Skip to content

Optimize QMix

Latest
Compare
Choose a tag to compare
@garrettj403 garrettj403 released this 09 Jul 04:03
· 34 commits to master since this release

Optimization

  • Added more detailed benchmarking and profiling scripts (found in QMix/profile/).
  • qmix.qtcurrent.qtcurrent is now ~20% faster for 1 tone.
  • qmix.harmonic_balance.harmonic_balance is ~20% faster for 1 or 2 tones, and ~10% faster for 3 tones.

Functionality Changes

  • Renamed cct.vph to cct.freq (in qmix.circuit.EmbeddingCircuit). The normalized photon voltage is equivalent to the normalized frequency. I think frequency is easier to understand so I changed it.
  • Deleted the command line scripts (bin/). There was only one script and it was pretty obscure so I descided to delete it.
  • Moved qtcurrent_all_freq() function from qmix.qtcurrent to qmix.harmonic_balance. Again, it's a pretty obscure function and it's only used in the harmonic_balance module, so I moved it.
  • Deleted qtcurrent_std from qmix.qtcurrent. This function calculated the "standard" output values, but this was pretty arbitrary. It also didn't really simplify the package at all so I deleted it.

Documentation

  • Travis CI now automatically rebuilds the online documentation on each new commit.
  • Added information on how to contribute to QMix to README.
  • Added docstrings to many private functions. This should allow new users to understand the inner workings a bit better.

Miscellaneous Minor Changes

  • QMix/examples/ renamed QMix/notebooks/
  • Removed codemeta.json and setup.cfg.
  • Fixed several bugs with Travis CI.