code/README

   1 To compile source code :
   2
   3    mkdir -p build/stage1/src
   4    cd build/stage1/src
   5    cmake ../../../stage1/src
   6    make
   7    #repeat previous lines for stage 2
   8
   9 Usage (stage 1) :
  10
  11    mpirun -np nbProcess ppam.exe cluster ifilename nbSeriesInChunk nbClusters randomize p_for_dissims
  12
  13 Where :
  14    nbProcess = number of simultaneous processes
  15    ifilename = path to serialized dataset (read below)
  16    nbSeriesInChunk = number of time-series to process sequentially
  17    nbClusters = number of clusters
  18    randomize = 1 to dispatch time-series at random. 0 to process them in order
  19    p_for_dissims = the 'p' of L_p distance used to compute dissimilarities
  20
  21 Note : custom [de]serialization. Consider writing your own
  22 in src/TimeSeries/ folder if you plan to test the package.
  23
  24 See also src/main.c for the details.