!!IMPORTANT!! The main steps from this plugin have been moved to cedar's main plugin.
This plugin provides steps related to self-organizing maps and other learning methods for cedar.
How do I get set up?
To get started with this plugin, copy the cedar.conf.example to cedar.conf and enter the path to the cedar version you want to use, either in CEDAR_HOME if you use a local version of cedar, or in CEDAR_HOME_INSTALLED if you are using a version of cedar installed with make install.
It is also recommended to set the deprecation level to one unless deprecated steps are explicitly needed.
Then cd into the plugin folder and type
mkdir build cd build cmake .. make
The plugin is then found in the build directory and can be loaded into cedar as usual.
What does the plugin provide?
The plugin provides many steps; here, we just highlight the most important ones.
- CrossCorrelation: A step to calculate cross-correlation between two input matrices. This step supports 1-4 dimensional inputs, as well as selecting along which dimensions cross correlation occurrs. It also provides functionality for padding and cropping the in- and output.
- PatternMemory: Can learn a discrete set of patterns corresponding to label.
- SteeringFilter, SteerG2, SteerH2, SteerC2C3: Preprocessing steps for steerable filter.
- SteeringEnergy: Calculates the steering energy based on C2 and C3 responses.
- SteeringAngle: Calculates the dominant angle based on C2 and C3 responses.
- ReceptiveFieldHistogram: Extracts histograms within localized receptive fields.
- WeightedSumOfSlices: Contracts a matrix along one dimension like the projection step, but weights each slice by another input.
- steering_filter_preprocessing: Extracts edges and edge orientations using steerable filters.