182,845 research outputs found
numbbo/coco: 2.6
Release 2.6 for BBOB-2022 workshop experiments
New Features:
bbob-constrained test suite with 54 problems in 6 dimensions in code-experiments
Rescale objective to avoid too easy targets
Disguise function gradient
Add inactive constraints
Add the rotated Rastrigin function
Post-processing of data with constraints
DataSetList.by method
New function groups (separable, ill-conditioned, multimodal and all; by number of constraints)
number_of_constraints, has_constraints attributes of a Testbed
Dynamic caption setting whether the testbed has constraints or not.
New files ppfigcons1.py and compall/ppfigcons.py for scaling plots vs number of constraints
addressing issue #1117: runtime distributions are by default instance-wise balanced, such that providing more runs/trials on one instance than others does not introduce a bias.
Documentation online
Bug fixes:
Improvements of the C code https://github.com/numbbo/coco/pull/2070
ERT computation https://github.com/numbbo/coco/commit/1bc7bad23b8b3c2e3245a438f5c7094faa807897
inconsistent data b/2009/CMA-ESPLUSSEL_auger https://github.com/numbbo/coco/commit/2af85f1a9cca9e3517e1c22fb9aa494965e057a8
improved sample size computation https://github.com/numbbo/coco/commit/793688026e730e8547aa4425dd32cf14d62cf28d
fix write_tables https://github.com/numbbo/coco/issues/2038
add install-user option in circleci https://github.com/numbbo/coco/commit/3c6fb2a16dad19e980de6d6a23b7a0cf632208e6
Other updates:
LaTeX templates updated for BBOB-2022 workshop
polishing of ECDFs and x limits in post-processing
Build files https://github.com/numbbo/coco/pull/207
A Guide to CoCo
CoCo is a program for estimation, test and model search among hierarchical interaction models for large complete contingency tables. The name CoCo is derivated of "Co"mplete "Co"ntingency tables, since the initial program could only handle complete tables, but the program has been enhanced to handle incomplete tables. CoCo works especially efficiently on graphical models, and some of the commands are designed to handle graphical models.
Cross-recurrence quantification analysis of categorical and continuous time series: an R package
Recommended from our members
Quantifying the Dynamics of Interpersonal Interaction: A Primer on Cross-Recurrence Quantification Analysis using R
Il sovraffollamento carcerario e l'ultimatum di Strasburgo
Il contributo analizza le molteplici cause che hanno contribuito a determinare il grave fenomeno del sovraffollamento carcerario. Particolare attenzione è riservata all'abuso della custodia cautelare in carcer
Le forze di polizia in Italia dal 1943 al 1945
Il ruolo della polizia durante la seconda guerra mondial
- …
