Update Planets dataset

[rlpitts]

I've been using a more recent and more detailed version of the 'planets' dataset for my teaching material, and I have to import it often enough that I think it would be useful as a replacement for the current planets.csv. It's derived from the same sources as the existing material, but since the data are from early 2023, there are more than 5 times as many planets and almost twice as many parameter columns.

I pared down the EU exoplanet dataset released on Kaggle (removed stellar parameters that most people wouldn't find relevant), sanitized the data (removed a few retracted planets and replaced some whitespaces-as-missing-values), and uploaded it to my personal GitHub page, here: https://github.com/rlpitts/rlpitts/blob/main/planets.csv

I hope you're interested. It's a very nice data set for demonstrating Seaborn, Pandas functions, and more general Matplotlib functions.

[mwaskom]

Wow, I thought we'd already found all the planets! ;)

Yeah I'd be interested in updating the built in planets dataset, in particular the new planet_type variable in this one is nice because the existing one was a little weak in terms of categorical variables.

[rlpitts]

Glad you like it!

FWIW, the "planet type" variable is just a proxy for certain mass or radius bins. Terrestrial planets have a mass and/or radius <= Earth's, Super-Earths go up to either M10 M_E or R2 R_E, and I think the boundary between Neptune-like and Gas Giants is around R5 R_E or M30 M_E. There were definitely a few misclassified planets to start with, but I think I fixed them.

But the detection method is fairly interesting to look at if you want to see how different methods are biased toward different orbital or physical parameters.