Carlos Hernández
Carlos Hernández
re: #11
Have a version of this working offline. Would be nice to add it to this repo.
As demonstrated in [Note: Variational Encoding of Protein Dynamics Benefits from Maximizing Latent Autocorrelation](https://arxiv.org/abs/1803.06449) by @HWaymentSteele
inspiration: https://garstats.wordpress.com/2017/11/28/trimmed-means/
e.g. scalers, dihedral featurizers, etc.
pinging @liusong299 The success of APM tests seems heavily dependent on this [random_state](https://github.com/msmbuilder/msmbuilder/blob/master/msmbuilder/cluster/apm.py#L222) being set to zero. Please review this code, and provide a more reproducible test; otherwise, it will...
From Xuhui's group http://aip.scitation.org/doi/abs/10.1063/1.4995558 (h/t @msultan)
As seen in - https://en.wikipedia.org/wiki/Hidden_Markov_model#Examples - http://emma-project.org/latest/api/generated/pyemma.msm.bayesian_hidden_markov_model.html#pyemma.msm.bayesian_hidden_markov_model