NEWS.md

In mrf2d: Markov Random Field Models for Image Analysis

mrf2d 1.0

• Adds references to the manuscript published in the Journal of Statistical Software.

mrf2d 0.6

• Adds symmetric family for equality constraints of the potentials array.

mrf2d 0.5.1

• Added vec_description() function to explicitly describe what relative positions and interactions are associated with each element of the summarized vector of parameters/sufficient statistics in each parameter restriction family.

mrf2d 0.5

• rpositions() can now be used as an alternative to create mrfi objects based on explicitly described relative positions with a simpler syntax than mrfi().
• Added a vignette on analyzing 3d data via 2d projections in mrf2d.
• Fixed bug that caused the R session to crash when calling smr_stat() with incomplete lattices.

mrf2d 0.4.1

• Added rmrf2d_mc to simplify the use of Monte-Carlo methods from randomly generated fields.
• Minor documentation updates.

mrf2d 0.4

• Added mrfout and hmrfout classes for model fitting functions output. A summary() and plot() methods are also available.
• Simplified structure of documentation files.

mrf2d 0.3

• Co-ocurrence histogram computing is now exported in the cohist() function.
• Computing of conditional probabilities for individual positions are now exported in the cp_mrf2d() function.
• All the conversions from/to arrays to/from vectors are now available. See the help file for smr_stat.
• Additional example data included in field1 and hfield1.
• Documentation improved, including paper available in: https://arxiv.org/abs/2006.00383

mrf2d documentation built on Jan. 26, 2022, 1:06 a.m.