Hi, all! I'm planning to write a how-to doc for ndarray indexing as per discussion in #19407. I would love to know which use-cases you'd like to see in the doc. These are a few I have in mind (with inspiration from StackOverflow):

• Accessing specific/random rows and columns
• Indexing along a specific axis
• Creating a sub-ndarray from a larger matrix
• Using variables for slicing e.g. having start and stop relative to each other
• Indexing based on conditional arguments
• Generating index arrays based on specific conditions
• Fetching indices of N max/min values
• Getting the number of indices generated for an indexing expression without performing the operation
• Indexing with index arrays of shapes different from the ndarray
• Indexing the same ndarray multiple times efficiently
• One-hot encoding with indices

As a new user of NumPy, I'm not sure if some of these cases are trivial or might be out of scope for a doc focus on indexing; I would appreciate any opinions regarding this.

cc: @melissawm @rossbar