These scripts were written by Irene Kaplow for running Shogun with various kernels. The following scripts involve kernels that do not account for positional information: runShogunSVMDNALinearStringKernel.py, runShogunSVMDNALocalAlignmentKernel.py, runShogunSVMDNAOligoStringKernel.py, runShogunSVMDNASpectrumKernel.py, runShogunSVMDNASubsequenceStringKernel.py, runShogunSVMDNAUlongSpectrumKernel.py, runShogunSVMDNAWeightedCommonWordKernel.py, and runSogunSVRSpectrumKernel.py. The following scripts involve kernels that do account for positional information: runShogunSVMDNAFixedDegreeKernel.py, runShogunSVMDNAWDKernel.py, runShogunSVMDNANoPositionKernel.py (this is a WD kernel without shifts), and runShogunSVRWDKernel.py. Every script with "SVM" in its name is for an SVM; every script with "SVR" in its name is for an SVR. Every script with "DNA" in its name is for a DNA alphabet; all other scripts are for protein alphabets. The following scripts involve SVMs that are extremly slow to train: runShogunSVMDNALinearStringKernel.py, runShogunSVMDNALocalAlignmentKernel.py, runShogunSVMDNAOligoStringKernel.py, and runShogunSVMDNASubsequenceStringKernel.py. All other SVMs should train in less than half a day on 10,000 examples with k/degree <= 11.
-
Notifications
You must be signed in to change notification settings - Fork 2
kundajelab/ShogunScripts
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
These scripts were written by Irene Kaplow for running Shogun with various kernels.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published