Author V2.8: Wolfgang Konen, wolfgang.konen@fh-koeln.de
Created:
Nov’2011
Last modified: Nov’2011
(Author V1.0: Pietro Berkes, berkes@brandeis.edu)
This directory contains four demo scripts
from sfa-tk, V1.0.1:
·
sfatk_demo.m
·
long_dataset_demo.m
·
expansion_demo.m
·
getHf_demo.m
six demo functions from sfa-tk, V2.7:
·
drive1.m
·
class_demo1.m
·
class_demo2.m
·
class_demo2A.m
·
class_demo2B.m
·
class_CVdemo2.m
and one demo script from sfa-tk, V2.7:
·
batch_2A.m
Alphabetical list of files in sfa-tk/demo/
- add_rotated_copies.m: (gesture data) add
records which are copies of the original gesture records, but randomly
rotated by small angles in each direction
- batch_2A.m: script to start class_demo2A.m, but w/o any graphics output
(useful for invocation from Java programs)
- class_demo1.m: demo for classification
experiments acc. to [Berkes05] on synthetic data.
- class_demo2.m: demo for classification
experiments acc. to [Berkes05] on real data (UCI repository or gesture data).
- class_CVdemo2.m: demo for classification
experiments acc. to [Berkes05] with cross validataion on real data (UCI
repository or gesture data). [Koch10a]
- class_demo2A.m: demo usage of sfaClassModel.m: do training and store all model elements on file.
- class_demo2B.m: demo usage of sfaClassPredict.m: reload model elements from file(s) created with class_demo2A.m and do prediction.
- class_demo2A_2B.m:
demo usage of both, 2A and 2B. Together both do the same as class_demo2.m,
but with storage of all model elements on file so that prediction can be
done at a later point in time or with another program (w/o the need to do
SFA again)
- dataLoad.m: load
several data sets, helper file for class_demo2.m and class_demo2A.m
- dataLoadCV.m:
load several data sets, helper file for class_CVdemo2.m
- divide_rand.m: divide data set randomly
into training and test set
- drive1.m: reproduces
the driving force experiment from [Konen09a]
(similar to [Wis03c])
- expansion_demo.m:
shows how to perform SFA on user-defined function spaces. This script uses
two functions defined in the subdirectory expansion_demo/.
- getHf_demo.m:
illustrates how to use the sfa_getHf function.
- long_dataset_demo.m:
solves the same problem as sfatk_demo.m. It illustrates how to perform SFA
on long data sets.
- mk_confmat.m: make confusion matrix
- parallel_plot.m: parallel plot
- sfatk_demo.m: reproduces an example from [WisS02],
Figure 2 and illustrates the basic sfa-tk functions.
Alphabetical list of datasets available in sfa-tk/demo/data
Notes:
·
hk_gesture_set1-norm1.mat is not part of the ZIP
software distribution.
·
When running certain demos, a subdirectory
sfa-tk/demo/datsave will be created for temporary save files.
Literature
- [WisS02] Wiskott, L.
and Sejnowski, T.J. (2002), Slow Feature Analysis: Unsupervised
Learning of Invariances, Neural Computation, 14(4):715-770
- [Wis03c] Wiskott, L. Estimating
driving forces of nonstationary time series with slow feature analysis. arXiv.org
e-Print archive, http://arxiv.org/abs/cond-mat/0312317, December 2003.
- [Berkes05] Berkes,
P. Pattern recognition with Slow Feature Analysis. Cognitive Sciences EPrint Archive (CogPrint) 4104, http://cogprints.org/4104/ (2005).
- [Berkes03] Berkes,
P. SFA-TK: Slow Feature Analysis Toolkit for Matlab
(v.1.0.1).
http://itb.biologie.hu-berlin.de/~berkes/software/sfa-tk/sfa-tk.shtml
or
http://people.brandeis.edu/~berkes/software/sfa-tk/index.html.
- [Konen09a] Konen, W., Koch, P.
(2009). How slow is slow? SFA detects signals
that are slower than the driving force.
arXiv.org e-Print archive, http://arxiv.org/abs/0911.4397v1, November 2009 (PDF)
- [Konen09b] Konen, W.
(2009). On the numeric stability of the SFA implementation sfa-tk. arXiv.org e-Print archive, http://arxiv.org/abs/0912.1064,
December 2009. (PDF)
·
[Koch10a] Koch, P., Konen,
W., Hein, K., Gesture
Recognition on Few Training Data using Slow Feature Analysis and Parametric
Bootstrap. In P. Sobrevilla (ed.), Proc. IEEE World Congress on Computational Intelligence (WCCI), Barcelona, July 2010. (PDF)
·
[Konen11a] Konen, W. (2011). Der
SFA-Algorithmus für Klassifikation. CIOP Technical Report 08/2011, Cologne
University of Applied Sciences.
·
[Konen11b] Konen, W. (2011). SFA
classification with few training data: Improvements with parametric bootstrap.
CIOP Technical Report 09/2011, Cologne University of Applied Sciences.