Results for "Author: brewguru99"

Submission 2 for the RGCC contest Uses a neural network to provide a "universal" logical operator replacement function. This wonderful function will do all the work that Xor, Or, And, Imp, Eqv, and Not would do - IN ONE SIMPLE FUNCTION!!! God I love this contest.

Neural networks (NN) emulate the thinking process of the human brain, including the ability to "learn". Many applications exsist for NN's, most popular of which is video games. This is a complete NN "Engine", all nicely packaged into one class module. v1.0.2: Some optimization and NN saving. v1.0.3: Actually made the NN export work (:Þ) and added NNImport. Toyed around a bit with Memory Banks, but have put them off in the background for now. v1.0.4: Added Noise and Momentum factors. Noise is useful for a small sample of scalar data, while momentum prevents drastic changes when training as well as promoating continued change in the right direction. Also see my chess program that uses this class. (Not yet updated with v1.0.4... on its way...) Just also want to say that this work is heavly based off of the work by Ulli. His code used to be available here on PSC, but I don't know what happened to it.

Demonstrates the capability of a neural net by teaching it to play chess. This is inheritly slow, and will take many many hours to train the neural network properly. For mor information on neural networks, see my "Neural Network" code post.

Uses the Luhn formula to quickly validate a credit card. Basically all the digits except for the last one are summed together and the output is a single digit (0 to 9). This digit is compared with the last digit ensure a proper credit card number is entered (Does not actually confirm that is is a real number, just that it is likely to be one. Example: Entering "4000-0000-0000-0002" will pass the check, but "4000-0000-0000-0003" will not pass.)

Submission 2 for the RGCC contest Uses a neural network to provide a "universal" logical operator replacement function. This wonderful function will do all the work that Xor, Or, And, Imp, Eqv, and Not would do - IN ONE SIMPLE FUNCTION!!! God I love this contest.

Neural networks (NN) emulate the thinking process of the human brain, including the ability to "learn". Many applications exsist for NN's, most popular of which is video games. This is a complete NN "Engine", all nicely packaged into one class module. v1.0.2: Some optimization and NN saving. v1.0.3: Actually made the NN export work (:Þ) and added NNImport. Toyed around a bit with Memory Banks, but have put them off in the background for now. v1.0.4: Added Noise and Momentum factors. Noise is useful for a small sample of scalar data, while momentum prevents drastic changes when training as well as promoating continued change in the right direction. Also see my chess program that uses this class. (Not yet updated with v1.0.4... on its way...) Just also want to say that this work is heavly based off of the work by Ulli. His code used to be available here on PSC, but I don't know what happened to it.

Demonstrates the capability of a neural net by teaching it to play chess. This is inheritly slow, and will take many many hours to train the neural network properly. For mor information on neural networks, see my "Neural Network" code post.

Uses the Luhn formula to quickly validate a credit card. Basically all the digits except for the last one are summed together and the output is a single digit (0 to 9). This digit is compared with the last digit ensure a proper credit card number is entered (Does not actually confirm that is is a real number, just that it is likely to be one. Example: Entering "4000-0000-0000-0002" will pass the check, but "4000-0000-0000-0003" will not pass.)

Neural networks (NN) emulate the thinking process of the human brain, including the ability to "learn". Many applications exsist for NN's, most popular of which is video games. This is a complete NN "Engine", all nicely packaged into one class module. v1.0.2: Some optimization and NN saving. v1.0.3: Actually made the NN export work (:Þ) and added NNImport. Toyed around a bit with Memory Banks, but have put them off in the background for now. v1.0.4: Added Noise and Momentum factors. Noise is useful for a small sample of scalar data, while momentum prevents drastic changes when training as well as promoating continued change in the right direction. Also see my chess program that uses this class. (Not yet updated with v1.0.4... on its way...) Just also want to say that this work is heavly based off of the work by Ulli. His code used to be available here on PSC, but I don't know what happened to it.

Demonstrates the capability of a neural net by teaching it to play chess. This is inheritly slow, and will take many many hours to train the neural network properly. For mor information on neural networks, see my "Neural Network" code post.

Submission 2 for the RGCC contest Uses a neural network to provide a "universal" logical operator replacement function. This wonderful function will do all the work that Xor, Or, And, Imp, Eqv, and Not would do - IN ONE SIMPLE FUNCTION!!! God I love this contest.

Uses the Luhn formula to quickly validate a credit card. Basically all the digits except for the last one are summed together and the output is a single digit (0 to 9). This digit is compared with the last digit ensure a proper credit card number is entered (Does not actually confirm that is is a real number, just that it is likely to be one. Example: Entering "4000-0000-0000-0002" will pass the check, but "4000-0000-0000-0003" will not pass.)

Submission 2 for the RGCC contest Uses a neural network to provide a "universal" logical operator replacement function. This wonderful function will do all the work that Xor, Or, And, Imp, Eqv, and Not would do - IN ONE SIMPLE FUNCTION!!! God I love this contest.

Neural networks (NN) emulate the thinking process of the human brain, including the ability to "learn". Many applications exsist for NN's, most popular of which is video games. This is a complete NN "Engine", all nicely packaged into one class module. v1.0.2: Some optimization and NN saving. v1.0.3: Actually made the NN export work (:Þ) and added NNImport. Toyed around a bit with Memory Banks, but have put them off in the background for now. v1.0.4: Added Noise and Momentum factors. Noise is useful for a small sample of scalar data, while momentum prevents drastic changes when training as well as promoating continued change in the right direction. Also see my chess program that uses this class. (Not yet updated with v1.0.4... on its way...) Just also want to say that this work is heavly based off of the work by Ulli. His code used to be available here on PSC, but I don't know what happened to it.

Demonstrates the capability of a neural net by teaching it to play chess. This is inheritly slow, and will take many many hours to train the neural network properly. For mor information on neural networks, see my "Neural Network" code post.

Uses the Luhn formula to quickly validate a credit card. Basically all the digits except for the last one are summed together and the output is a single digit (0 to 9). This digit is compared with the last digit ensure a proper credit card number is entered (Does not actually confirm that is is a real number, just that it is likely to be one. Example: Entering "4000-0000-0000-0002" will pass the check, but "4000-0000-0000-0003" will not pass.)

Submission 2 for the RGCC contest Uses a neural network to provide a "universal" logical operator replacement function. This wonderful function will do all the work that Xor, Or, And, Imp, Eqv, and Not would do - IN ONE SIMPLE FUNCTION!!! God I love this contest.

Neural networks (NN) emulate the thinking process of the human brain, including the ability to "learn". Many applications exsist for NN's, most popular of which is video games. This is a complete NN "Engine", all nicely packaged into one class module. v1.0.2: Some optimization and NN saving. v1.0.3: Actually made the NN export work (:Þ) and added NNImport. Toyed around a bit with Memory Banks, but have put them off in the background for now. v1.0.4: Added Noise and Momentum factors. Noise is useful for a small sample of scalar data, while momentum prevents drastic changes when training as well as promoating continued change in the right direction. Also see my chess program that uses this class. (Not yet updated with v1.0.4... on its way...) Just also want to say that this work is heavly based off of the work by Ulli. His code used to be available here on PSC, but I don't know what happened to it.

Demonstrates the capability of a neural net by teaching it to play chess. This is inheritly slow, and will take many many hours to train the neural network properly. For mor information on neural networks, see my "Neural Network" code post.

Uses the Luhn formula to quickly validate a credit card. Basically all the digits except for the last one are summed together and the output is a single digit (0 to 9). This digit is compared with the last digit ensure a proper credit card number is entered (Does not actually confirm that is is a real number, just that it is likely to be one. Example: Entering "4000-0000-0000-0002" will pass the check, but "4000-0000-0000-0003" will not pass.)