Added a folder qutrit_lstm with a copy of the vanilla lstm network. It can handle three different values for the strong readout results: 0, 1 and 2. Right now it's set to discard trajectories with strong ro = 2, such that it can output qubit expectations X, Y and Z.
Switched to a more user friendly settings.yaml file which stores all settings for running prep.py, train.py and analyze.py in a single yaml file.
Note: does not affect any files in the vanilla_lstm folder.
qutrit_lstmwith a copy of the vanilla lstm network. It can handle three different values for the strong readout results: 0, 1 and 2. Right now it's set to discard trajectories with strong ro = 2, such that it can output qubit expectations X, Y and Z.settings.yamlfile which stores all settings for runningprep.py,train.pyandanalyze.pyin a single yaml file.Note: does not affect any files in the
vanilla_lstmfolder.