PRSD Studio - version 1.1 (26-Oct-2008)
name
description
Visualization
sdscatter
Interactive scatter plot and visualization of classifier outputs
sdimage
Visualize image stored in a dataset
sddrawroc
Draw interactive ROC plot
Decisions
sddecide
Define an operating point or perform decisions
sdp_decide
Pipeline action: Decisions at a given operating point
getcurop
Get index of the current operating point
setcurop
Set current operating point in decision mapping or ROC object
getcuropdata
Returns data (threshold or weights) for the current op.point
ROC analysis
sdrelab
Relabel the dataset lablist and featlab
sdroc
Estimate ROC on a dataset with classifier outputs
sddrawroc
Draw interactive ROC plot
setcurop
Set current operating point in decision mapping or ROC object
Evaluation
sdcrossval
Perform cross-validation
sdconfmatind
Get indices of samples in a confusion matrix field
sdstackgen
Stacked generalization set of classifier outputs
Algorithm deployment
sdconvert
Convert object into a pipeline or post-process a pipeline
sdexe
Execute pipeline or prtools mapping using the libPRSD mex library
sdexport
Export the trained mapping as a recognition pipeline
getoutput
Returns output type of a pipeline or a mapping
getdectable
Returns the decision table for a pipeline or mapping
getclftable
Returns the list of classifiers in a pipeline
Execution pipelines
sdp_1nn
Pipeline action: Distance to the 1-st nearest neighbor (one-class and multi-class)
sdp_affine
Pipeline action: Affine transformation
sdp_cascade
Pipeline action: two-stage cascade
sdp_decide
Pipeline action: Decisions at a given operating point
sdp_fsel
Pipeline action: Feature selection
sdp_knnmc
Pipeline action: multi-class k-th nearest neighbor with output normalization
sdp_knnoc
Pipeline action: One-class k-th nearest neighbor data description
sdp_norm
Pipeline action: Normalize output to sum to one
sdp_normal
Pipeline action: Gaussian model
sdp_parallel
Pipeline action: Parallel combiner
sdp_parzen
Pipeline action: Parzen density estimator or classifier
sdp_prox
Pipeline action: Proximity computation
sdp_sigmoid
Pipeline action: Sigmoid transformation
sdp_stack
Pipeline action: Stack combiner
sdp_svc
Pipeline action: Support vector machine classifier
sdp_svdd
Pipeline action: Support vector data description
sdp_tree
Pipeline action: Decision tree classifier
Cluster Analysis
sdeaclust
Evidence Accumulation Clustering
Per-sample meta-data in datasets
setprop
Set dataset property (meta-data)
getprop
Get dataset property (meta-data)
findprop
Find indices of examples with meta-data values
printprop
Print property values for samples in the dataset
sduniqueprop
Create a dataset with a sample per unique property
| name | description |
|---|---|
| Visualization | |
| sdscatter | Interactive scatter plot and visualization of classifier outputs |
| sdimage | Visualize image stored in a dataset |
| sddrawroc | Draw interactive ROC plot |
| Decisions | |
| sddecide | Define an operating point or perform decisions |
| sdp_decide | Pipeline action: Decisions at a given operating point |
| getcurop | Get index of the current operating point |
| setcurop | Set current operating point in decision mapping or ROC object |
| getcuropdata | Returns data (threshold or weights) for the current op.point |
| ROC analysis | |
| sdrelab | Relabel the dataset lablist and featlab |
| sdroc | Estimate ROC on a dataset with classifier outputs |
| sddrawroc | Draw interactive ROC plot |
| setcurop | Set current operating point in decision mapping or ROC object |
| Evaluation | |
| sdcrossval | Perform cross-validation |
| sdconfmatind | Get indices of samples in a confusion matrix field |
| sdstackgen | Stacked generalization set of classifier outputs |
| Algorithm deployment | |
| sdconvert | Convert object into a pipeline or post-process a pipeline |
| sdexe | Execute pipeline or prtools mapping using the libPRSD mex library |
| sdexport | Export the trained mapping as a recognition pipeline |
| getoutput | Returns output type of a pipeline or a mapping |
| getdectable | Returns the decision table for a pipeline or mapping |
| getclftable | Returns the list of classifiers in a pipeline |
| Execution pipelines | |
| sdp_1nn | Pipeline action: Distance to the 1-st nearest neighbor (one-class and multi-class) |
| sdp_affine | Pipeline action: Affine transformation |
| sdp_cascade | Pipeline action: two-stage cascade |
| sdp_decide | Pipeline action: Decisions at a given operating point |
| sdp_fsel | Pipeline action: Feature selection |
| sdp_knnmc | Pipeline action: multi-class k-th nearest neighbor with output normalization |
| sdp_knnoc | Pipeline action: One-class k-th nearest neighbor data description |
| sdp_norm | Pipeline action: Normalize output to sum to one |
| sdp_normal | Pipeline action: Gaussian model |
| sdp_parallel | Pipeline action: Parallel combiner |
| sdp_parzen | Pipeline action: Parzen density estimator or classifier |
| sdp_prox | Pipeline action: Proximity computation |
| sdp_sigmoid | Pipeline action: Sigmoid transformation |
| sdp_stack | Pipeline action: Stack combiner |
| sdp_svc | Pipeline action: Support vector machine classifier |
| sdp_svdd | Pipeline action: Support vector data description |
| sdp_tree | Pipeline action: Decision tree classifier |
| Cluster Analysis | |
| sdeaclust | Evidence Accumulation Clustering |
| Per-sample meta-data in datasets | |
| setprop | Set dataset property (meta-data) |
| getprop | Get dataset property (meta-data) |
| findprop | Find indices of examples with meta-data values |
| printprop | Print property values for samples in the dataset |
| sduniqueprop | Create a dataset with a sample per unique property |
Next step: Reference: Supported PRTools and DDTools mappings