openModeller / News: Recent posts

This release contains a new algorithm to generate virtual niches and a decent description was finally included for the Niche Mosaic algorithm. The Web service contains a few changes and a new scheduler compatible with HTCondor DagMan. The most important changes are on the Windows installer, which now: 1) includes some MSVC DLLs that could be missing in some machines, 2) fixes a problem that could erase the content of the PATH environmental variable, and 3) contains a new option to install the Python module.

This release contains a new "meta" algorithm that can use several algorithms behind the scenes to generate a consensus model. A new command-line tool (om_evaluate) can be used to return raw model values given a set of points and an environmental scenario. A new version of the web service (OMWS 2.0) is also available, including new operations - one of them allowing whole experiments to be specified in a single request. A new raster driver now allows remote files to the fetched and used. There were also many changes in the framework itself.

This release contains a new algorithm Random Forests and new versions of ENFA, Maxent and Environmental Distance. There were also changes in the Web Service, the command line interface and the framework itself.

This release includes two new algorithms - ENFA (Ecological Niche Factor Analysis) and Niche Mosaic - and a new version of the Maximum Entropy algorithm based on the Maxent paper (Phillips et al., 2006). It also contains a few adjustments in other existing algorithms (GARP, AquaMaps, ANN, CSM and Mahalanobis distance).

This release includes adjustments in many parts of the framework, including ROC curve, algorithms (ANN and SVM), and command line tools (om_model and om_niche). There were also improvements in the modelling protocol and model statistics (possibility to use lowest presence threshold).

This release contains a new algorithm using Artificial Neural Networks, it also includes support to generate distribution maps in ARC/Info ASCII grid format, as well as important changes in the ROC Curve class. It also includes a few bugfixes and other improvements.

openModeller is a cross-platform C++ library for potential distribution modeling.

This release contains mainly new features, including:

- Drivers to directly read occurrences from GBIF or TAPIR/DarwinCore providers.
- A new command-line tool (om_points) that can be used to retrieve occurrence data using any of the available drivers.
- A new command-line tool (om_algorithm) to get information about the available algorithms.
- A new Maximum Entropy algorithm with two training methods: GIS (Generalized Iterative Scaling) and L-BFGS (Limited-Memory Variable Metric).... read more

It is our great pleasure to announce the immediate availability of openModeller Desktop Version 1.0.5 openModeller is a generic framework for carrying out fundamental niche modelling experiments - typically used to predict species distribution given a set of environmental raster layers. The openModeller Desktop application builds on the openModeller library by providing a user friendly graphical user interface for carrying out niche modelling experiments.... read more

openModeller is a cross-platform C++ library for static spatial distribution modelling, normally used to predict species distribution (fundamental niche).

This release contains some bugfixes, a change in the pattern of the Log object (now a singleton) and the following new features:

- ROC curve as part of model statistics.
- Metric Chebyshev as a new parameter of the Environmental distance algorithm.

It is our great pleasure to announce the immediate availability of openModeller Desktop Version 1.0.5 openModeller is a generic framework for carrying out fundamental niche modelling experiments - typically used to predict species distribution given a set of environmental raster layers. The openModeller Desktop application builds on the openModeller library by providing a user friendly graphical user interface for carrying out niche modelling experiments.... read more

openModeller is a cross-platform C++ library for static spatial distribution modelling, normally used to predict species distribution (fundamental niche).

This release contains some changes and feature enhancements:

- New algorithm EnvironmentalDistance supporting three different metrics: Euclidean, Mahalanobis and Gower. When used with Gower and maximum distance of 1, this algorithm should be equivalent to DOMAIN.
- It is now possible to specify absence points in the localities file, which has a new format: four mandatory columns (id, label, longitude and latitude) and one optional column (abundance). Absences are indicated with abundance equal to zero. All values must now be separated by TABs, so labels (usually scientific names) can contain spaces. The new column "id" is used to identify that particular record.
- For developers, there's initial support to cmake (should compile on GNU/Linux and Mac OSX).... read more

We are pleased to announce the first public release of openModeller Desktop (version 1.0.2).

openModeller is a cross-platform C++ library for static spatial distribution modelling, normally used to predict species distribution (fundamental niche).

openModeller Desktop provides a user friendly graphical user interface for the openModeller libary. This release is available as source code (all platforms) and as a Windows binary setup file.... read more

openModeller is a cross-platform C++ library for static spatial distribution modelling, normally used to predict species distribution (fundamental niche).

This release contains major feature enhancements and also some bugfixes. The main changes are:

- Integration with TerraLib. In addition to GDAL, openModeller's IO functions now work with Terralib raster and point data.
- Refactoring of the SOAP interface which now covers all modelling functionality, allowing remote interaction with a modelling service.
- The new version of GARP (v3) has been re-enabled.
- The default output map file type was changed to ERDAS Imagine ".img" to achieve wider compatibility with GIS software. This format provides a floating-point representation in each raster cell (reflecting probability of presence).
- Several fixes in the SWIG-Python binding.
- An installation issue with the RPM packages was fixed.... read more

This version contains a major bugfix in the model projector that could cause localized distortions in the distribution map, therefore we recommend all users to upgrade. There is also a new version of the CSM algorithm with several adjustments, and new command line tools (om_sampledump, om_create, om_project).

This version includes a new serialization mechanism for the GARP Best Subsets Procedure, and it fixes some issues related to the Mac OS X build. A new release of the GUI should be following soon.

An error during normalization offsets calculation was making DG_GARP and DG_GARP_BS produce null models most part of the time (100% omission on projected maps). Apparently the other algorithms were not affected. This problem was fixed on version 0.3.2 of both the library and the GUI. We recommend all users to upgrade.

New versions of the library and the GUI have been co-released. There were lots of changes in the library, with many improvements and bugfixes. We encourage users to upgrade.