Differences

This shows you the differences between two versions of the page.

Link to this comparison view

projects:fingertip_distance [2010/09/20 13:33] – created amaldoprojects:fingertip_distance [2010/09/24 10:44] (current) ehmann
Line 1: Line 1:
 +====== Distance classification using the fingertip sensor ======
 +
 +
 +===== Client programs =====
 +
 +The project contains client programs to train and test the sensors.
 +
 +
 +
 +==== training file format ====
 +The training data needs to be in the following format.
 +This allows the use in the distance evaluation program and in the libsvm-tools programs.
 +
 + <classid> 1:<SQUAL> 2:<PixAVG> 3:<PixMin> 4:<PixMax> 5:<Shutter> 6:<ShutterMax> 7:<FramePeriod> 8:<FramePeriodMin> 9:<FramePeriodMax>
 + .
 + .
 + .
 +
 +The parameters 1-9 have to be in ascending order.
 +The ''classid''s do not have to be sorted.
 +
 +
 +==== evaluate ====
 +
 +This program prints values from selected sensors.
 +It can also be used to generate training data in libsvm format
 +
 +Usage:
 + Sensor-Register-Reader
 + Parameters
 + -f <file> File to write data to - "out.txt" is used if not specified
 + -c <comment> Comment to show up atop each read
 + -n <cnt> Number of rounds per read
 + -m <map> Bitmap of sensors to be read (e.g. for sensors 1,3 value is 5)
 + -v <1/0> Save values (with description) (default=1)
 + -s <1/0> Save values in libsvm-training file format (comment=classid)(default=0)
 +
 +==== distance ====
 +
 +This program uses libsvm to predict the sensors distance to the surface.
 +It can use both training data and model files to generate the internal model.
 +The training data can also be saved as a model file.
 +Beside the distance-prediction on the sensor it can be used to test the training data. Either perform a cross validation check or test it against given test data.
 +
 +Usage:
 + Distance-Evaluation
 + Parameters
 + -f <file> Inut file - libsvm-model-file or training data
 + -m <1/0> Use model as input file
 + -n <cnt> Number of lines (used) in training-data-file
 + -t <file> Testfile (instead of sensor input)
 + -p Print probability estimates instead of class predictions
 + -T Make cross validation test
 + -s Save model file (use -o)
 + -o <file> Output filename for model file
 +
 +
  
 
Recent changes RSS feed Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki