-A15.) Freely available software packages for NN simulation ? [This is a bit chaotic and needs reorganization. A bit more information about what the various programs can do, on which platform they run, and how big they are would also be nice. And some important packages are still missing (?) Who volunteers for that ?] 1. Rochester Connectionist Simulator A quite versatile simulator program for arbitrary types of neural nets. Comes with a backprop package and a X11/Sunview interface. anonymous FTP from cs.rochester.edu (192.5.53.209) directory : pub/simulator files: README (8 KB) (documentation:) rcs_v4.2.justdoc.tar.Z (1.6 MB) (source code:) rcs_v4.2.justsrc.tar.Z (1.4 MB) 2. UCLA-SFINX ftp 131.179.16.6 (retina.cs.ucla.edu) Name: sfinxftp Password: joshua directory: pub/ files : README sfinx_v2.0.tar.Z Email info request : sfinx@retina.cs.ucla.edu 3. NeurDS request from mcclanahan%cookie.dec.com@decwrl.dec.com simulator for DEC systems supporting VT100 terminal. OR anonymous ftp gatekeeper.dec.com [16.1.0.2] directory: pub/DEC file: NeurDS031.tar.Z ( please check may be NeurDSO31.tar.Z ) 4. PlaNet5.7 (also known as SunNet) ftp 133.15.240.3 (tutserver.tut.ac.jp) pub/misc/PlaNet5.7.tar.Z or ftp 128.138.240.1 (boulder.colorado.edu) pub/generic-sources/PlaNet5.7.tar.Z (also the old PlaNet5.6.tar.Z) A popular connectionist simulator with versions to run under X Windows, and non-graphics terminals created by Yoshiro Miyata (Chukyo Univ., Japan). 60-page User's Guide in Postscript. Send any questions to miyata@sccs.chukyo-u.ac.jp 5. GENESIS GENESIS 1.4.1 (GEneral NEural SImulation System) is a general purpose simulation platform which was developed to support the simulation of neural systems ranging from complex models of single neurons to simulations of large networks made up of more abstract neuronal components. Most current GENESIS applications involve realistic simulations of biological neural systems. Although the software can also model more abstract networks, other simulators are more suitable for backpropagation and similar connectionist modeling. May be obtained via FTP from genesis.cns.caltech.edu [131.215.137.64]. Use 'telnet' to genesis.cns.caltech.edu beforehands and login as the user "genesis" (no password required). If you answer all the questions asked of you an 'ftp' account will automatically be created for you. You can then 'ftp' back to the machine and download the software (ca. 3 MB). Contact: genesis@cns.caltech.edu. 6. Mactivation anonymous ftp from bruno.cs.colorado.edu [128.138.243.151] directory: /pub/cs/misc file: Mactivation-3.3.sea.hqx 7. CMU Connectionist Archive There is a lisp backprop simulator in the connectionist archive. unix> ftp b.gp.cs.cmu.edu (or 128.2.242.8) Name: ftpguest Password: cmunix ftp> cd connectionists/archives ftp> get backprop.lisp 8. Cascade Correlation Simulator There is a LISP and C version of the simulator based on Scott Fahlman's Cascade Correlation algorithm, who also created the LISP version. The C version was created by Scott Crowder. Anonymous ftp from pt.cs.cmu.edu (or 128.2.254.155) directory /afs/cs/project/connect/code files cascor1.lisp (56 KB) cascor1.c (108 KB) 9. Quickprop A variation of the back-propagation algorithm developed by Scott Fahlman. A LISP and C version can be obtained in the same directory as the cascade correlation simulator above. (25 KB) 10. DartNet DartNet is a Macintosh-based Neural Network Simulator. It makes full use of the Mac's graphical interface, and provides a number of powerful tools for building, editing, training, testing and examining networks. This program is available by anonymous ftp from dartvax.dartmouth.edu [129.170.16.4] as /pub/mac/dartnet.sit.hqx (124 KB) Copies may also be obtained through email from bharucha@dartmouth.edu. Along with a number of interface improvements and feature additions, v2.0 is an extensible simulator. That is, new network architectures and learning algorithms can be added to the system by writing small XCMD-like CODE resources called nDEF's ("Network Definitions"). A number of such architectures are included with v2.0, as well as header files for creating new nDEF's. Contact: sean@coos.dartmouth.edu (Sean P. Nolan) 11. SNNS "Stuttgart Neural Network Simulator" from the University of Stuttgart, Germany. A luxurious simulator for many types of nets; with X11 interface: Graphical 2D and 3D topology editor/visualizer, training visualisation, etc. Currently supports backpropagation (vanilla, online, with momentum term and flat spot elimination, batch, time delay), counterpropagation, quickprop, backpercolation 1, generalized radial basis functions (RBF), RProp, ART1, ART2, ARTMAP, Cascade Correlation, Recurrent Cascade Correlation, Dynamic LVQ, Backpropagation through time (for recurrent networks), batch backpropagation through time (for recurrent networks), Quickpropagation through time (for recurrent networks), and is user-extendable. ftp: ftp.informatik.uni-stuttgart.de [129.69.211.2] directory /pub/SNNS file SNNSv3.0.tar.Z OR SNNSv3.0.tar.Za[a-d] ( 826 KB) manual SNNSv2.1.Manual.ps.Z (1270 KB) SNNSv2.1.Readme (8052 Bytes) 12. Aspirin/MIGRAINES Aspirin/MIGRAINES 6.0 consists of a code generator that builds neural network simulations by reading a network description (written in a language called "Aspirin") and generates a C simulation. An interface (called "MIGRAINES") is provided to export data from the neural network to visualization tools. The system has been ported to a large number of platforms. The goal of Aspirin is to provide a common extendible front-end language and parser for different network paradigms. The MIGRAINES interface is a terminal based interface that allows you to open Unix pipes to data in the neural network. This replaces the NeWS1.1 graphical interface in version 4.0 of the Aspirin/MIGRAINES software. The new interface is not a simple to use as the version 4.0 interface but is much more portable and flexible. The MIGRAINES interface allows users to output neural network weight and node vectors to disk or to other Unix processes. Users can display the data using either public or commercial graphics/analysis tools. Example filters are included that convert data exported through MIGRAINES to formats readable by Gnuplot 3.0, Matlab, Mathematica, and xgobi. The software is available from two FTP sites: CMU's simulator collection on "pt.cs.cmu.edu" (128.2.254.155) in /afs/cs/project/connect/code/am6.tar.Z". and UCLA's cognitive science machine "ftp.cognet.ucla.edu" (128.97.50.19) in alexis/am6.tar.Z The compressed tar file is a little less than 2 megabytes. 13. Adaptive Logic Network kit Available from menaik.cs.ualberta.ca. This package differs from the traditional nets in that it uses logic functions rather than floating point; for many tasks, ALN's can show many orders of magnitude gain in training and performance speed. Anonymous ftp from menaik.cs.ualberta.ca [129.128.4.241] README /pub/atree/atree.readme (7 KB) unix source code and examples: /pub/atree/atree2.tar.Z (145 KB) Postscript documentation: /pub/atree/atree2.ps.Z ( 76 KB) MS-Windows 3.x executable: /pub/atree/a27exe.exe (412 KB) MS-Windows 3.x source code: /pub/atree/atre27.exe (572 KB) 14. NeuralShell Available from FTP site quanta.eng.ohio-state.edu (128.146.35.1) in directory "pub/NeuralShell", filename "NeuralShell.tar". 15. PDP The PDP simulator package is available via anonymous FTP at nic.funet.fi (128.214.6.100) in /pub/sci/neural/sims/pdp.tar.Z (0.2 MB) The simulator is also available with the book "Explorations in Parallel Distributed Processing: A Handbook of Models, Programs, and Exercises" by McClelland and Rumelhart. MIT Press, 1988. Comment: "This book is often referred to as PDP vol III which is a very misleading practice! The book comes with software on an IBM disk but includes a makefile for compiling on UNIX systems. The version of PDP available at nic.funet.fi seems identical to the one with the book except for a bug in bp.c which occurs when you try to run a script of PDP commands using the DO command. This can be found and fixed easily." 16. Xerion Xerion is available via anonymous ftp from ftp.cs.toronto.edu in the directory /pub/xerion. xerion-3.1.ps.Z (153 kB) and xerion-3.1.tar.Z (1322 kB) plus several concrete simulators built with xerion (about 40 kB each). Xerion runs on SGI and Sun machines and uses X Windows for graphics. The software contains modules that implement Back Propagation, Recurrent Back Propagation, Boltzmann Machine, Mean Field Theory, Free Energy Manipulation, Hard and Soft Competitive Learning, and Kohonen Networks. Sample networks built for each of the modules are also included. Contact: xerion@ai.toronto.edu 17. Neocognitron simulator An implementation is available for anonymous ftp at [128.194.15.32] tamsun.tamu.edu as /pub/neocognitron.Z.tar or [129.12.21.7] unix.hensa.ac.uk as /pub/uunet/pub/ai/neural/neocognitron.tar.Z The simulator is written in C and comes with a list of references which are necessary to read to understand the specifics of the implementation. The unsupervised version is coded without (!) C-cell inhibition. 18. Multi-Module Neural Computing Environment (MUME) MUME is a simulation environment for multi-modules neural computing. It provides an object oriented facility for the simulation and training of multiple nets with various architectures and learning algorithms. MUME includes a library of network architectures including feedforward, simple recurrent, and continuously running recurrent neural networks. Each architecture is supported by a variety of learning algorithms. MUME can be used for large scale neural network simulations as it provides support for learning in multi-net environments. It also provide pre- and post-processing facilities. The modules are provided in a library. Several "front-ends" or clients are also available. MUME can be used to include non-neural computing modules (decision trees, ...) in applications. The software is written in 'C' and is being used on Sun and DEC workstations. Efforts are underway to port it to the Fujitsu VP2200 vector processor using the VCC vectorising C compiler. MUME is made available to research institutions on media/doc/postage cost arrangements. Contact: Marwan Jabri, SEDAL, Sydney University Electrical Engineering, NSW 2006 Australia, marwan@sedal.su.oz.au 19. LVQ_PAK, SOM_PAK These are packages for Learning Vector Quantization and Self-Organizing Maps, respectively. They have been built by the LVQ/SOM Programming Team of the Helsinki University of Technology, Laboratory of Computer and Information Science, Rakentajanaukio 2 C, SF-02150 Espoo, FINLAND There are versions for Unix and MS-DOS available from cochlea.hut.fi (130.233.168.48) in /pub/lvq_pak/lvq_pak-2.1.tar.Z (340 kB, Unix) /pub/lvq_pak/lvq_p2r1.exe (310 kB, MS-DOS self-extract archive) /pub/som_pak/som_pak-1.1.tar.Z (246 kB, Unix) /pub/som_pak/som_p1r1.exe (215 kB, MS-DOS self-extract archive) 20. SESAME (Software Environment for the Simulation of Adaptive Modular Systems) SESAME is a prototypical software implementation which facilitates * Object-oriented building blocks approach. * Contains a large set of C++ classes useful for neural nets, neurocontrol and pattern recognition. No C++ classes can be used as stand alone, though! * C++ classes include CartPole, nondynamic two-robot arms, Lunar Lander, Backpropagation, Feature Maps, Radial Basis Functions, TimeWindows, Fuzzy Set Coding, Potential Fields, Pandemonium, and diverse utility building blocks. * A kernel which is the framework for the C++ classes and allows run-time manipulation, construction, and integration of arbitrary complex and hybrid experiments. * Currently no graphic interface for construction, only for visualization. * Platform is SUN4, XWindows Unfortunately no reasonable good introduction has been written until now. We hope to have something soon. For now we provide papers (eg. NIPS-92), a reference manual (>220 pages), source code (ca. 35.000 lines of code), and a SUN4-executable by ftp only. Sesame and its description is available for anonymous ftp on ftp ftp.gmd.de [129.26.8.90] in the directories gmd/as/sesame and gmd/as/paper Questions please to sesame-request@gmd.de There is only very limited support available. Currently we can not handle many users. 21. Nevada Backpropagation (NevProp) NevProp is a user-friendly backpropagation program written in C for UNIX, Macintosh, and DOS. The original version was Quickprop 1.0 by Scott Fahlman, as translated from Common Lisp into C by Terry Regier. The quickprop algorithm itself was not substantively changed, but we inserted options to force gradient descent (per-epoch or per-pattern) and added generalization & stopped training, c index, and interface enhancements. *** FEATURES: NevProp version 1.15... o UNLIMITED (except by machine memory) number of input PATTERNS; o UNLIMITED number of input, hidden, and output UNITS; o Arbitrary CONNECTIONS among the various layers' units; o Clock-time or user-specified RANDOM SEED for initial random weights; o Choice of regular GRADIENT DESCENT or QUICKPROP; o Choice of LOGISTIC or TANH activation functions; o Choice of PER-EPOCH or PER-PATTERN (stochastic) weight updating; o GENERALIZATION to a test dataset; o AUTOMATICALLY STOPPED TRAINING based on generalization; o RETENTION of best-generalizing weights and predictions; o Simple but useful bar GRAPH to show smoothness of generalization; o SAVING of results to a file while working interactively; o SAVING of weights file and reloading for continued training; o PREDICTION-only on datasets by applying an existing weights file; o In addition to RMS error, the concordance, or c index is displayed. The c index shows the correctness of the RELATIVE ordering of predictions AMONG the cases; ie, it considers all possible PAIRS of vectors. This statistic is identical to the area under the receiver operating characteristic (ROC) curve, widely used in technology assessment. *** AVAILABILITY: The most updated version of NevProp will be made available by anonymous ftp from the University of Nevada, Reno: "unssun.scs.unr.edu" [134.197.10.128] directory "pub/goodman/nevpropdir" *** SUPPORT: Limited support is available from Phil Goodman (goodman@unr.edu), University of Nevada Center for Biomedical Research. 22. Fuzzy ARTmap Available for anonymous ftp from park.bu.edu [128.176.121.56] as /pub/fuzzy-artmap.tar.Z (44 kB) (This is just a small example program.) For some of these simulators there are user mailing lists. Get the packages and look into their documentation for further info. If you are using a small computer (PC, Mac, etc.) you may want to have a look at the Central Neural System Electronic Bulletin Board (see Answer 14) Modem: 509-627-6CNS; Sysop: Wesley R. Elsberry; P.O. Box 1187, Richland, WA 99352; welsberr@sandbox.kenn.wa.us There are lots of small simulator packages, the CNS ANNSIM file set. There is an ftp mirror site for the CNS ANNSIM file set at me.uta.edu (129.107.2.20) in the /pub/neural directory. Most ANN offerings are in /pub/neural/annsim.