New matplotlib notebook issues

[Zabamund]

A few issues in new notebook:

• perhaps worth adding a link to pooch on pypi as I'm 100% sure at least one student will ask what spot = pooch.create means
• ax.set_ylim(830, 880) no longer displays any data as depth.min(), depth.max() are (1140.0, 13950.0) when using https://geocomp.s3.amazonaws.com/data/L-30.las, the data loaded with numpy 'https://geocomp.s3.amazonaws.com/data/L-30_Depth-DT-RHOB.csv' have a different depth range (as previously)
• name 'ma' is not defined presumably we need to run ma = np.percentile(vol, 98)
• reading fname = spot.fetch('F3_horizon.npy') in context manager yields UnicodeDecodeError: 'utf-8' codec can't decode byte 0x93 in position 0: invalid start byte
• in scatter = ax.scatter(np.arange(200), hor[200, :200]/0.004, c=attribute[200, :200]), name 'attribute' is not defined
• in 'Adding decoration' section, the displayed data don't match the selected regions to display

I'm dropping these here so as not to loose track of them.
If I understand how we use kosu correctly, any corrections should be done in a local copy of geocomputing in the develop branch, then after testing, these changes should be merged to master so that the Github Action can then push these changes to S3?

[kwinkunks]

I'm not certain we'll be using this notebook in the next classes, but I remember now I did leave it in a half-baked state when we changed the seismic data.

[Zabamund]

Ah ok, I thought we were as it's listed in geocomp.yaml in the curriculum section, day 4.