... I will include .ens in the next release. It does not seem to be used. The .ens extension will need to be associated with a suitable editor. Like all...
... The .ens extension will need to be associated with a suitable editor. Like all scripts .ens is plain text so I am using Notepad.<< Sounds good, thanks....
<I apologies if you recevied the multiple copies of the CFP> SECOND CALL FOR PAPERS KBCS-2004 THE FIFTH INTERNATIONAL CONFERENCE ON KNOWLEDGE BASED COMPUTER...
Group, I have been using a free text editor, Context, which highlights scripts according to customizable syntax files. I have attached a file for the Easy NN...
Hi All, I've just started trying to use EasyNN-Plus. I'm hoping to model software development costing with it. My problem is many times all I need to do is...
... Chuck, I've thought of a possible cause. It could happen on blank (new) grids if the edit menu starts to open. Is that the problem? If so, it's fixed in...
... The method I used may be of interest to you. I worked on a mainframe OS development team that used one of my neural netwroks to produce implementation,...
Hi Steve, Thanks for the info. I'm going to try and use the COCOMO II early design factors and a function point type system to model the code lines. The hard...
I have an input file that is a CSV file I created in Excel. When I import the file, all 24000 rows are set to training. Is there a fast way to set a few...
... Set the inputs and outputs so you can create a network. Then use the random selection in the Validating section of the Controls dialog. The validating rows...
Dear All * Is there any source of general info, and in particular, info on the format for text input and output in the "Who wrote this?" application? * Am I...
... What is the "Who wrote this?" application? Steve -- EasyNN-plus. The easy way to build neural networks. Build networks from numeric, text and image files. ...
... Yes, I remember now - it was a long time ago! The exercise was done by a student who was investigating the possibility of classifying writing styles from a...
Steve, I have 24 boolean columns represented by 0 & 1. Should I indicate they are boolean or integer? Output is also 0 & 1. If I indicate they are boolean,...
... Change them to bool. If you need they can be changed back to integer. With small integers don't worry about the warning message. It is for converting very...
Great! Now, a question on network format; In general, if you have three items that are compared to 2 items, (6 columns) would it help if I defined the net as...
... How do get 6 columns? What do you mean by "items"? ... Sorry Walt, I don't understand the question. Steve -- EasyNN-plus. The easy way to build neural...
Does anyone have any signal processing example networks in easyNN or have any information on them? Examples could be: EKG filtering, sine wave classification,...
Clarification; My input columns a>v1 b>v1 c>v1 a>v2 b>v2 c>v2 Since I know first 3 columns are related to v1 and next three are related to v2; would two...
I am trying to present some data in six columns (the maximum number of changes). There are three possible states for each cell, namely lower, unchanged or...
... I still don't understand the question. Are a, b, c the inputs and v1 and v2 the outputs? If so, what do you mean by the first three and next three? I can...
Steve, I'm sorry for being so obtuse, Let me try again; these six comparisons (boolean) are my input columns: a > v1, b > v1, c > v1 , a > v2, b > v2, c >...
... Now I understand. Stick to one layer. You should never need two layers when using validating. I can't think of any advantages of more than one layer. ... I...
... John, Three integers - I don't see any other possible format. -1, 0, 1 would be the logical choice but that won't work any better than 1, 2, 3. Steve -- ...
The first beta version of EasyNN-plus 5.0a is ready for testing. The changes are as follows:- 1. Four new toolbar buttons provide fast querying. The buttons...
Steve, I have a procedural question; I am using the following approach to determine inputs to my net: 1. I create various columns of data that I suspect...
... It is worth saving all the clusters found while associating is running. The clustering process does take much longer than associating. When it has...