Martin K. Scherer

On 9/24/18 11:40 PM, Eugen Hruska wrote: > Here https://drive.google.com/file/d/1hYZoE72H2C-2ahiAzOLcOzHXfRtER8TQ/view?usp=sharing > > Thanks. The eigenvectors have indeed only a dimension of 4. This means however that solving the generalized eigenvalue...

On 10/12/18 1:11 AM, Eugen Hruska wrote: > any update? > the bug occurs because the matrix real rank is lower than the requested dimension, this should of course not...

I'd replace it.

On 09/07/2017 11:25 AM, fabian-paul wrote: > So what is the workaround that you would suggest for the moment to call > methods from the superclass explicitly in a derived...

Are you talking about makeing properties of mdtraj.Topology available? So you do not need to write: featurizer.topology.n_atoms ?

In our debug session we figured out that the samples are already restricted to the observable set, but during the final submodel call in estimate, we try to restrict again...

@kochaneks thanks for reporting. I think that for larger lag times, the active set collapses (e.g. contains unvisited states, and is therefore not connected anymore). This means that you can...

Could you please provide the dtrajs as a non-pickled format, like compressed ascii?

Thanks for sharing. Could you provide the code snippet producing the error. I can compute a cktest with 5 metastable states on your data with a BayesianMSM without any warning/error.

We wanna allow our users to change/extend our provided schemata, e.g. adding required attributes etc.