1 2How to configure the 3dsvm plugin for real-time experiments using plugout_drive: 3=============================================================================== 4 5 plugout_drive is a command-line program that can be used to drive (control) 6 AFNI (please see README.driver for more details) and allows the user to automate 7 the configuration of the 3dsvm plugin for real-time experiments. 8 9 Using plugout_drive to set up the 3dsvm plugin for real-time experiments is 10 very similar to the usage of the command-line program 3dsvm for off-line 11 SVM analysis. Most of the 3dsvm (and SVM-Light) command-line options can be 12 used in conjunction with plugout_drive. 13 14 Usage: 15 ------ 16 17 18 plugout_drive -com '3DSVM [options]' 19 20 21 Examples: 22 --------- 23 24 Training: 25 plugout_drive -com '3DSVM -rt_train -trainlabels run1_categories.1D ... 26 -mask mask+orig -model model_run1' 27 28 Testing: 29 plugout_drive -com '3DSVM -rt_test -model model_run1+orig ... 30 -stim_ip 111.222.333.444 -stim_port 5000' 31 32 N.B.: -rt_train and -rt_test serve as flags for the real-time training 33 and testing modes, respectively. No brik or nifti file is 34 specified since it is expected from the scanner (or rtfeedme). 35 36 Options: 37 -------- 38 N.B. The plugout_drive options are almost identical to the "normal" 3dsvm usage, 39 (see 3dsvm -help) but restricted to 2-class classification and regression. 40 Coming soon (or someday when asked): multi-class classification 41 42 43 44Reference: 45LaConte, S., Strother, S., Cherkassky, V. and Hu, X. 2005. Support vector 46 machines for temporal classification of block design fMRI data. 47 NeuroImage, 26, 317-329. 48 49Specific to real-time fMRI: 50S. M. LaConte. (2011). Decoding fMRI brain states in real-time. NeuroImage, 56:440-54. 51S. M. LaConte, S. J. Peltier, and X. P. Hu. (2007). Real-time fMRI using brain-state classification. Hum Brain Mapp, 208:1033–1044. 52 53 54Please also consider to reference: 55T. Joachims, Making Large-Scale SVM Learning Practical. 56 Advances in Kernel Methods - Support Vector Learning, 57 B. Schoelkopf and C. Burges and A. Smola (ed.), MIT Press, 1999. 58 59RW Cox. AFNI: Software for analysis and visualization of 60 functional magnetic resonance neuroimages. 61 Computers and Biomedical Research, 29:162-173, 1996. 62 63