Weeks 9 & 10: Slicing, Profiles and a New Factory
Slicing, Profiles and a New Factory 🚀 The last two weeks have been a pretty interesting mix of working on spectrogram slicing and starting to look at one of the bigger architectural changes in radiospectra . A lot of the work this time was about understanding how the different pieces fit together and figuring out how to make things easier for users and contributors. Exploring Slicing and Profiles 🔍 After the ndcube refactor, I started looking more closely at how users could work with individual parts of a spectrogram. One thing I wanted to make easier was getting a time profile at a particular frequency or a frequency profile at a particular time . Initially, I worked on adding time_profile() and line_profile() methods to GenericSpectrogram . To make this work, I had to deal with something I hadn't worked with much before—converting physical coordinates into the actual array indices behind the spectrogram. I added a couple of internal helpers to handle this conversio...