Skip to content


Subversion checkout URL

You can clone with
Download ZIP
RSNNS: Neural Networks in R using the Stuttgart Neural Network Simulator (SNNS)
C++ C R
Branch: master

fixed two minor problems

latest commit 9b13aa4b61
Christoph Bergmeir authored
Failed to load latest commit information.
R fixed two minor problems
data RSNNS 0.4-3: see ChangeLog
tests applied patch cran version 0.4-3 to 0.4-4, and changes towards 0.4-5
ChangeLog fixed two minor problems
DESCRIPTION initial commit 0.4-8
NAMESPACE initial commit 0.4-8 preparations for version 0.4-7
RSNNS.log memory leak fix for saveNet, changelog for version 0.4-6


RSNNS: Neural Networks in R using the Stuttgart Neural Network Simulator (SNNS)

Possible TODOs for the next version:

  • fix remaining memory leaks, detected by valgrind, e.g. this one:

==32137== 206,480 (80 direct, 206,400 indirect) bytes in 1 blocks are definitely lost in loss record 2,312 of 2,347 ==32137== at 0x4A081D4: calloc (in /usr/lib64/valgrind/ ==32137== by 0x1019F8DD: SnnsCLib::allocMixupArray() (dlvq_learn.cpp:212) ==32137== by 0x101A0B28: SnnsCLib::allocArrays() (dlvq_learn.cpp:942) ==32137== by 0x101A10B7: SnnsCLib::LEARN_DLVQ(int, int, float, int, float, int) (dlvq_learn.cpp:1072) ==32137== by 0x101AB52C: SnnsCLib::kr_callNetworkFunctionSTD(int, float, int, float, int, int, int) (kernel.cpp:3952) ==32137== by 0x101AB5D6: SnnsCLib::kr_callNetworkFunction(int, float, int, float, int, int, int) (kernel.cpp:4019) ==32137== by 0x101D276F: SnnsCLib::krui_learnAllPatterns(float, int, float, int) (kr_ui.cpp:3505) ==32137== by 0x1018347A: SnnsCLib__learnAllPatterns (SnnsCLibWrapper.cpp:1278)

  • Remove printf throughout the code (replace commented printf's with Rprintf, diff of commit May-15h)
  • use the "colorspace" package for heatmaps
  • add JSS paper as vignette

TODOs for a far away future:

  • Implement Softmax/Entropy for elman?
  • Make it possible to implement learning functions in R
  • Implement more sophisticated convergence detection
Something went wrong with that request. Please try again.