search

markovmodel / thermotools

Winter is coming...
spdy://probably.not.this.year
GNU Lesser General Public License v3.0
12 stars 4 forks source link

Estimators need to (optimally) return convergence values #10

Closed franknoe closed 8 years ago

franknoe commented 9 years ago

Iterative estimators such as WHAM etc must be able to return information about the convergence, i.e. the max-error history. It would also be useful to (optionally) access the likelihood history.

We have something like this for msmtools.estimation.transition_matrix (although that could be improved too)

cwehmeyer commented 8 years ago

WHAM and dTRAM now give (optional) access to the max-error and loglikelihood history via the arguments err_out and lll_out (output every *_out steps, default=0) in their estimate() functions.

Is that sufficient to close this issue?

franknoe commented 8 years ago

yes

Am 18/11/15 um 16:48 schrieb Christoph Wehmeyer:

WHAM and dTRAM now give (optional) access to the max-error and loglikelihood history via the arguments |err_out| and |lll_out| (output every |*_out| steps, default=0) in their |estimate()| functions.

Is that sufficient to close this issue?

— Reply to this email directly or view it on GitHub https://github.com/markovmodel/thermotools/issues/10#issuecomment-157755072.

Prof. Dr. Frank Noe Head of Computational Molecular Biology group Freie Universitaet Berlin

Phone: (+49) (0)30 838 75354 Web: research.franknoe.de

Mail: Arnimallee 6, 14195 Berlin, Germany