HPAVC

home *** CD-ROM | disk | FTP | other *** search

/ HPAVC / HPAVC CD-ROM.iso / BRAIN12.ZIP / brain12.doc < prev next >

Wrap

Text File | 1994-03-05 | 39KB | 1,011 lines

The Brain Version 1.2s User's Manual _____________________________ (c) Copyright, All Rights Reserved, 1994, DP Computing DP Computing PO Box 712 Noarlunga Center SA 5168 Australia Internet: dpc@mep.com perkovic@cleese.apana.org.au The Brain v1.2 - User's Manual Page 2 ____________________________________________________________________ Table Of Contents ----------------- Introduction ............................. 3 Program Files ............................ 4 Introduction To Neural Networks .......... 5 What Are Neural Networks? ............. 5 How Do Neural Networks Learn? ......... 5 Uses of Neural Networks ............... 7 Training And Testing The Brain ........... 8 Training The Network .................. 8 Hints On Training A Network ........... 9 Testing The Network ................... 9 Input File Layout (with examples) ..... 10 Tutorial ................................. 14 General Texts On Neural Networks ......... 17 License Agreement ........................ 18 Support Policy ........................... 19 Distribution Policy ...................... 19 About This Manual ........................ 20 The Brain v1.2 - User's Manual Page 3 ____________________________________________________________________ Introduction ------------ Get ready to explore the exciting world of artificial intelligence. The Brain is an advanced neural network simulator that is simple enough to be used by non-technical people, yet sophisticated enough for serious research work. Based upon the backpropagation learning algorithm, The Brain allows you to train the computer to learn what you want it to learn. The Brain gives you a glimpse into the future of computing. With The Brain, you can create, train, and test your own neural networks. Three sample networks have been included with this distribution package: - a network to recognise the numerals 1, 2, and 3. - a network to process the logical AND function. - a network to process the logical XOR function. This manual will outline the features and capabilities of The Brain as well as providing you with a brief overview of the concepts and applications of neural networks. Several excellent books are listed at the end of this manual for those interested in a more thorough introduction to neural networks. The Brain is a shareware product. That means you get to try before you buy. If you are satisfied with The Brain and intend to continue using it, you are required to register it. See the 'register.doc' file for details on how and where you can register your copy of The Brain. The Brain v1.2 - User's Manual Page 4 ____________________________________________________________________ Program Files ------------- The distribution package of The Brain contains the following files: - brain12.exe The stand alone executable version of The Brain. - brain12.doc The documentation for The Brain (this file). - start-me.bat A batch file to start The Brain in Beginners Mode. - char123.net A sample input file to train a network to recognize the numerals 1, 2, and 3. - test123.net A sample input file to test a network using the numerals 1, 2, and 3. - test123.wts The weights saved after a network was trained to recognize the numerals 1, 2, and 3. - xor.net A sample input file to train a network to recognize the 'xor' function. - and.net A sample input file to train a network to recognize the 'and' function. - vendor.doc Documentation for vendors, sysops, and others outlining our policy on distributing this program. - license.doc The license for The Brain. - register.doc Registration details. This file explains how to register your copy of The Brain. Includes addresses for our worldwide registration sites. - order.doc Order form for The Brain. - support.doc A list of our worldwide support sites. - nn.faq The neural network faq (frequently asked questions) from the Internet news group 'comp.ai.neural.nets'. The Brain v1.2 - User's Manual Page 5 ____________________________________________________________________ Introduction to Neural Networks =============================== What Are Neural Networks? ------------------------- Expert Definition: A neural network is a parallel distributed information processing structure in the form of a directed graph with the most popular type being feed forward networks. Non-Technical Definition: A neural network can be thought of as a pattern recognition system. The computer learns to associate a certain pattern with a given result. For example, once a neural network has been taught the characteristics of the numerals 1, 2, and 3, it should be able to recognize those same numerals even if presented in a different font or in a different person's handwriting. How Do Neural Networks Learn? ----------------------------- Think of a group of interconnected units (as in figure 1). Data is first presented to the system at the input layer in the form of zeros and ones (representing 'off' and 'on' respectively). It then goes through a series of hidden mathematical calculations (within the hidden layer) before being passed to the output layer, whose job is to provide you with some sort of sensible and understandable result. | | O O Output Layer |/\| O O Hidden Layer |/\| 0 0 Input Layer | | (Figure 1) In general, neural networks can consist of any number of input, hidden, and output units, as well as any number of hidden layers. However, since any problem can be solved with only one hidden layer, and since generalization on unseen data is enhanced with the use of only one hidden layer, The Brain restricts you to a single hidden layer. Because of memory considerations, the unregistered version The Brain v1.2 - User's Manual Page 6 ____________________________________________________________________ of The Brain can contain a maximum of 30 units. The registered version allows you to build bigger networks by taking advantage of any extra memory you have available. Once the information has been processed through the input layer, it is fanned out to form the input to each unit in the hidden layer. Units in the hidden layer perform a calculation on the input that results in a decimal number between 0 and 1. This result then serves as the input for the output layer which again performs a calculation and produces an output in the range 0 to 1. The calculations the network performs in the hidden and output layers depend upon decimal numbers known as 'weights'. Since each network performs a unique task, the weights appropriate for that network are also unique. To determine a network's unique set of weights, the network must first be trained to learn how to recognize a specified input and output pattern. For example, to train the network to learn the numerals 1, 2, and 3, we must supply the network with a representation of those numerals. We must also tell the network what output is appropriate for each numeral. Thus, if we input the number one we would like the network to have the 1st output node 'on' (i.e. 1) while the other two output nodes are 'off' (i.e. 0). When the input is 2, the 2nd output node should be 'on' while the other two are 'off', and when the input is 3, the 3rd output node should be 'on' while the other two output nodes are 'off'. Training a network consists of an iterative process in which the network is given the desired inputs along with the correct outputs for those inputs. It then seeks to alter its weights to try and produce the correct output (within a reasonable error margin). If it succeeds, it has learned the training set and is ready to perform upon previously unseen data. If it fails to produce the correct output it re-reads the input and again tries to produce the correct output. The weights are slightly adjusted during each iteration through the training set (known as a training cycle) until the appropriate weights have been established. Depending upon the complexity of the task to be learned, many thousands of training cycles may be needed for the network to correctly identify the training set. Once the output is correct the weights can be used with the same network on unseen data to examine how well it performs. For the example of learning the numerals 1, 2, and 3, the test data could be one of those numerals entered in a different font or in someone else's handwriting. The whole idea of a neural network is to train the network on an input set and then to show the network a similar but different data set which it hasn't seen before. Hopefully the network can The Brain v1.2 - User's Manual Page 7 ____________________________________________________________________ correctly recognize it. This is very important in areas such as handwriting recognition. While humans can usually read handwriting that they've never seen before, this 'simple' task is much more difficult for computers. Uses of Neural Networks ----------------------- The main driving force behind neural network research is the desire to create a machine which works similar to the manner our own brain works. Neural networks have been used in a variety of different areas to solve a wide range of problems. The types of problems solved by (or currently being researched using) neural networks include: - voice recognition - image recognition - stock-market prediction - car navigation - data compression - backgammon - character recognition - chess - horse racing prediction - sonar recognition In theory, neural networks can compute any function a normal computer can. In practice, neural networks are useful for problems with a high error rate, that have many examples, and where no algorithm exists to solve the problem. Professions using neural networks include: - Computer scientists requiring solutions to problems where currently no algorithms exist. - Engineers wanting to exploit the capabilities of neural networks in their particular application areas. - Cognitive scientists using neural networks to describe models of thinking and conscience. - Neuro-physiologists using neural networks to describe and explore brain functions. - Physicists using neural networks to model phenomena in statistical mechanics. - Biologists using neural networks to model various biological processes. The Brain v1.2 - User's Manual Page 8 ____________________________________________________________________ Training and Testing The Brain ============================== Training The Brain ------------------ You can start The Brain in either of two ways: 1) For beginners: Type 'start-me' (without the quotes) at the DOS prompt. This will guide the beginner through the included tutorial/demo (see the Tutorial for more details). 2) For advanced users: Type 'brain12' (without the quotes) at the DOS prompt. This will start The Brain in normal mode. After starting The Brain, the first prompt will ask you to enter the filename in which the input data is stored. The input file contains the details about the problem, including the training examples and the correct output for each example. See the 'Input File Layout' section for details on creating/specifying input files. After the input file has been correctly loaded you'll be asked whether you want to load in a file containing a stored set of weights. This file must have been created by a previous training session. NOTE: An error will occur if you try to load a set of weights saved for a different sized network. Training of the network will then occur. Updates will be displayed after every 100 training cycles (i.e. after 100 presentations of the complete training set). Training is halted when the error result (a quadratic error function is used) drops below 0.2 or after 10000 training cycles have been processed. Training can be terminated prematurely by pressing control-c (i.e. while holding down the control key press the 'c' key). Training times vary depending upon the network size and the speed of the machine you are using. Large networks can sometimes take a long time to complete the training session. Once training has been completed a prompt appears asking whether you would like to save the weights. A set of weights is needed for testing the network. If the network was able to correctly identify the desired input patterns, and you intend to test the network, answer "y" to this prompt. See the section on "Testing the Network". Testing on unseen data can be carried out once the network has been trained. Unseen data refers to data similar to but different from the data used to train the network. For example, if you've trained The Brain v1.2 - User's Manual Page 9 ____________________________________________________________________ the network on your own handwriting, test it with someone else's handwriting). Hints on Training a Network --------------------------- Experimentation is required in order to enable a network to correctly learn a particular training set. Some networks may learn a particular problem using only one hidden unit while others may require 20 or 30 hidden units. The only way to find out is to try it! Networks also occasionally become stuck during learning. Either they take a very long time to learn a problem or they fail completely; it all depends on the initially chosen values for the weights. To overcome this problem, simply train the network again. Since The Brain randomly chooses the initial weights for the network (assuming you haven't loaded a set of weights), one or more additional training sessions should find an acceptable starting set of weights . Restarting the network should solve the problem of the network becoming stuck. Research has found that a network performs best on unseen data with a network using the least number of hidden units that can successfully learn the training set. If you would like good generalization on unseen data, train the network using the minimum number of hidden units that can successfully learn the training set. If the network is still unable to learn the desired input after training, try the following: - add more units to the hidden layer. - decrease/increase the size of the training set. - increase the training times. - alter some of the parameters used in backpropagation (NOTE: this can only be done in the registered version). Testing The Network ------------------- To test a network, start The Brain (by entering 'brain12' at the DOS prompt). When you are prompted for the input filename, enter the name of the test input file. A testing input file is almost identical to a training input file. The main differences are that you provide testing examples rather than training examples and you The Brain v1.2 - User's Manual Page 10 ____________________________________________________________________ omit the correct output results (The Brain should now be trained to figure out the correct results on its own). See the 'Input File Layout' section for details. After the input file has been correctly loaded, you'll be asked if you want to load in a previously stored set of weights. Enter 'y', then enter the name of the file containing the weights. This file should have been created after successfully completing a previous training session. NOTE: An error will occur if you try to load in a set of weights saved for a different sized network. The inputs contained in the input file will then be fed through the network using the weights loaded in. The output will be displayed once all calculations are completed. An example input file for testing (along with its associated weights file) has been included with the distribution version of The Brain: test123.net Testing data for the numerals 1, 2, and 3. test123.wts The weights saved after the network was trained to recognize the numerals 1, 2, and 3. If the network performs poorly during testing, you can improve its performance in two ways: - Add more examples into the training (for example, if you are training the network to learn handwriting, add handwriting samples from different/more people). - Ensure that the network contains the minimum number of hidden units. Input File Layout ----------------- The input file is a text file consisting of (in the following order): - The unit number for each of the following units: . the first input unit. (usually 1) . the last input unit. . the first hidden unit. . the last hidden unit. . the first output unit. . the last output unit. (a maximum of 30 due to memory considerations. The registered version uses any available extra memory to allow for more units). - The number of training/testing examples. - Either the word 'test' or the word 'train'. If the input file The Brain v1.2 - User's Manual Page 11 ____________________________________________________________________ represents a training session, 'train' should be used. If the input file represents a testing session, the word 'test' should be used. - The training/testing input data. This must be real (i.e. decimal) numbers consisting of 1.0 or 0.0 values ('on' or 'off'); - The actual output wanted, which also must be real (i.e. decimal) numbers. These are omitted if you are just testing the network. ---XOR Input Example--- The following is an example input file for the XOR problem using a network with 2 inputs, 2 hidden units, and 1 output unit. The following example input data is contained in the file 'xor.net', included with the distribution package. The XOR (exclusive or) problem is to determine when only one of two given inputs is 'on' (i.e. The result should be 'on', a 1, if only one of the inputs is 'on' and produce a result of 'off', a 0, when the two inputs are either both 'on' or both 'off'). | O output unit inputs output / \ 0 0 0 O O hidden units 1 0 1 |\ /| 0 1 1 O O input units 1 1 0 | | XOR Input File --------------------------cut here-------------------------- 1 2 3 4 5 5 4 train 0.0 0.0 1.0 0.0 0.0 1.0 1.0 1.0 0.0 1.0 1.0 0.0 --------------------------cut here-------------------------- The Brain v1.2 - User's Manual Page 12 ____________________________________________________________________ ---AND Input Example--- The following is an example input file for the AND problem using a network with 2 inputs, 2 hidden units, and 1 output unit. The following example input data is contained in the file 'and.net', included in the distribution package. The AND function is to produce an 'on' output (a 1) only if both the inputs are 'on' otherwise produce an 'off' (a 0). | O output unit inputs output / \ 0 0 0 O O hidden units 1 0 0 |\ /| 0 1 0 O O input units 1 1 1 | | AND Input File --------------------------cut here-------------------------- 1 2 3 4 5 5 4 train 0.0 0.0 1.0 0.0 0.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 --------------------------cut here-------------------------- ---Character Recognition Input Example--- This network is designed to recognise the characters 1, 2, and 3. - If the network has recognised a 1 the 1st output node should be 'on' (i.e. close to 1.0) and the other output nodes 'off' (i.e. close to 0.0). - If the network has recognised a 2 the 2nd output node should be 'on' (i.e. close to 1.0) and the other output nodes 'off' (i.e. close to 0.0). - If the networks have recognised a 3 the 3rd output node should be 'on' (i.e. close to 1.0) and the other output nodes 'off' (i.e. close to 0.0). The network consists of 20 input nodes, 3 hidden nodes, and 3 output nodes. The following example input file is contained in the file 'char123.net', included with the distribution package. The Brain v1.2 - User's Manual Page 13 ____________________________________________________________________ Character Recognition Input File --------------------------cut here-------------------------- 1 20 21 23 24 26 3 train 0.0 0.0 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 0.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 0.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 --------------------------cut here-------------------------- NOTE: Both the XOR and the AND example training data (described above) will most likely require a few training runs due to them both getting stuck. The character recognition training file should learn the desired output about every time. The Brain v1.2 - User's Manual Page 14 ____________________________________________________________________ Tutorial ======== To demonstrate a use for neural networks let us train The Brain to learn some numerals, namely 1, 2, and 3. 1 1111 1111 1 1 1 1 1111 111 1 1 1 1 1111 1111 These numerals can be represented using 0's and 1's as: 0010 1111 1111 0010 0001 0001 0010 1111 0111 0010 1000 0001 0010 1111 1111 Note: For illustrative purposes we are using whole numbers, '0' and '1'. Keep in mind that in the input file these numbers must be presented as decimal values ('0.0' and '1.0'). As each of the numerals consist of 20 0's and 1's, a network is constructed that consisted of 20 inputs. To represent a 1 the following is presented to the network: 00100010001000100010 This is taken from the above representation on the numeral one. Similarly for a 2: 11110001111110001111 and for a 3: 11110001011100011111. For the output layer we use 3 units. If the network has recognized the numeral one, the first output unit's result should be one (or close to one) and the second and third output units should be zero or close to zero. If the network has recognized the numeral two, the second output node should be one or close to one (i.e. 'on') and the first and third units should be zero or close to zero (i.e. 'off'). Similarly, if the numeral three is recognized the first and second output units should be 'off' and the third output unit 'on'. The Brain v1.2 - User's Manual Page 15 ____________________________________________________________________ From experimentation we've found that the minimum number of hidden units needed to correctly learn the above numerals was three. The input file (available in the distribution package as 'char123.net') needed to learn the three numerals (1, 2, and 3) is: 1 First input node is # 1. -| 20 Last input node is # 20. -|-->(i.e. 20 input units). 21 First hidden unit. -| 23 Last hidden unit. -|--> (i.e. 3 hidden units). 24 First input unit. -| 26 Last input unit. -|--> (i.e. 3 output units). 3 Number of training examples. train Indicates training the network (rather than testing). 0.0 0.0 1.0 0.0 -| 0.0 0.0 1.0 0.0 | 0.0 0.0 1.0 0.0 |--> The input pattern for the numeral one. 0.0 0.0 1.0 0.0 | 0.0 0.0 1.0 0.0 -| 1.0 1.0 1.0 1.0 -| 0.0 0.0 0.0 1.0 | 1.0 1.0 1.0 1.0 |--> The input pattern for the numeral two. 1.0 0.0 0.0 0.0 | 1.0 1.0 1.0 1.0 -| 1.0 1.0 1.0 1.0 -| 0.0 0.0 0.0 1.0 | 0.0 1.0 1.0 1.0 |--> The input pattern for the numeral three. 0.0 0.0 0.0 1.0 | 1.0 1.0 1.0 1.0 -| 1.0 0.0 0.0 The output needed when the numeral one is presented. 0.0 1.0 0.0 The output needed when the numeral two is presented. 0.0 0.0 1.0 The output needed when the numeral three is presented. Note: the comments on the right should not be included in the file. Each input pattern is presented on 5 lines to aid in visualization of each numeral; presentation of each pattern on a single line would make no difference. To run through the tutorial, enter 'start-me' at the DOS prompt. Enter 'char123.net' when the input file is requested. Press 'n' when asked if you want to load a set of saved weights. Learning of the training set will then start (i.e. it is now learning how to distinguish between the numerals 1, 2, and 3). The network should take less than one thousand training cycles (i.e. less than one thousand presentations of the 3 input patterns) to learn the 3 patterns. If it is taking longer it has most likely become stuck (as described in the section 'Training the Network') and you should stop the network (by pressing 'control-c') and start the training again by restarting the program. The Brain v1.2 - User's Manual Page 16 ____________________________________________________________________ After training has been completed you are given the option of saving the weights. The output of the network is then displayed. The output will look something similar to the following: Training example 1 0.920013 0.012343 0.211126 Training example 2 0.000129 0.834562 0.123983 Training example 3 0.300403 0.203044 0.970030 From the output we can see that for the first example the first output unit is close to being fully 'on' (i.e. close to 1) while the other two output units are close to being fully 'off' (i.e. close to 0). This indicates that the network has recognized that the first training example is, in fact, a numeral one. With the 2nd training example, the first and third output units are 'off' while the 2nd output unit is 'on', indicating we have identified the numeral two. A similar output is noted for the third training example: the first and second output units are 'off' with the third being 'on', showing the network correctly identifying the numeral three. Once the network is trained, the weights can be saved and then used to test the network on unseen data (see "Testing the Network"). For this example we can test the network using different representations of the numerals 1, 2, and 3 (i.e. alter their shapes by moving, shrinking, or expanding them). Most likely you will notice that the performance on the test data is poor. To get around this problem try expanding the training set by adding more examples of the numerals 1, 2, and 3, and by retraining and retesting the network. The Brain v1.2 - User's Manual Page 17 ____________________________________________________________________ General Texts on Neural Networks -------------------------------- Hertz, J., Krogh, A., and Palmer, R. (1991). Introduction to the Theory of Neural Computation. Addison-Wesley: Redwood City, California. ISBN 0-201-50395-6 (hardbound) and 0-201-51560-1 (paperbound) Hecht-Nielsen, R. (1990). Neurocomputing. Addison Wesley. Aleksander, I. and Morton, H. (1990). An Introduction to Neural Computing. Chapman and Hall. (ISBN 0-412-37780-2). Beale, R. and Jackson, T. (1990). Neural Computing, an Introduction. Adam Hilger, IOP Publishing Ltd : Bristol. (ISBN 0-85274-262-2). Dayhoff, J. E. (1990). Neural Network Architectures: An Introduction. Van Nostrand Reinhold: New York. McClelland, J. L. and Rumelhart, D. E. (1988). Explorations in Parallel Distributed Processing: Computational Models of Cognition and Perception (software manual). The MIT Press. McCord Nelson, M. and Illingworth, W.T. (1990). A Practical Guide to Neural Nets. Addison-Wesley Publishing Company, Inc. (ISBN 0-201-52376-0). Orchard, G.A. & Phillips, W.A. (1991). Neural Computation: A Beginner's Guide. Lawrence Earlbaum Associates: London. Wasserman, P. D. (1989). Neural Computing: Theory & Practice. Van Nostrand Reinhold: New York. (ISBN 0-442-20743-3) The Brain v1.2 - User's Manual Page 18 ____________________________________________________________________ License Agreement ----------------- You may use the distribution version of The Brain for evaluation purposes only. Once you start using this program regularly for educational, commercial, or private use you must register this product (see the file 'register.doc'). You may share this distribution version with anyone you choose so long as it is unaltered, and so long as you follow the distribution policy outlined in the file 'vendor.doc'. You are not permitted to share or otherwise distribute, in whole or in part, the registered version. By registering this product you acknowledge that this product represents a trade secret and agree to protect it. Misuse of a registered version is subject to collection of 100 times the registration fee and all legal fees and costs. Licenses are not transferable and may not be modified. DP COMPUTING DISCLAIMS ALL WARRANTIES RELATING TO THIS PRODUCT, WHETHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, AND ALL SUCH WARRANTIES ARE EXPRESSLY AND SPECIFICALLY DISCLAIMED. NEITHER DP COMPUTING NOR ANYONE ELSE WHO HAS BEEN INVOLVED IN THE CREATION, PRODUCTION, DELIVERY, OR DISTRIBUTION OF THIS PRODUCT SHALL BE LIABLE FOR ANY INDIRECT, CONSEQUENTIAL, OR INCIDENTAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE SUCH SOFTWARE EVEN IF DP COMPUTING HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES OR CLAIMS. IN NO EVENT SHALL DP COMPUTING'S LIABILITY FOR ANY DAMAGES EVER EXCEED THE PRICE PAID FOR THE LICENSE TO USE THE SOFTWARE REGARDLESS OF THE FORM OF THE CLAIM. THE PERSON USING THE SOFTWARE BEARS ALL RISKS AS TO THE QUALITY AND PERFORMANCE OF THE SOFTWARE. It is the users responsibility to determine whether the product will work reliably on their equipment and for their specific needs. That is the purpose of this evaluation version. DP Computing does not imply in any manner that this software is suitable for any given application or purpose. If any bugs are found please let us know and return a copy of the product with the bugs in it and we will do our best to fix it. If you don't agree with these conditions, delete this product from your disks. The Brain v1.2 - User's Manual Page 19 ____________________________________________________________________ Support Policy -------------- DP Computing is fully committed to providing the best possible support for our clients. If you have any problems at all please feel free to contact DP Computing or one of our support sites (listed in the file 'support.doc'). We will provide support for registered users for up to 3 months following the registration of the product. This support is for correcting bugs in the software and manuals and does not include advice on how to solve various problems using the neural network. Questions and advice on "The Brain" and neural networks in general may be answered depending upon time constraints. DP Computing is also available for contract work on the use of neural networks to solve specific problems. Please call for details. Distribution Policy ------------------- All DP Computing products are protected under Australian and International (c) copyright laws. As a shareware distributor you have permission to distribute any of DP Computing's shareware products as long as: - it is kept in its present electronic form. - it is clearly identified as shareware. - all copyright notices remain intact. - no file in this distribution package is modified or deleted. - we don't request you stop. You may archive the programs, unarchive them, use your own installation routines, include them with other programs on a disk, etc., so long as you follow the above rules. If you are a shareware distributor we would appreciate a copy of your catalog and the disk on which the program is placed. In return, we will send you all updates to the program. BBS owners are also urged to drop us a line so that we can keep you up to date with future releases. Shareware distributors and bulletin board systems may be named in our documentation as distribution sites if so desired. The Brain v1.2 - User's Manual Page 20 ____________________________________________________________________ If you have any questions, complaints, or concerns please contact me: David Perkovic DP Computing PO Box 712 Noarlunga Center SA 5162 Australia Ph: +61 8 326 4364 (International) 08 326 4364 (Within Australia) Internet: perkovic@cleese.apana.org.au dpc@mep.com International: If you are a publisher interested in supporting or translating our products please contact us for further information. Anyone interested in providing registration and/or distribution services outside of Australia please contact us at the above address. Site licenses available. About This Manual ----------------- This manual was written by David Perkovic (author of The Brain) and laid out / edited by Eugene Mallay. Eugene Mallay is a freelance writer and editor specializing in handbooks and manuals. He can be contacted at: Internet: emallay@io.org Voice: (416) 261-4241 Surface Mail: 945 Midland Ave, Suite 1003 Fax: (416) 261-7374 Scarborough, Ontario, Canada M1K 4G5 International clients are welcomed.