Highest Priority

• Set a modulation index threshold
• Produce dynamic spectrum plots (with guiding dispersion sweep lines) for the best candidates
  • This code is already written, just implement in pipeline
• Updated pipeline design requirements
• Some basic documentation

Medium Priority

• Re-write modulation index calculation in python
  • dmean - either obtain from .singlepulse files or calculate
• Error checking of input parameters
  • i.e. check that the ranges of the inputs to prepsubband make sense instead of letting it segfault
• Make a new and improved container
  • install PyQT5 via conda
  • other small changes

Lowest Priority

• Should be using python3
• prepare for using machine learning algorithm
• Revisit decimation once we have a full-resolution version working