documented with examples) to index the data container using the survey's sources and receivers...

Hi @nwilliams-kobold. Thanks for your inputs on this PR. We do plan to include a constructor function for ellipsoids that would work like you describe it. @KellySavanna also started adding...

@KellySavanna, I started pushing a few changes to your PR. Remember to pull them if you want to continue working on them. I'll probably keep working more on it this...

@KellySavanna, congrats for this! Amazing to see it merged! 🚀 🏅 I'll keep working on the ellipsoids code in separate PRs. I have some ideas for it that I started...

I totally agree that the data misfits should return real values when called. Nonetheless, I think the responsibility of ensuring that should come from the simulations: the `residual` method of...

I'm not very familiar with how the simulations and the data misfits were designed to behave with complex data. From [this example](https://docs.simpeg.xyz/latest/content/tutorials/07-fdem/plot_fwd_3_fem_3d.html) I see that the returned data is complex...

Thanks @prisae for the clarification. I think I wasn't understanding the issue very well. I agree that doing `np.vdot(residual, residual).real` won't loose any information, since the imaginary part of the...